23 files changed, 238 insertions, 118 deletions

diff --git a/fs/block_dev.c b/fs/block_dev.c
index 5147bdd3b8e..257b00e9842 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1102,6 +1102,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 			if (!bdev->bd_part)
 				goto out_clear;
 
+			ret = 0;
 			if (disk->fops->open) {
 				ret = disk->fops->open(bdev, mode);
 				if (ret == -ERESTARTSYS) {
@@ -1118,9 +1119,18 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 					put_disk(disk);
 					goto restart;
 				}
-				if (ret)
-					goto out_clear;
 			}
+			/*
+			 * If the device is invalidated, rescan partition
+			 * if open succeeded or failed with -ENOMEDIUM.
+			 * The latter is necessary to prevent ghost
+			 * partitions on a removed medium.
+			 */
+			if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM))
+				rescan_partitions(disk, bdev);
+			if (ret)
+				goto out_clear;
+
 			if (!bdev->bd_openers) {
 				bd_set_size(bdev,(loff_t)get_capacity(disk)<<9);
 				bdi = blk_get_backing_dev_info(bdev);
@@ -1128,8 +1138,6 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 					bdi = &default_backing_dev_info;
 				bdev_inode_switch_bdi(bdev->bd_inode, bdi);
 			}
-			if (bdev->bd_invalidated)
-				rescan_partitions(disk, bdev);
 		} else {
 			struct block_device *whole;
 			whole = bdget_disk(disk, 0);
@@ -1153,13 +1161,14 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 		}
 	} else {
 		if (bdev->bd_contains == bdev) {
-			if (bdev->bd_disk->fops->open) {
+			ret = 0;
+			if (bdev->bd_disk->fops->open)
 				ret = bdev->bd_disk->fops->open(bdev, mode);
-				if (ret)
-					goto out_unlock_bdev;
-			}
-			if (bdev->bd_invalidated)
+			/* the same as first opener case, read comment there */
+			if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM))
 				rescan_partitions(bdev->bd_disk, bdev);
+			if (ret)
+				goto out_unlock_bdev;
 		}
 		/* only one opener holds refs to the module and disk */
 		module_put(disk->fops->owner);
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 5d505aaa72f..44ea5b92e1b 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -178,12 +178,13 @@ static int btrfs_xattr_acl_set(struct dentry *dentry, const char *name,
 
 	if (value) {
 		acl = posix_acl_from_xattr(value, size);
+		if (IS_ERR(acl))
+			return PTR_ERR(acl);
+
 		if (acl) {
 			ret = posix_acl_valid(acl);
 			if (ret)
 				goto out;
-		} else if (IS_ERR(acl)) {
-			return PTR_ERR(acl);
 		}
 	}
 
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index cd52f7f556e..9ee6bd55e16 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -8856,23 +8856,38 @@ out:
 int btrfs_init_space_info(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_space_info *space_info;
+	struct btrfs_super_block *disk_super;
+	u64 features;
+	u64 flags;
+	int mixed = 0;
 	int ret;
 
-	ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_SYSTEM, 0, 0,
-								 &space_info);
-	if (ret)
-		return ret;
+	disk_super = &fs_info->super_copy;
+	if (!btrfs_super_root(disk_super))
+		return 1;
 
-	ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA, 0, 0,
-								 &space_info);
-	if (ret)
-		return ret;
+	features = btrfs_super_incompat_flags(disk_super);
+	if (features & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS)
+		mixed = 1;
 
-	ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_DATA, 0, 0,
-								 &space_info);
+	flags = BTRFS_BLOCK_GROUP_SYSTEM;
+	ret = update_space_info(fs_info, flags, 0, 0, &space_info);
 	if (ret)
-		return ret;
+		goto out;
 
+	if (mixed) {
+		flags = BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA;
+		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+	} else {
+		flags = BTRFS_BLOCK_GROUP_METADATA;
+		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+		if (ret)
+			goto out;
+
+		flags = BTRFS_BLOCK_GROUP_DATA;
+		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+	}
+out:
 	return ret;
 }
 
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index ffb48d6c543..2616f7ed479 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -81,6 +81,13 @@ static unsigned int btrfs_flags_to_ioctl(unsigned int flags)
 		iflags |= FS_NOATIME_FL;
 	if (flags & BTRFS_INODE_DIRSYNC)
 		iflags |= FS_DIRSYNC_FL;
+	if (flags & BTRFS_INODE_NODATACOW)
+		iflags |= FS_NOCOW_FL;
+
+	if ((flags & BTRFS_INODE_COMPRESS) && !(flags & BTRFS_INODE_NOCOMPRESS))
+		iflags |= FS_COMPR_FL;
+	else if (flags & BTRFS_INODE_NOCOMPRESS)
+		iflags |= FS_NOCOMP_FL;
 
 	return iflags;
 }
@@ -144,16 +151,13 @@ static int check_flags(unsigned int flags)
 	if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \
 		      FS_NOATIME_FL | FS_NODUMP_FL | \
 		      FS_SYNC_FL | FS_DIRSYNC_FL | \
-		      FS_NOCOMP_FL | FS_COMPR_FL | \
-		      FS_NOCOW_FL | FS_COW_FL))
+		      FS_NOCOMP_FL | FS_COMPR_FL |
+		      FS_NOCOW_FL))
 		return -EOPNOTSUPP;
 
 	if ((flags & FS_NOCOMP_FL) && (flags & FS_COMPR_FL))
 		return -EINVAL;
 
-	if ((flags & FS_NOCOW_FL) && (flags & FS_COW_FL))
-		return -EINVAL;
-
 	return 0;
 }
 
@@ -218,6 +222,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 		ip->flags |= BTRFS_INODE_DIRSYNC;
 	else
 		ip->flags &= ~BTRFS_INODE_DIRSYNC;
+	if (flags & FS_NOCOW_FL)
+		ip->flags |= BTRFS_INODE_NODATACOW;
+	else
+		ip->flags &= ~BTRFS_INODE_NODATACOW;
 
 	/*
 	 * The COMPRESS flag can only be changed by users, while the NOCOMPRESS
@@ -230,11 +238,9 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 	} else if (flags & FS_COMPR_FL) {
 		ip->flags |= BTRFS_INODE_COMPRESS;
 		ip->flags &= ~BTRFS_INODE_NOCOMPRESS;
+	} else {
+		ip->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS);
 	}
-	if (flags & FS_NOCOW_FL)
-		ip->flags |= BTRFS_INODE_NODATACOW;
-	else if (flags & FS_COW_FL)
-		ip->flags &= ~BTRFS_INODE_NODATACOW;
 
 	trans = btrfs_join_transaction(root, 1);
 	BUG_ON(IS_ERR(trans));
diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c
index 23d43cde430..1b2e180b018 100644
--- a/fs/cifs/cifs_unicode.c
+++ b/fs/cifs/cifs_unicode.c
@@ -277,6 +277,7 @@ cifsConvertToUCS(__le16 *target, const char *source, int srclen,
 
 	for (i = 0, j = 0; i < srclen; j++) {
 		src_char = source[i];
+		charlen = 1;
 		switch (src_char) {
 		case 0:
 			put_unaligned(0, &target[j]);
@@ -316,16 +317,13 @@ cifsConvertToUCS(__le16 *target, const char *source, int srclen,
 				dst_char = cpu_to_le16(0x003f);
 				charlen = 1;
 			}
-			/*
-			 * character may take more than one byte in the source
-			 * string, but will take exactly two bytes in the
-			 * target string
-			 */
-			i += charlen;
-			continue;
 		}
+		/*
+		 * character may take more than one byte in the source string,
+		 * but will take exactly two bytes in the target string
+		 */
+		i += charlen;
 		put_unaligned(dst_char, &target[j]);
-		i++; /* move to next char in source string */
 	}
 
 ctoUCS_out:
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 05f1dcf7d79..277262a8e82 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -2673,6 +2673,11 @@ is_path_accessible(int xid, struct cifsTconInfo *tcon,
 			      0 /* not legacy */, cifs_sb->local_nls,
 			      cifs_sb->mnt_cifs_flags &
 				CIFS_MOUNT_MAP_SPECIAL_CHR);
+
+	if (rc == -EOPNOTSUPP || rc == -EINVAL)
+		rc = SMBQueryInformation(xid, tcon, full_path, pfile_info,
+				cifs_sb->local_nls, cifs_sb->mnt_cifs_flags &
+				  CIFS_MOUNT_MAP_SPECIAL_CHR);
 	kfree(pfile_info);
 	return rc;
 }
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 3313dd19f54..9a37a9b6de3 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -53,11 +53,14 @@ DEFINE_SPINLOCK(configfs_dirent_lock);
 static void configfs_d_iput(struct dentry * dentry,
 			    struct inode * inode)
 {
-	struct configfs_dirent * sd = dentry->d_fsdata;
+	struct configfs_dirent *sd = dentry->d_fsdata;
 
 	if (sd) {
 		BUG_ON(sd->s_dentry != dentry);
+		/* Coordinate with configfs_readdir */
+		spin_lock(&configfs_dirent_lock);
 		sd->s_dentry = NULL;
+		spin_unlock(&configfs_dirent_lock);
 		configfs_put(sd);
 	}
 	iput(inode);
@@ -689,7 +692,8 @@ static int create_default_group(struct config_group *parent_group,
 			sd = child->d_fsdata;
 			sd->s_type |= CONFIGFS_USET_DEFAULT;
 		} else {
-			d_delete(child);
+			BUG_ON(child->d_inode);
+			d_drop(child);
 			dput(child);
 		}
 	}
@@ -1545,7 +1549,7 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir
 	struct configfs_dirent * parent_sd = dentry->d_fsdata;
 	struct configfs_dirent *cursor = filp->private_data;
 	struct list_head *p, *q = &cursor->s_sibling;
-	ino_t ino;
+	ino_t ino = 0;
 	int i = filp->f_pos;
 
 	switch (i) {
@@ -1573,6 +1577,7 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir
 				struct configfs_dirent *next;
 				const char * name;
 				int len;
+				struct inode *inode = NULL;
 
 				next = list_entry(p, struct configfs_dirent,
 						   s_sibling);
@@ -1581,9 +1586,28 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir
 
 				name = configfs_get_name(next);
 				len = strlen(name);
-				if (next->s_dentry)
-					ino = next->s_dentry->d_inode->i_ino;
-				else
+
+				/*
+				 * We'll have a dentry and an inode for
+				 * PINNED items and for open attribute
+				 * files.  We lock here to prevent a race
+				 * with configfs_d_iput() clearing
+				 * s_dentry before calling iput().
+				 *
+				 * Why do we go to the trouble?  If
+				 * someone has an attribute file open,
+				 * the inode number should match until
+				 * they close it.  Beyond that, we don't
+				 * care.
+				 */
+				spin_lock(&configfs_dirent_lock);
+				dentry = next->s_dentry;
+				if (dentry)
+					inode = dentry->d_inode;
+				if (inode)
+					ino = inode->i_ino;
+				spin_unlock(&configfs_dirent_lock);
+				if (!inode)
 					ino = iunique(configfs_sb, 2);
 
 				if (filldir(dirent, name, len, filp->f_pos, ino,
@@ -1683,7 +1707,8 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys)
 		err = configfs_attach_group(sd->s_element, &group->cg_item,
 					    dentry);
 		if (err) {
-			d_delete(dentry);
+			BUG_ON(dentry->d_inode);
+			d_drop(dentry);
 			dput(dentry);
 		} else {
 			spin_lock(&configfs_dirent_lock);
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index c6ba49bd95b..b32eb29a4e6 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -174,7 +174,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd)
 		if (!inode)
 			return 0;
 
-		if (nd->flags & LOOKUP_RCU)
+		if (nd && (nd->flags & LOOKUP_RCU))
 			return -ECHILD;
 
 		fc = get_fuse_conn(inode);
diff --git a/fs/namei.c b/fs/namei.c
index 54fc993e302..e3c4f112ebf 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -179,7 +179,7 @@ EXPORT_SYMBOL(putname);
 static int acl_permission_check(struct inode *inode, int mask, unsigned int flags,
 		int (*check_acl)(struct inode *inode, int mask, unsigned int flags))
 {
-	umode_t			mode = inode->i_mode;
+	unsigned int mode = inode->i_mode;
 
 	mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
 
diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c
index 6f8192f4cfc..be79dc9f386 100644
--- a/fs/nfs/nfs4filelayout.c
+++ b/fs/nfs/nfs4filelayout.c
@@ -117,6 +117,8 @@ static int filelayout_async_handle_error(struct rpc_task *task,
 	case -EKEYEXPIRED:
 		rpc_delay(task, FILELAYOUT_POLL_RETRY_MAX);
 		break;
+	case -NFS4ERR_RETRY_UNCACHED_REP:
+		break;
 	default:
 		dprintk("%s DS error. Retry through MDS %d\n", __func__,
 			task->tk_status);
@@ -416,7 +418,8 @@ static int
 filelayout_check_layout(struct pnfs_layout_hdr *lo,
 			struct nfs4_filelayout_segment *fl,
 			struct nfs4_layoutget_res *lgr,
-			struct nfs4_deviceid *id)
+			struct nfs4_deviceid *id,
+			gfp_t gfp_flags)
 {
 	struct nfs4_file_layout_dsaddr *dsaddr;
 	int status = -EINVAL;
@@ -439,7 +442,7 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo,
 	/* find and reference the deviceid */
 	dsaddr = nfs4_fl_find_get_deviceid(id);
 	if (dsaddr == NULL) {
-		dsaddr = get_device_info(lo->plh_inode, id);
+		dsaddr = get_device_info(lo->plh_inode, id, gfp_flags);
 		if (dsaddr == NULL)
 			goto out;
 	}
@@ -500,7 +503,8 @@ static int
 filelayout_decode_layout(struct pnfs_layout_hdr *flo,
 			 struct nfs4_filelayout_segment *fl,
 			 struct nfs4_layoutget_res *lgr,
-			 struct nfs4_deviceid *id)
+			 struct nfs4_deviceid *id,
+			 gfp_t gfp_flags)
 {
 	struct xdr_stream stream;
 	struct xdr_buf buf = {
@@ -516,7 +520,7 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo,
 
 	dprintk("%s: set_layout_map Begin\n", __func__);
 
-	scratch = alloc_page(GFP_KERNEL);
+	scratch = alloc_page(gfp_flags);
 	if (!scratch)
 		return -ENOMEM;
 
@@ -554,13 +558,13 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo,
 		goto out_err;
 
 	fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *),
-			       GFP_KERNEL);
+			       gfp_flags);
 	if (!fl->fh_array)
 		goto out_err;
 
 	for (i = 0; i < fl->num_fh; i++) {
 		/* Do we want to use a mempool here? */
-		fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), GFP_KERNEL);
+		fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), gfp_flags);
 		if (!fl->fh_array[i])
 			goto out_err_free;
 
@@ -605,19 +609,20 @@ filelayout_free_lseg(struct pnfs_layout_segment *lseg)
 
 static struct pnfs_layout_segment *
 filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
-		      struct nfs4_layoutget_res *lgr)
+		      struct nfs4_layoutget_res *lgr,
+		      gfp_t gfp_flags)
 {
 	struct nfs4_filelayout_segment *fl;
 	int rc;
 	struct nfs4_deviceid id;
 
 	dprintk("--> %s\n", __func__);
-	fl = kzalloc(sizeof(*fl), GFP_KERNEL);
+	fl = kzalloc(sizeof(*fl), gfp_flags);
 	if (!fl)
 		return NULL;
 
-	rc = filelayout_decode_layout(layoutid, fl, lgr, &id);
-	if (rc != 0 || filelayout_check_layout(layoutid, fl, lgr, &id)) {
+	rc = filelayout_decode_layout(layoutid, fl, lgr, &id, gfp_flags);
+	if (rc != 0 || filelayout_check_layout(layoutid, fl, lgr, &id, gfp_flags)) {
 		_filelayout_free_lseg(fl);
 		return NULL;
 	}
@@ -633,7 +638,7 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
 		int size = (fl->stripe_type == STRIPE_SPARSE) ?
 			fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
 
-		fl->commit_buckets = kcalloc(size, sizeof(struct list_head), GFP_KERNEL);
+		fl->commit_buckets = kcalloc(size, sizeof(struct list_head), gfp_flags);
 		if (!fl->commit_buckets) {
 			filelayout_free_lseg(&fl->generic_hdr);
 			return NULL;
diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/nfs4filelayout.h
index 7c44579f583..2b461d77b43 100644
--- a/fs/nfs/nfs4filelayout.h
+++ b/fs/nfs/nfs4filelayout.h
@@ -104,6 +104,6 @@ extern struct nfs4_file_layout_dsaddr *
 nfs4_fl_find_get_deviceid(struct nfs4_deviceid *dev_id);
 extern void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr);
 struct nfs4_file_layout_dsaddr *
-get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id);
+get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags);
 
 #endif /* FS_NFS_NFS4FILELAYOUT_H */
diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c
index de5350f2b24..db07c7af139 100644
--- a/fs/nfs/nfs4filelayoutdev.c
+++ b/fs/nfs/nfs4filelayoutdev.c
@@ -225,11 +225,11 @@ nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr)
 }
 
 static struct nfs4_pnfs_ds *
-nfs4_pnfs_ds_add(struct inode *inode, u32 ip_addr, u32 port)
+nfs4_pnfs_ds_add(struct inode *inode, u32 ip_addr, u32 port, gfp_t gfp_flags)
 {
 	struct nfs4_pnfs_ds *tmp_ds, *ds;
 
-	ds = kzalloc(sizeof(*tmp_ds), GFP_KERNEL);
+	ds = kzalloc(sizeof(*tmp_ds), gfp_flags);
 	if (!ds)
 		goto out;
 
@@ -261,7 +261,7 @@ out:
  * Currently only support ipv4, and one multi-path address.
  */
 static struct nfs4_pnfs_ds *
-decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode)
+decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode, gfp_t gfp_flags)
 {
 	struct nfs4_pnfs_ds *ds = NULL;
 	char *buf;
@@ -303,7 +303,7 @@ decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode)
 			rlen);
 		goto out_err;
 	}
-	buf = kmalloc(rlen + 1, GFP_KERNEL);
+	buf = kmalloc(rlen + 1, gfp_flags);
 	if (!buf) {
 		dprintk("%s: Not enough memory\n", __func__);
 		goto out_err;
@@ -333,7 +333,7 @@ decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode)
 	sscanf(pstr, "-%d-%d", &tmp[0], &tmp[1]);
 	port = htons((tmp[0] << 8) | (tmp[1]));
 
-	ds = nfs4_pnfs_ds_add(inode, ip_addr, port);
+	ds = nfs4_pnfs_ds_add(inode, ip_addr, port, gfp_flags);
 	dprintk("%s: Decoded address and port %s\n", __func__, buf);
 out_free:
 	kfree(buf);
@@ -343,7 +343,7 @@ out_err:
 
 /* Decode opaque device data and return the result */
 static struct nfs4_file_layout_dsaddr*
-decode_device(struct inode *ino, struct pnfs_device *pdev)
+decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags)
 {
 	int i;
 	u32 cnt, num;
@@ -362,7 +362,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev)
 	struct page *scratch;
 
 	/* set up xdr stream */
-	scratch = alloc_page(GFP_KERNEL);
+	scratch = alloc_page(gfp_flags);
 	if (!scratch)
 		goto out_err;
 
@@ -384,7 +384,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev)
 	}
 
 	/* read stripe indices */
-	stripe_indices = kcalloc(cnt, sizeof(u8), GFP_KERNEL);
+	stripe_indices = kcalloc(cnt, sizeof(u8), gfp_flags);
 	if (!stripe_indices)
 		goto out_err_free_scratch;
 
@@ -423,7 +423,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev)
 
 	dsaddr = kzalloc(sizeof(*dsaddr) +
 			(sizeof(struct nfs4_pnfs_ds *) * (num - 1)),
-			GFP_KERNEL);
+			gfp_flags);
 	if (!dsaddr)
 		goto out_err_free_stripe_indices;
 
@@ -452,7 +452,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev)
 		for (j = 0; j < mp_count; j++) {
 			if (j == 0) {
 				dsaddr->ds_list[i] = decode_and_add_ds(&stream,
-					ino);
+					ino, gfp_flags);
 				if (dsaddr->ds_list[i] == NULL)
 					goto out_err_free_deviceid;
 			} else {
@@ -503,12 +503,12 @@ out_err:
  * available devices.
  */
 static struct nfs4_file_layout_dsaddr *
-decode_and_add_device(struct inode *inode, struct pnfs_device *dev)
+decode_and_add_device(struct inode *inode, struct pnfs_device *dev, gfp_t gfp_flags)
 {
 	struct nfs4_file_layout_dsaddr *d, *new;
 	long hash;
 
-	new = decode_device(inode, dev);
+	new = decode_device(inode, dev, gfp_flags);
 	if (!new) {
 		printk(KERN_WARNING "%s: Could not decode or add device\n",
 			__func__);
@@ -537,7 +537,7 @@ decode_and_add_device(struct inode *inode, struct pnfs_device *dev)
  * of available devices, and return it.
  */
 struct nfs4_file_layout_dsaddr *
-get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id)
+get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags)
 {
 	struct pnfs_device *pdev = NULL;
 	u32 max_resp_sz;
@@ -556,17 +556,17 @@ get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id)
 	dprintk("%s inode %p max_resp_sz %u max_pages %d\n",
 		__func__, inode, max_resp_sz, max_pages);
 
-	pdev = kzalloc(sizeof(struct pnfs_device), GFP_KERNEL);
+	pdev = kzalloc(sizeof(struct pnfs_device), gfp_flags);
 	if (pdev == NULL)
 		return NULL;
 
-	pages = kzalloc(max_pages * sizeof(struct page *), GFP_KERNEL);
+	pages = kzalloc(max_pages * sizeof(struct page *), gfp_flags);
 	if (pages == NULL) {
 		kfree(pdev);
 		return NULL;
 	}
 	for (i = 0; i < max_pages; i++) {
-		pages[i] = alloc_page(GFP_KERNEL);
+		pages[i] = alloc_page(gfp_flags);
 		if (!pages[i])
 			goto out_free;
 	}
@@ -587,7 +587,7 @@ get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id)
 	 * Found new device, need to decode it and then add it to the
 	 * list of known devices for this mountpoint.
 	 */
-	dsaddr = decode_and_add_device(inode, pdev);
+	dsaddr = decode_and_add_device(inode, pdev, gfp_flags);
 out_free:
 	for (i = 0; i < max_pages; i++)
 		__free_page(pages[i]);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 69c0f3c5ee7..cf1b339c393 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -300,6 +300,7 @@ static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struc
 			ret = nfs4_delay(server->client, &exception->timeout);
 			if (ret != 0)
 				break;
+		case -NFS4ERR_RETRY_UNCACHED_REP:
 		case -NFS4ERR_OLD_STATEID:
 			exception->retry = 1;
 			break;
@@ -3695,6 +3696,7 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 			rpc_delay(task, NFS4_POLL_RETRY_MAX);
 			task->tk_status = 0;
 			return -EAGAIN;
+		case -NFS4ERR_RETRY_UNCACHED_REP:
 		case -NFS4ERR_OLD_STATEID:
 			task->tk_status = 0;
 			return -EAGAIN;
@@ -4844,6 +4846,8 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
 		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
 		rpc_delay(task, NFS4_POLL_RETRY_MIN);
 		task->tk_status = 0;
+		/* fall through */
+	case -NFS4ERR_RETRY_UNCACHED_REP:
 		nfs_restart_rpc(task, data->clp);
 		return;
 	}
@@ -5479,6 +5483,8 @@ static int nfs41_reclaim_complete_handle_errors(struct rpc_task *task, struct nf
 		break;
 	case -NFS4ERR_DELAY:
 		rpc_delay(task, NFS4_POLL_RETRY_MAX);
+		/* fall through */
+	case -NFS4ERR_RETRY_UNCACHED_REP:
 		return -EAGAIN;
 	default:
 		nfs4_schedule_lease_recovery(clp);
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index ff681ab65d3..f57f5281a52 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -383,6 +383,7 @@ pnfs_destroy_all_layouts(struct nfs_client *clp)
 				plh_layouts);
 		dprintk("%s freeing layout for inode %lu\n", __func__,
 			lo->plh_inode->i_ino);
+		list_del_init(&lo->plh_layouts);
 		pnfs_destroy_layout(NFS_I(lo->plh_inode));
 	}
 }
@@ -466,7 +467,8 @@ pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct pnfs_layout_hdr *lo,
 static struct pnfs_layout_segment *
 send_layoutget(struct pnfs_layout_hdr *lo,
 	   struct nfs_open_context *ctx,
-	   u32 iomode)
+	   u32 iomode,
+	   gfp_t gfp_flags)
 {
 	struct inode *ino = lo->plh_inode;
 	struct nfs_server *server = NFS_SERVER(ino);
@@ -479,7 +481,7 @@ send_layoutget(struct pnfs_layout_hdr *lo,
 	dprintk("--> %s\n", __func__);
 
 	BUG_ON(ctx == NULL);
-	lgp = kzalloc(sizeof(*lgp), GFP_KERNEL);
+	lgp = kzalloc(sizeof(*lgp), gfp_flags);
 	if (lgp == NULL)
 		return NULL;
 
@@ -487,12 +489,12 @@ send_layoutget(struct pnfs_layout_hdr *lo,
 	max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
 	max_pages = max_resp_sz >> PAGE_SHIFT;
 
-	pages = kzalloc(max_pages * sizeof(struct page *), GFP_KERNEL);
+	pages = kzalloc(max_pages * sizeof(struct page *), gfp_flags);
 	if (!pages)
 		goto out_err_free;
 
 	for (i = 0; i < max_pages; i++) {
-		pages[i] = alloc_page(GFP_KERNEL);
+		pages[i] = alloc_page(gfp_flags);
 		if (!pages[i])
 			goto out_err_free;
 	}
@@ -508,6 +510,7 @@ send_layoutget(struct pnfs_layout_hdr *lo,
 	lgp->args.layout.pages = pages;
 	lgp->args.layout.pglen = max_pages * PAGE_SIZE;
 	lgp->lsegpp = &lseg;
+	lgp->gfp_flags = gfp_flags;
 
 	/* Synchronously retrieve layout information from server and
 	 * store in lseg.
@@ -665,11 +668,11 @@ pnfs_insert_layout(struct pnfs_layout_hdr *lo,
 }
 
 static struct pnfs_layout_hdr *
-alloc_init_layout_hdr(struct inode *ino)
+alloc_init_layout_hdr(struct inode *ino, gfp_t gfp_flags)
 {
 	struct pnfs_layout_hdr *lo;
 
-	lo = kzalloc(sizeof(struct pnfs_layout_hdr), GFP_KERNEL);
+	lo = kzalloc(sizeof(struct pnfs_layout_hdr), gfp_flags);
 	if (!lo)
 		return NULL;
 	atomic_set(&lo->plh_refcount, 1);
@@ -681,7 +684,7 @@ alloc_init_layout_hdr(struct inode *ino)
 }
 
 static struct pnfs_layout_hdr *
-pnfs_find_alloc_layout(struct inode *ino)
+pnfs_find_alloc_layout(struct inode *ino, gfp_t gfp_flags)
 {
 	struct nfs_inode *nfsi = NFS_I(ino);
 	struct pnfs_layout_hdr *new = NULL;
@@ -696,7 +699,7 @@ pnfs_find_alloc_layout(struct inode *ino)
 			return nfsi->layout;
 	}
 	spin_unlock(&ino->i_lock);
-	new = alloc_init_layout_hdr(ino);
+	new = alloc_init_layout_hdr(ino, gfp_flags);
 	spin_lock(&ino->i_lock);
 
 	if (likely(nfsi->layout == NULL))	/* Won the race? */
@@ -756,7 +759,8 @@ pnfs_find_lseg(struct pnfs_layout_hdr *lo, u32 iomode)
 struct pnfs_layout_segment *
 pnfs_update_layout(struct inode *ino,
 		   struct nfs_open_context *ctx,
-		   enum pnfs_iomode iomode)
+		   enum pnfs_iomode iomode,
+		   gfp_t gfp_flags)
 {
 	struct nfs_inode *nfsi = NFS_I(ino);
 	struct nfs_client *clp = NFS_SERVER(ino)->nfs_client;
@@ -767,7 +771,7 @@ pnfs_update_layout(struct inode *ino,
 	if (!pnfs_enabled_sb(NFS_SERVER(ino)))
 		return NULL;
 	spin_lock(&ino->i_lock);
-	lo = pnfs_find_alloc_layout(ino);
+	lo = pnfs_find_alloc_layout(ino, gfp_flags);
 	if (lo == NULL) {
 		dprintk("%s ERROR: can't get pnfs_layout_hdr\n", __func__);
 		goto out_unlock;
@@ -807,7 +811,7 @@ pnfs_update_layout(struct inode *ino,
 		spin_unlock(&clp->cl_lock);
 	}
 
-	lseg = send_layoutget(lo, ctx, iomode);
+	lseg = send_layoutget(lo, ctx, iomode, gfp_flags);
 	if (!lseg && first) {
 		spin_lock(&clp->cl_lock);
 		list_del_init(&lo->plh_layouts);
@@ -846,7 +850,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
 		goto out;
 	}
 	/* Inject layout blob into I/O device driver */
-	lseg = NFS_SERVER(ino)->pnfs_curr_ld->alloc_lseg(lo, res);
+	lseg = NFS_SERVER(ino)->pnfs_curr_ld->alloc_lseg(lo, res, lgp->gfp_flags);
 	if (!lseg || IS_ERR(lseg)) {
 		if (!lseg)
 			status = -ENOMEM;
@@ -899,7 +903,8 @@ static int pnfs_read_pg_test(struct nfs_pageio_descriptor *pgio,
 		/* This is first coelesce call for a series of nfs_pages */
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   prev->wb_context,
-						   IOMODE_READ);
+						   IOMODE_READ,
+						   GFP_KERNEL);
 	}
 	return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req);
 }
@@ -921,7 +926,8 @@ static int pnfs_write_pg_test(struct nfs_pageio_descriptor *pgio,
 		/* This is first coelesce call for a series of nfs_pages */
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   prev->wb_context,
-						   IOMODE_RW);
+						   IOMODE_RW,
+						   GFP_NOFS);
 	}
 	return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req);
 }
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index bc4827202e7..0c015bad9e7 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -70,7 +70,7 @@ struct pnfs_layoutdriver_type {
 	const u32 id;
 	const char *name;
 	struct module *owner;
-	struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr);
+	struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr, gfp_t gfp_flags);
 	void (*free_lseg) (struct pnfs_layout_segment *lseg);
 
 	/* test for nfs page cache coalescing */
@@ -126,7 +126,7 @@ void get_layout_hdr(struct pnfs_layout_hdr *lo);
 void put_lseg(struct pnfs_layout_segment *lseg);
 struct pnfs_layout_segment *
 pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx,
-		   enum pnfs_iomode access_type);
+		   enum pnfs_iomode access_type, gfp_t gfp_flags);
 void set_pnfs_layoutdriver(struct nfs_server *, u32 id);
 void unset_pnfs_layoutdriver(struct nfs_server *);
 enum pnfs_try_status pnfs_try_to_write_data(struct nfs_write_data *,
@@ -245,7 +245,7 @@ static inline void put_lseg(struct pnfs_layout_segment *lseg)
 
 static inline struct pnfs_layout_segment *
 pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx,
-		   enum pnfs_iomode access_type)
+		   enum pnfs_iomode access_type, gfp_t gfp_flags)
 {
 	return NULL;
 }
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 7cded2b12a0..2bcf0dc306a 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -288,7 +288,7 @@ static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc)
 	atomic_set(&req->wb_complete, requests);
 
 	BUG_ON(desc->pg_lseg != NULL);
-	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ);
+	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ, GFP_KERNEL);
 	ClearPageError(page);
 	offset = 0;
 	nbytes = desc->pg_count;
@@ -351,7 +351,7 @@ static int nfs_pagein_one(struct nfs_pageio_descriptor *desc)
 	}
 	req = nfs_list_entry(data->pages.next);
 	if ((!lseg) && list_is_singular(&data->pages))
-		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ);
+		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ, GFP_KERNEL);
 
 	ret = nfs_read_rpcsetup(req, data, &nfs_read_full_ops, desc->pg_count,
 				0, lseg);
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index 3bd5d7e80f6..49c715b4ac9 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -939,7 +939,7 @@ static int nfs_flush_multi(struct nfs_pageio_descriptor *desc)
 	atomic_set(&req->wb_complete, requests);
 
 	BUG_ON(desc->pg_lseg);
-	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW);
+	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW, GFP_NOFS);
 	ClearPageError(page);
 	offset = 0;
 	nbytes = desc->pg_count;
@@ -1013,7 +1013,7 @@ static int nfs_flush_one(struct nfs_pageio_descriptor *desc)
 	}
 	req = nfs_list_entry(data->pages.next);
 	if ((!lseg) && list_is_singular(&data->pages))
-		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW);
+		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW, GFP_NOFS);
 
 	if ((desc->pg_ioflags & FLUSH_COND_STABLE) &&
 	    (desc->pg_moreio || NFS_I(desc->pg_inode)->ncommit))
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index 643720209a9..9a3e6bbff27 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -539,25 +539,41 @@ static int o2hb_verify_crc(struct o2hb_region *reg,
 
 /* We want to make sure that nobody is heartbeating on top of us --
  * this will help detect an invalid configuration. */
-static int o2hb_check_last_timestamp(struct o2hb_region *reg)
+static void o2hb_check_last_timestamp(struct o2hb_region *reg)
 {
-	int node_num, ret;
 	struct o2hb_disk_slot *slot;
 	struct o2hb_disk_heartbeat_block *hb_block;
+	char *errstr;
 
-	node_num = o2nm_this_node();
-
-	ret = 1;
-	slot = &reg->hr_slots[node_num];
+	slot = &reg->hr_slots[o2nm_this_node()];
 	/* Don't check on our 1st timestamp */
-	if (slot->ds_last_time) {
-		hb_block = slot->ds_raw_block;
+	if (!slot->ds_last_time)
+		return;
 
-		if (le64_to_cpu(hb_block->hb_seq) != slot->ds_last_time)
-			ret = 0;
-	}
+	hb_block = slot->ds_raw_block;
+	if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time &&
+	    le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation &&
+	    hb_block->hb_node == slot->ds_node_num)
+		return;
 
-	return ret;
+#define ERRSTR1		"Another node is heartbeating on device"
+#define ERRSTR2		"Heartbeat generation mismatch on device"
+#define ERRSTR3		"Heartbeat sequence mismatch on device"
+
+	if (hb_block->hb_node != slot->ds_node_num)
+		errstr = ERRSTR1;
+	else if (le64_to_cpu(hb_block->hb_generation) !=
+		 slot->ds_last_generation)
+		errstr = ERRSTR2;
+	else
+		errstr = ERRSTR3;
+
+	mlog(ML_ERROR, "%s (%s): expected(%u:0x%llx, 0x%llx), "
+	     "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg->hr_dev_name,
+	     slot->ds_node_num, (unsigned long long)slot->ds_last_generation,
+	     (unsigned long long)slot->ds_last_time, hb_block->hb_node,
+	     (unsigned long long)le64_to_cpu(hb_block->hb_generation),
+	     (unsigned long long)le64_to_cpu(hb_block->hb_seq));
 }
 
 static inline void o2hb_prepare_block(struct o2hb_region *reg,
@@ -983,9 +999,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 	/* With an up to date view of the slots, we can check that no
 	 * other node has been improperly configured to heartbeat in
 	 * our slot. */
-	if (!o2hb_check_last_timestamp(reg))
-		mlog(ML_ERROR, "Device \"%s\": another node is heartbeating "
-		     "in our slot!\n", reg->hr_dev_name);
+	o2hb_check_last_timestamp(reg);
 
 	/* fill in the proper info for our next heartbeat */
 	o2hb_prepare_block(reg, reg->hr_generation);
@@ -999,8 +1013,8 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 	}
 
 	i = -1;
-	while((i = find_next_bit(configured_nodes, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) {
-
+	while((i = find_next_bit(configured_nodes,
+				 O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) {
 		change |= o2hb_check_slot(reg, &reg->hr_slots[i]);
 	}
 
@@ -1690,6 +1704,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg,
 	struct file *filp = NULL;
 	struct inode *inode = NULL;
 	ssize_t ret = -EINVAL;
+	int live_threshold;
 
 	if (reg->hr_bdev)
 		goto out;
@@ -1766,8 +1781,18 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg,
 	 * A node is considered live after it has beat LIVE_THRESHOLD
 	 * times.  We're not steady until we've given them a chance
 	 * _after_ our first read.
+	 * The default threshold is bare minimum so as to limit the delay
+	 * during mounts. For global heartbeat, the threshold doubled for the
+	 * first region.
 	 */
-	atomic_set(&reg->hr_steady_iterations, O2HB_LIVE_THRESHOLD + 1);
+	live_threshold = O2HB_LIVE_THRESHOLD;
+	if (o2hb_global_heartbeat_active()) {
+		spin_lock(&o2hb_live_lock);
+		if (o2hb_pop_count(&o2hb_region_bitmap, O2NM_MAX_REGIONS) == 1)
+			live_threshold <<= 1;
+		spin_unlock(&o2hb_live_lock);
+	}
+	atomic_set(&reg->hr_steady_iterations, live_threshold + 1);
 
 	hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s",
 			      reg->hr_item.ci_name);
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index 9fe5b8fd658..8582e3f4f12 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -2868,7 +2868,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh,
 		bytes = blocks_wanted << sb->s_blocksize_bits;
 	struct ocfs2_super *osb = OCFS2_SB(dir->i_sb);
 	struct ocfs2_inode_info *oi = OCFS2_I(dir);
-	struct ocfs2_alloc_context *data_ac;
+	struct ocfs2_alloc_context *data_ac = NULL;
 	struct ocfs2_alloc_context *meta_ac = NULL;
 	struct buffer_head *dirdata_bh = NULL;
 	struct buffer_head *dx_root_bh = NULL;
diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index 7540a492eab..3b179d6cbde 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -1614,7 +1614,8 @@ static int dlm_try_to_join_domain(struct dlm_ctxt *dlm)
 	spin_unlock(&dlm->spinlock);
 
 	/* Support for global heartbeat and node info was added in 1.1 */
-	if (dlm_protocol.pv_major > 1 || dlm_protocol.pv_minor > 0) {
+	if (dlm->dlm_locking_proto.pv_major > 1 ||
+	    dlm->dlm_locking_proto.pv_minor > 0) {
 		status = dlm_send_nodeinfo(dlm, ctxt->yes_resp_map);
 		if (status) {
 			mlog_errno(status);
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index fede57ed005..84d166328cf 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -2574,6 +2574,9 @@ fail:
 		res->state &= ~DLM_LOCK_RES_MIGRATING;
 		wake = 1;
 		spin_unlock(&res->spinlock);
+		if (dlm_is_host_down(ret))
+			dlm_wait_for_node_death(dlm, target,
+						DLM_NODE_DEATH_WAIT_MAX);
 		goto leave;
 	}
 
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 41565ae5285..89659d6dc20 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1607,6 +1607,9 @@ static void ocfs2_calc_trunc_pos(struct inode *inode,
 	range = le32_to_cpu(rec->e_cpos) + ocfs2_rec_clusters(el, rec);
 
 	if (le32_to_cpu(rec->e_cpos) >= trunc_start) {
+		/*
+		 * remove an entire extent record.
+		 */
 		*trunc_cpos = le32_to_cpu(rec->e_cpos);
 		/*
 		 * Skip holes if any.
@@ -1617,7 +1620,16 @@ static void ocfs2_calc_trunc_pos(struct inode *inode,
 		*blkno = le64_to_cpu(rec->e_blkno);
 		*trunc_end = le32_to_cpu(rec->e_cpos);
 	} else if (range > trunc_start) {
+		/*
+		 * remove a partial extent record, which means we're
+		 * removing the last extent record.
+		 */
 		*trunc_cpos = trunc_start;
+		/*
+		 * skip hole if any.
+		 */
+		if (range < *trunc_end)
+			*trunc_end = range;
 		*trunc_len = *trunc_end - trunc_start;
 		coff = trunc_start - le32_to_cpu(rec->e_cpos);
 		*blkno = le64_to_cpu(rec->e_blkno) +
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index b141a44605c..295d56454e8 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1260,6 +1260,9 @@ void ocfs2_complete_mount_recovery(struct ocfs2_super *osb)
 {
 	struct ocfs2_journal *journal = osb->journal;
 
+	if (ocfs2_is_hard_readonly(osb))
+		return;
+
 	/* No need to queue up our truncate_log as regular cleanup will catch
 	 * that */
 	ocfs2_queue_recovery_completion(journal, osb->slot_num,