Merge branch 'mlog_replace_for_39' of git://repo.or.cz/taoma-kernel into ocfs2-merge-window-fix

141 files changed, 6132 insertions, 3234 deletions

diff --git a/fs/Kconfig b/fs/Kconfig
index 3db9caa57ed..7cb53aafac1 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -47,7 +47,7 @@ config FS_POSIX_ACL
 	def_bool n
 
 config EXPORTFS
-	tristate
+	bool
 
 config FILE_LOCKING
 	bool "Enable POSIX file locking API" if EXPERT
diff --git a/fs/Makefile b/fs/Makefile
index a7f7cef0c0c..ba01202844c 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -48,6 +48,8 @@ obj-$(CONFIG_FS_POSIX_ACL)	+= posix_acl.o xattr_acl.o
 obj-$(CONFIG_NFS_COMMON)	+= nfs_common/
 obj-$(CONFIG_GENERIC_ACL)	+= generic_acl.o
 
+obj-$(CONFIG_FHANDLE)		+= fhandle.o
+
 obj-y				+= quota/
 
 obj-$(CONFIG_PROC_FS)		+= proc/
diff --git a/fs/afs/write.c b/fs/afs/write.c
index 15690bb1d3b..789b3afb342 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -140,6 +140,7 @@ int afs_write_begin(struct file *file, struct address_space *mapping,
 	candidate->first = candidate->last = index;
 	candidate->offset_first = from;
 	candidate->to_last = to;
+	INIT_LIST_HEAD(&candidate->link);
 	candidate->usage = 1;
 	candidate->state = AFS_WBACK_PENDING;
 	init_waitqueue_head(&candidate->waitq);
diff --git a/fs/aio.c b/fs/aio.c
index fc557a3be0a..26869cde395 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -239,15 +239,23 @@ static void __put_ioctx(struct kioctx *ctx)
 	call_rcu(&ctx->rcu_head, ctx_rcu_free);
 }
 
-#define get_ioctx(kioctx) do {						\
-	BUG_ON(atomic_read(&(kioctx)->users) <= 0);			\
-	atomic_inc(&(kioctx)->users);					\
-} while (0)
-#define put_ioctx(kioctx) do {						\
-	BUG_ON(atomic_read(&(kioctx)->users) <= 0);			\
-	if (unlikely(atomic_dec_and_test(&(kioctx)->users))) 		\
-		__put_ioctx(kioctx);					\
-} while (0)
+static inline void get_ioctx(struct kioctx *kioctx)
+{
+	BUG_ON(atomic_read(&kioctx->users) <= 0);
+	atomic_inc(&kioctx->users);
+}
+
+static inline int try_get_ioctx(struct kioctx *kioctx)
+{
+	return atomic_inc_not_zero(&kioctx->users);
+}
+
+static inline void put_ioctx(struct kioctx *kioctx)
+{
+	BUG_ON(atomic_read(&kioctx->users) <= 0);
+	if (unlikely(atomic_dec_and_test(&kioctx->users)))
+		__put_ioctx(kioctx);
+}
 
 /* ioctx_alloc
  *	Allocates and initializes an ioctx.  Returns an ERR_PTR if it failed.
@@ -601,8 +609,13 @@ static struct kioctx *lookup_ioctx(unsigned long ctx_id)
 	rcu_read_lock();
 
 	hlist_for_each_entry_rcu(ctx, n, &mm->ioctx_list, list) {
-		if (ctx->user_id == ctx_id && !ctx->dead) {
-			get_ioctx(ctx);
+		/*
+		 * RCU protects us against accessing freed memory but
+		 * we have to be careful not to get a reference when the
+		 * reference count already dropped to 0 (ctx->dead test
+		 * is unreliable because of races).
+		 */
+		if (ctx->user_id == ctx_id && !ctx->dead && try_get_ioctx(ctx)){
 			ret = ctx;
 			break;
 		}
@@ -1629,6 +1642,23 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
 		goto out_put_req;
 
 	spin_lock_irq(&ctx->ctx_lock);
+	/*
+	 * We could have raced with io_destroy() and are currently holding a
+	 * reference to ctx which should be destroyed. We cannot submit IO
+	 * since ctx gets freed as soon as io_submit() puts its reference.  The
+	 * check here is reliable: io_destroy() sets ctx->dead before waiting
+	 * for outstanding IO and the barrier between these two is realized by
+	 * unlock of mm->ioctx_lock and lock of ctx->ctx_lock.  Analogously we
+	 * increment ctx->reqs_active before checking for ctx->dead and the
+	 * barrier is realized by unlock and lock of ctx->ctx_lock. Thus if we
+	 * don't see ctx->dead set here, io_destroy() waits for our IO to
+	 * finish.
+	 */
+	if (ctx->dead) {
+		spin_unlock_irq(&ctx->ctx_lock);
+		ret = -EINVAL;
+		goto out_put_req;
+	}
 	aio_run_iocb(req);
 	if (!list_empty(&ctx->run_list)) {
 		/* drain the run list */
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 4fb8a343153..88928701959 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -873,6 +873,11 @@ int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk)
 	ret = add_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj);
 	if (ret)
 		goto out_del;
+	/*
+	 * bdev could be deleted beneath us which would implicitly destroy
+	 * the holder directory.  Hold on to it.
+	 */
+	kobject_get(bdev->bd_part->holder_dir);
 
 	list_add(&holder->list, &bdev->bd_holder_disks);
 	goto out_unlock;
@@ -909,6 +914,7 @@ void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk)
 		del_symlink(disk->slave_dir, &part_to_dev(bdev->bd_part)->kobj);
 		del_symlink(bdev->bd_part->holder_dir,
 			    &disk_to_dev(disk)->kobj);
+		kobject_put(bdev->bd_part->holder_dir);
 		list_del_init(&holder->list);
 		kfree(holder);
 	}
@@ -922,14 +928,15 @@ EXPORT_SYMBOL_GPL(bd_unlink_disk_holder);
  * flush_disk - invalidates all buffer-cache entries on a disk
  *
  * @bdev:      struct block device to be flushed
+ * @kill_dirty: flag to guide handling of dirty inodes
  *
  * Invalidates all buffer-cache entries on a disk. It should be called
  * when a disk has been changed -- either by a media change or online
  * resize.
  */
-static void flush_disk(struct block_device *bdev)
+static void flush_disk(struct block_device *bdev, bool kill_dirty)
 {
-	if (__invalidate_device(bdev)) {
+	if (__invalidate_device(bdev, kill_dirty)) {
 		char name[BDEVNAME_SIZE] = "";
 
 		if (bdev->bd_disk)
@@ -966,7 +973,7 @@ void check_disk_size_change(struct gendisk *disk, struct block_device *bdev)
 		       "%s: detected capacity change from %lld to %lld\n",
 		       name, bdev_size, disk_size);
 		i_size_write(bdev->bd_inode, disk_size);
-		flush_disk(bdev);
+		flush_disk(bdev, false);
 	}
 }
 EXPORT_SYMBOL(check_disk_size_change);
@@ -1019,7 +1026,7 @@ int check_disk_change(struct block_device *bdev)
 	if (!(events & DISK_EVENT_MEDIA_CHANGE))
 		return 0;
 
-	flush_disk(bdev);
+	flush_disk(bdev, true);
 	if (bdops->revalidate_disk)
 		bdops->revalidate_disk(bdev->bd_disk);
 	return 1;
@@ -1600,7 +1607,7 @@ fail:
 }
 EXPORT_SYMBOL(lookup_bdev);
 
-int __invalidate_device(struct block_device *bdev)
+int __invalidate_device(struct block_device *bdev, bool kill_dirty)
 {
 	struct super_block *sb = get_super(bdev);
 	int res = 0;
@@ -1613,7 +1620,7 @@ int __invalidate_device(struct block_device *bdev)
 		 * hold).
 		 */
 		shrink_dcache_sb(sb);
-		res = invalidate_inodes(sb);
+		res = invalidate_inodes(sb, kill_dirty);
 		drop_super(sb);
 	}
 	invalidate_bdev(bdev);
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2c98b3af605..7f78cc78fdd 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -729,6 +729,15 @@ struct btrfs_space_info {
 	u64 disk_total;		/* total bytes on disk, takes mirrors into
 				   account */
 
+	/*
+	 * we bump reservation progress every time we decrement
+	 * bytes_reserved.  This way people waiting for reservations
+	 * know something good has happened and they can check
+	 * for progress.  The number here isn't to be trusted, it
+	 * just shows reclaim activity
+	 */
+	unsigned long reservation_progress;
+
 	int full;		/* indicates that we cannot allocate any more
 				   chunks for this space */
 	int force_alloc;	/* set if we need to force a chunk alloc for
@@ -1254,6 +1263,7 @@ struct btrfs_root {
 #define BTRFS_MOUNT_SPACE_CACHE		(1 << 12)
 #define BTRFS_MOUNT_CLEAR_CACHE		(1 << 13)
 #define BTRFS_MOUNT_USER_SUBVOL_RM_ALLOWED (1 << 14)
+#define BTRFS_MOUNT_ENOSPC_DEBUG	 (1 << 15)
 
 #define btrfs_clear_opt(o, opt)		((o) &= ~BTRFS_MOUNT_##opt)
 #define btrfs_set_opt(o, opt)		((o) |= BTRFS_MOUNT_##opt)
@@ -2218,6 +2228,8 @@ int btrfs_error_unpin_extent_range(struct btrfs_root *root,
 				   u64 start, u64 end);
 int btrfs_error_discard_extent(struct btrfs_root *root, u64 bytenr,
 			       u64 num_bytes);
+int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans,
+			    struct btrfs_root *root, u64 type);
 
 /* ctree.c */
 int btrfs_bin_search(struct extent_buffer *eb, struct btrfs_key *key,
diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c
index ff27d7a477b..b4ffad859ad 100644
--- a/fs/btrfs/export.c
+++ b/fs/btrfs/export.c
@@ -21,9 +21,13 @@ static int btrfs_encode_fh(struct dentry *dentry, u32 *fh, int *max_len,
 	int len = *max_len;
 	int type;
 
-	if ((len < BTRFS_FID_SIZE_NON_CONNECTABLE) ||
-	    (connectable && len < BTRFS_FID_SIZE_CONNECTABLE))
+	if (connectable && (len < BTRFS_FID_SIZE_CONNECTABLE)) {
+		*max_len = BTRFS_FID_SIZE_CONNECTABLE;
 		return 255;
+	} else if (len < BTRFS_FID_SIZE_NON_CONNECTABLE) {
+		*max_len = BTRFS_FID_SIZE_NON_CONNECTABLE;
+		return 255;
+	}
 
 	len  = BTRFS_FID_SIZE_NON_CONNECTABLE;
 	type = FILEID_BTRFS_WITHOUT_PARENT;
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index f3c96fc0143..7b3089b5c2d 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -3342,15 +3342,16 @@ static int shrink_delalloc(struct btrfs_trans_handle *trans,
 	u64 max_reclaim;
 	u64 reclaimed = 0;
 	long time_left;
-	int pause = 1;
 	int nr_pages = (2 * 1024 * 1024) >> PAGE_CACHE_SHIFT;
 	int loops = 0;
+	unsigned long progress;
 
 	block_rsv = &root->fs_info->delalloc_block_rsv;
 	space_info = block_rsv->space_info;
 
 	smp_mb();
 	reserved = space_info->bytes_reserved;
+	progress = space_info->reservation_progress;
 
 	if (reserved == 0)
 		return 0;
@@ -3365,31 +3366,36 @@ static int shrink_delalloc(struct btrfs_trans_handle *trans,
 		writeback_inodes_sb_nr_if_idle(root->fs_info->sb, nr_pages);
 
 		spin_lock(&space_info->lock);
-		if (reserved > space_info->bytes_reserved) {
-			loops = 0;
+		if (reserved > space_info->bytes_reserved)
 			reclaimed += reserved - space_info->bytes_reserved;
-		} else {
-			loops++;
-		}
 		reserved = space_info->bytes_reserved;
 		spin_unlock(&space_info->lock);
 
+		loops++;
+
 		if (reserved == 0 || reclaimed >= max_reclaim)
 			break;
 
 		if (trans && trans->transaction->blocked)
 			return -EAGAIN;
 
-		__set_current_state(TASK_INTERRUPTIBLE);
-		time_left = schedule_timeout(pause);
+		time_left = schedule_timeout_interruptible(1);
 
 		/* We were interrupted, exit */
 		if (time_left)
 			break;
 
-		pause <<= 1;
-		if (pause > HZ / 10)
-			pause = HZ / 10;
+		/* we've kicked the IO a few times, if anything has been freed,
+		 * exit.  There is no sense in looping here for a long time
+		 * when we really need to commit the transaction, or there are
+		 * just too many writers without enough free space
+		 */
+
+		if (loops > 3) {
+			smp_mb();
+			if (progress != space_info->reservation_progress)
+				break;
+		}
 
 	}
 	return reclaimed >= to_reclaim;
@@ -3612,6 +3618,7 @@ void block_rsv_release_bytes(struct btrfs_block_rsv *block_rsv,
 		if (num_bytes) {
 			spin_lock(&space_info->lock);
 			space_info->bytes_reserved -= num_bytes;
+			space_info->reservation_progress++;
 			spin_unlock(&space_info->lock);
 		}
 	}
@@ -3844,6 +3851,7 @@ static void update_global_block_rsv(struct btrfs_fs_info *fs_info)
 	if (block_rsv->reserved >= block_rsv->size) {
 		num_bytes = block_rsv->reserved - block_rsv->size;
 		sinfo->bytes_reserved -= num_bytes;
+		sinfo->reservation_progress++;
 		block_rsv->reserved = block_rsv->size;
 		block_rsv->full = 1;
 	}
@@ -4005,7 +4013,6 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes)
 		to_reserve = 0;
 	}
 	spin_unlock(&BTRFS_I(inode)->accounting_lock);
-
 	to_reserve += calc_csum_metadata_size(inode, num_bytes);
 	ret = reserve_metadata_bytes(NULL, root, block_rsv, to_reserve, 1);
 	if (ret)
@@ -4133,6 +4140,7 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 			btrfs_set_block_group_used(&cache->item, old_val);
 			cache->reserved -= num_bytes;
 			cache->space_info->bytes_reserved -= num_bytes;
+			cache->space_info->reservation_progress++;
 			cache->space_info->bytes_used += num_bytes;
 			cache->space_info->disk_used += num_bytes * factor;
 			spin_unlock(&cache->lock);
@@ -4184,6 +4192,7 @@ static int pin_down_extent(struct btrfs_root *root,
 	if (reserved) {
 		cache->reserved -= num_bytes;
 		cache->space_info->bytes_reserved -= num_bytes;
+		cache->space_info->reservation_progress++;
 	}
 	spin_unlock(&cache->lock);
 	spin_unlock(&cache->space_info->lock);
@@ -4234,6 +4243,7 @@ static int update_reserved_bytes(struct btrfs_block_group_cache *cache,
 				space_info->bytes_readonly += num_bytes;
 			cache->reserved -= num_bytes;
 			space_info->bytes_reserved -= num_bytes;
+			space_info->reservation_progress++;
 		}
 		spin_unlock(&cache->lock);
 		spin_unlock(&space_info->lock);
@@ -4712,6 +4722,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans,
 		if (ret) {
 			spin_lock(&cache->space_info->lock);
 			cache->space_info->bytes_reserved -= buf->len;
+			cache->space_info->reservation_progress++;
 			spin_unlock(&cache->space_info->lock);
 		}
 		goto out;
@@ -5376,7 +5387,7 @@ again:
 			       num_bytes, data, 1);
 		goto again;
 	}
-	if (ret == -ENOSPC) {
+	if (ret == -ENOSPC && btrfs_test_opt(root, ENOSPC_DEBUG)) {
 		struct btrfs_space_info *sinfo;
 
 		sinfo = __find_space_info(root->fs_info, data);
@@ -8065,6 +8076,13 @@ out:
 	return ret;
 }
 
+int btrfs_force_chunk_alloc(struct btrfs_trans_handle *trans,
+			    struct btrfs_root *root, u64 type)
+{
+	u64 alloc_flags = get_alloc_profile(root, type);
+	return do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, 1);
+}
+
 /*
  * helper to account the unused space of all the readonly block group in the
  * list. takes mirrors into account.
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 92ac5192c51..714adc4ac4c 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -1433,12 +1433,13 @@ int extent_clear_unlock_delalloc(struct inode *inode,
  */
 u64 count_range_bits(struct extent_io_tree *tree,
 		     u64 *start, u64 search_end, u64 max_bytes,
-		     unsigned long bits)
+		     unsigned long bits, int contig)
 {
 	struct rb_node *node;
 	struct extent_state *state;
 	u64 cur_start = *start;
 	u64 total_bytes = 0;
+	u64 last = 0;
 	int found = 0;
 
 	if (search_end <= cur_start) {
@@ -1463,7 +1464,9 @@ u64 count_range_bits(struct extent_io_tree *tree,
 		state = rb_entry(node, struct extent_state, rb_node);
 		if (state->start > search_end)
 			break;
-		if (state->end >= cur_start && (state->state & bits)) {
+		if (contig && found && state->start > last + 1)
+			break;
+		if (state->end >= cur_start && (state->state & bits) == bits) {
 			total_bytes += min(search_end, state->end) + 1 -
 				       max(cur_start, state->start);
 			if (total_bytes >= max_bytes)
@@ -1472,6 +1475,9 @@ u64 count_range_bits(struct extent_io_tree *tree,
 				*start = state->start;
 				found = 1;
 			}
+			last = state->end;
+		} else if (contig && found) {
+			break;
 		}
 		node = rb_next(node);
 		if (!node)
@@ -2912,6 +2918,46 @@ out:
 	return sector;
 }
 
+/*
+ * helper function for fiemap, which doesn't want to see any holes.
+ * This maps until we find something past 'last'
+ */
+static struct extent_map *get_extent_skip_holes(struct inode *inode,
+						u64 offset,
+						u64 last,
+						get_extent_t *get_extent)
+{
+	u64 sectorsize = BTRFS_I(inode)->root->sectorsize;
+	struct extent_map *em;
+	u64 len;
+
+	if (offset >= last)
+		return NULL;
+
+	while(1) {
+		len = last - offset;
+		if (len == 0)
+			break;
+		len = (len + sectorsize - 1) & ~(sectorsize - 1);
+		em = get_extent(inode, NULL, 0, offset, len, 0);
+		if (!em || IS_ERR(em))
+			return em;
+
+		/* if this isn't a hole return it */
+		if (!test_bit(EXTENT_FLAG_VACANCY, &em->flags) &&
+		    em->block_start != EXTENT_MAP_HOLE) {
+			return em;
+		}
+
+		/* this is a hole, advance to the next extent */
+		offset = extent_map_end(em);
+		free_extent_map(em);
+		if (offset >= last)
+			break;
+	}
+	return NULL;
+}
+
 int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		__u64 start, __u64 len, get_extent_t *get_extent)
 {
@@ -2921,16 +2967,19 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	u32 flags = 0;
 	u32 found_type;
 	u64 last;
+	u64 last_for_get_extent = 0;
 	u64 disko = 0;
+	u64 isize = i_size_read(inode);
 	struct btrfs_key found_key;
 	struct extent_map *em = NULL;
 	struct extent_state *cached_state = NULL;
 	struct btrfs_path *path;
 	struct btrfs_file_extent_item *item;
 	int end = 0;
-	u64 em_start = 0, em_len = 0;
+	u64 em_start = 0;
+	u64 em_len = 0;
+	u64 em_end = 0;
 	unsigned long emflags;
-	int hole = 0;
 
 	if (len == 0)
 		return -EINVAL;
@@ -2940,6 +2989,10 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		return -ENOMEM;
 	path->leave_spinning = 1;
 
+	/*
+	 * lookup the last file extent.  We're not using i_size here
+	 * because there might be preallocation past i_size
+	 */
 	ret = btrfs_lookup_file_extent(NULL, BTRFS_I(inode)->root,
 				       path, inode->i_ino, -1, 0);
 	if (ret < 0) {
@@ -2953,18 +3006,38 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	btrfs_item_key_to_cpu(path->nodes[0], &found_key, path->slots[0]);
 	found_type = btrfs_key_type(&found_key);
 
-	/* No extents, just return */
+	/* No extents, but there might be delalloc bits */
 	if (found_key.objectid != inode->i_ino ||
 	    found_type != BTRFS_EXTENT_DATA_KEY) {
-		btrfs_free_path(path);
-		return 0;
+		/* have to trust i_size as the end */
+		last = (u64)-1;
+		last_for_get_extent = isize;
+	} else {
+		/*
+		 * remember the start of the last extent.  There are a
+		 * bunch of different factors that go into the length of the
+		 * extent, so its much less complex to remember where it started
+		 */
+		last = found_key.offset;
+		last_for_get_extent = last + 1;
 	}
-	last = found_key.offset;
 	btrfs_free_path(path);
 
+	/*
+	 * we might have some extents allocated but more delalloc past those
+	 * extents.  so, we trust isize unless the start of the last extent is
+	 * beyond isize
+	 */
+	if (last < isize) {
+		last = (u64)-1;
+		last_for_get_extent = isize;
+	}
+
 	lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len, 0,
 			 &cached_state, GFP_NOFS);
-	em = get_extent(inode, NULL, 0, off, max - off, 0);
+
+	em = get_extent_skip_holes(inode, off, last_for_get_extent,
+				   get_extent);
 	if (!em)
 		goto out;
 	if (IS_ERR(em)) {
@@ -2973,22 +3046,38 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	}
 
 	while (!end) {
-		hole = 0;
-		off = em->start + em->len;
-		if (off >= max)
-			end = 1;
+		u64 offset_in_extent;
 
-		if (em->block_start == EXTENT_MAP_HOLE) {
-			hole = 1;
-			goto next;
-		}
+		/* break if the extent we found is outside the range */
+		if (em->start >= max || extent_map_end(em) < off)
+			break;
 
-		em_start = em->start;
-		em_len = em->len;
+		/*
+		 * get_extent may return an extent that starts before our
+		 * requested range.  We have to make sure the ranges
+		 * we return to fiemap always move forward and don't
+		 * overlap, so adjust the offsets here
+		 */
+		em_start = max(em->start, off);
 
+		/*
+		 * record the offset from the start of the extent
+		 * for adjusting the disk offset below
+		 */
+		offset_in_extent = em_start - em->start;
+		em_end = extent_map_end(em);
+		em_len = em_end - em_start;
+		emflags = em->flags;
 		disko = 0;
 		flags = 0;
 
+		/*
+		 * bump off for our next call to get_extent
+		 */
+		off = extent_map_end(em);
+		if (off >= max)
+			end = 1;
+
 		if (em->block_start == EXTENT_MAP_LAST_BYTE) {
 			end = 1;
 			flags |= FIEMAP_EXTENT_LAST;
@@ -2999,42 +3088,34 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			flags |= (FIEMAP_EXTENT_DELALLOC |
 				  FIEMAP_EXTENT_UNKNOWN);
 		} else {
-			disko = em->block_start;
+			disko = em->block_start + offset_in_extent;
 		}
 		if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags))
 			flags |= FIEMAP_EXTENT_ENCODED;
 
-next:
-		emflags = em->flags;
 		free_extent_map(em);
 		em = NULL;
-		if (!end) {
-			em = get_extent(inode, NULL, 0, off, max - off, 0);
-			if (!em)
-				goto out;
-			if (IS_ERR(em)) {
-				ret = PTR_ERR(em);
-				goto out;
-			}
-			emflags = em->flags;
-		}
-
-		if (test_bit(EXTENT_FLAG_VACANCY, &emflags)) {
+		if ((em_start >= last) || em_len == (u64)-1 ||
+		   (last == (u64)-1 && isize <= em_end)) {
 			flags |= FIEMAP_EXTENT_LAST;
 			end = 1;
 		}
 
-		if (em_start == last) {
+		/* now scan forward to see if this is really the last extent. */
+		em = get_extent_skip_holes(inode, off, last_for_get_extent,
+					   get_extent);
+		if (IS_ERR(em)) {
+			ret = PTR_ERR(em);
+			goto out;
+		}
+		if (!em) {
 			flags |= FIEMAP_EXTENT_LAST;
 			end = 1;
 		}
-
-		if (!hole) {
-			ret = fiemap_fill_next_extent(fieinfo, em_start, disko,
-						em_len, flags);
-			if (ret)
-				goto out_free;
-		}
+		ret = fiemap_fill_next_extent(fieinfo, em_start, disko,
+					      em_len, flags);
+		if (ret)
+			goto out_free;
 	}
 out_free:
 	free_extent_map(em);
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 7083cfafd06..9318dfefd59 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -191,7 +191,7 @@ void extent_io_exit(void);
 
 u64 count_range_bits(struct extent_io_tree *tree,
 		     u64 *start, u64 search_end,
-		     u64 max_bytes, unsigned long bits);
+		     u64 max_bytes, unsigned long bits, int contig);
 
 void free_extent_state(struct extent_state *state);
 int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end,
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 7084140d594..f447b783bb8 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -70,6 +70,19 @@ static noinline int btrfs_copy_from_user(loff_t pos, int num_pages,
 
 		/* Flush processor's dcache for this page */
 		flush_dcache_page(page);
+
+		/*
+		 * if we get a partial write, we can end up with
+		 * partially up to date pages.  These add
+		 * a lot of complexity, so make sure they don't
+		 * happen by forcing this copy to be retried.
+		 *
+		 * The rest of the btrfs_file_write code will fall
+		 * back to page at a time copies after we return 0.
+		 */
+		if (!PageUptodate(page) && copied < count)
+			copied = 0;
+
 		iov_iter_advance(i, copied);
 		write_bytes -= copied;
 		total_copied += copied;
@@ -763,6 +776,27 @@ out:
 }
 
 /*
+ * on error we return an unlocked page and the error value
+ * on success we return a locked page and 0
+ */
+static int prepare_uptodate_page(struct page *page, u64 pos)
+{
+	int ret = 0;
+
+	if ((pos & (PAGE_CACHE_SIZE - 1)) && !PageUptodate(page)) {
+		ret = btrfs_readpage(NULL, page);
+		if (ret)
+			return ret;
+		lock_page(page);
+		if (!PageUptodate(page)) {
+			unlock_page(page);
+			return -EIO;
+		}
+	}
+	return 0;
+}
+
+/*
  * this gets pages into the page cache and locks them down, it also properly
  * waits for data=ordered extents to finish before allowing the pages to be
  * modified.
@@ -777,6 +811,7 @@ static noinline int prepare_pages(struct btrfs_root *root, struct file *file,
 	unsigned long index = pos >> PAGE_CACHE_SHIFT;
 	struct inode *inode = fdentry(file)->d_inode;
 	int err = 0;
+	int faili = 0;
 	u64 start_pos;
 	u64 last_pos;
 
@@ -794,15 +829,24 @@ again:
 	for (i = 0; i < num_pages; i++) {
 		pages[i] = grab_cache_page(inode->i_mapping, index + i);
 		if (!pages[i]) {
-			int c;
-			for (c = i - 1; c >= 0; c--) {
-				unlock_page(pages[c]);
-				page_cache_release(pages[c]);
-			}
-			return -ENOMEM;
+			faili = i - 1;
+			err = -ENOMEM;
+			goto fail;
+		}
+
+		if (i == 0)
+			err = prepare_uptodate_page(pages[i], pos);
+		if (i == num_pages - 1)
+			err = prepare_uptodate_page(pages[i],
+						    pos + write_bytes);
+		if (err) {
+			page_cache_release(pages[i]);
+			faili = i - 1;
+			goto fail;
 		}
 		wait_on_page_writeback(pages[i]);
 	}
+	err = 0;
 	if (start_pos < inode->i_size) {
 		struct btrfs_ordered_extent *ordered;
 		lock_extent_bits(&BTRFS_I(inode)->io_tree,
@@ -842,6 +886,14 @@ again:
 		WARN_ON(!PageLocked(pages[i]));
 	}
 	return 0;
+fail:
+	while (faili >= 0) {
+		unlock_page(pages[faili]);
+		page_cache_release(pages[faili]);
+		faili--;
+	}
+	return err;
+
 }
 
 static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
@@ -851,7 +903,6 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = fdentry(file)->d_inode;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct page *pinned[2];
 	struct page **pages = NULL;
 	struct iov_iter i;
 	loff_t *ppos = &iocb->ki_pos;
@@ -872,9 +923,6 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 	will_write = ((file->f_flags & O_DSYNC) || IS_SYNC(inode) ||
 		      (file->f_flags & O_DIRECT));
 
-	pinned[0] = NULL;
-	pinned[1] = NULL;
-
 	start_pos = pos;
 
 	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);
@@ -962,32 +1010,6 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 	first_index = pos >> PAGE_CACHE_SHIFT;
 	last_index = (pos + iov_iter_count(&i)) >> PAGE_CACHE_SHIFT;
 
-	/*
-	 * there are lots of better ways to do this, but this code
-	 * makes sure the first and last page in the file range are
-	 * up to date and ready for cow
-	 */
-	if ((pos & (PAGE_CACHE_SIZE - 1))) {
-		pinned[0] = grab_cache_page(inode->i_mapping, first_index);
-		if (!PageUptodate(pinned[0])) {
-			ret = btrfs_readpage(NULL, pinned[0]);
-			BUG_ON(ret);
-			wait_on_page_locked(pinned[0]);
-		} else {
-			unlock_page(pinned[0]);
-		}
-	}
-	if ((pos + iov_iter_count(&i)) & (PAGE_CACHE_SIZE - 1)) {
-		pinned[1] = grab_cache_page(inode->i_mapping, last_index);
-		if (!PageUptodate(pinned[1])) {
-			ret = btrfs_readpage(NULL, pinned[1]);
-			BUG_ON(ret);
-			wait_on_page_locked(pinned[1]);
-		} else {
-			unlock_page(pinned[1]);
-		}
-	}
-
 	while (iov_iter_count(&i) > 0) {
 		size_t offset = pos & (PAGE_CACHE_SIZE - 1);
 		size_t write_bytes = min(iov_iter_count(&i),
@@ -1024,8 +1046,20 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb,
 
 		copied = btrfs_copy_from_user(pos, num_pages,
 					   write_bytes, pages, &i);
-		dirty_pages = (copied + offset + PAGE_CACHE_SIZE - 1) >>
-				PAGE_CACHE_SHIFT;
+
+		/*
+		 * if we have trouble faulting in the pages, fall
+		 * back to one page at a time
+		 */
+		if (copied < write_bytes)
+			nrptrs = 1;
+
+		if (copied == 0)
+			dirty_pages = 0;
+		else
+			dirty_pages = (copied + offset +
+				       PAGE_CACHE_SIZE - 1) >>
+				       PAGE_CACHE_SHIFT;
 
 		if (num_pages > dirty_pages) {
 			if (copied > 0)
@@ -1069,10 +1103,6 @@ out:
 		err = ret;
 
 	kfree(pages);
-	if (pinned[0])
-		page_cache_release(pinned[0]);
-	if (pinned[1])
-		page_cache_release(pinned[1]);
 	*ppos = pos;
 
 	/*
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index fb9bd7832b6..4a0107e1874 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1913,7 +1913,7 @@ static int btrfs_clean_io_failures(struct inode *inode, u64 start)
 
 	private = 0;
 	if (count_range_bits(&BTRFS_I(inode)->io_failure_tree, &private,
-			     (u64)-1, 1, EXTENT_DIRTY)) {
+			     (u64)-1, 1, EXTENT_DIRTY, 0)) {
 		ret = get_state_private(&BTRFS_I(inode)->io_failure_tree,
 					start, &private_failure);
 		if (ret == 0) {
@@ -4806,9 +4806,6 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	int err;
 	int drop_inode = 0;
 
-	if (inode->i_nlink == 0)
-		return -ENOENT;
-
 	/* do not allow sys_link's with other subvols of the same device */
 	if (root->objectid != BTRFS_I(inode)->root->objectid)
 		return -EPERM;
@@ -4821,10 +4818,11 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 		goto fail;
 
 	/*
-	 * 1 item for inode ref
+	 * 2 items for inode and inode ref
 	 * 2 items for dir items
+	 * 1 item for parent inode
 	 */
-	trans = btrfs_start_transaction(root, 3);
+	trans = btrfs_start_transaction(root, 5);
 	if (IS_ERR(trans)) {
 		err = PTR_ERR(trans);
 		goto fail;
@@ -5280,6 +5278,128 @@ out:
 	return em;
 }
 
+struct extent_map *btrfs_get_extent_fiemap(struct inode *inode, struct page *page,
+					   size_t pg_offset, u64 start, u64 len,
+					   int create)
+{
+	struct extent_map *em;
+	struct extent_map *hole_em = NULL;
+	u64 range_start = start;
+	u64 end;
+	u64 found;
+	u64 found_end;
+	int err = 0;
+
+	em = btrfs_get_extent(inode, page, pg_offset, start, len, create);
+	if (IS_ERR(em))
+		return em;
+	if (em) {
+		/*
+		 * if our em maps to a hole, there might
+		 * actually be delalloc bytes behind it
+		 */
+		if (em->block_start != EXTENT_MAP_HOLE)
+			return em;
+		else
+			hole_em = em;
+	}
+
+	/* check to see if we've wrapped (len == -1 or similar) */
+	end = start + len;
+	if (end < start)
+		end = (u64)-1;
+	else
+		end -= 1;
+
+	em = NULL;
+
+	/* ok, we didn't find anything, lets look for delalloc */
+	found = count_range_bits(&BTRFS_I(inode)->io_tree, &range_start,
+				 end, len, EXTENT_DELALLOC, 1);
+	found_end = range_start + found;
+	if (found_end < range_start)
+		found_end = (u64)-1;
+
+	/*
+	 * we didn't find anything useful, return
+	 * the original results from get_extent()
+	 */
+	if (range_start > end || found_end <= start) {
+		em = hole_em;
+		hole_em = NULL;
+		goto out;
+	}
+
+	/* adjust the range_start to make sure it doesn't
+	 * go backwards from the start they passed in
+	 */
+	range_start = max(start,range_start);
+	found = found_end - range_start;
+
+	if (found > 0) {
+		u64 hole_start = start;
+		u64 hole_len = len;
+
+		em = alloc_extent_map(GFP_NOFS);
+		if (!em) {
+			err = -ENOMEM;
+			goto out;
+		}
+		/*
+		 * when btrfs_get_extent can't find anything it
+		 * returns one huge hole
+		 *
+		 * make sure what it found really fits our range, and
+		 * adjust to make sure it is based on the start from
+		 * the caller
+		 */
+		if (hole_em) {
+			u64 calc_end = extent_map_end(hole_em);
+
+			if (calc_end <= start || (hole_em->start > end)) {
+				free_extent_map(hole_em);
+				hole_em = NULL;
+			} else {
+				hole_start = max(hole_em->start, start);
+				hole_len = calc_end - hole_start;
+			}
+		}
+		em->bdev = NULL;
+		if (hole_em && range_start > hole_start) {
+			/* our hole starts before our delalloc, so we
+			 * have to return just the parts of the hole
+			 * that go until  the delalloc starts
+			 */
+			em->len = min(hole_len,
+				      range_start - hole_start);
+			em->start = hole_start;
+			em->orig_start = hole_start;
+			/*
+			 * don't adjust block start at all,
+			 * it is fixed at EXTENT_MAP_HOLE
+			 */
+			em->block_start = hole_em->block_start;
+			em->block_len = hole_len;
+		} else {
+			em->start = range_start;
+			em->len = found;
+			em->orig_start = range_start;
+			em->block_start = EXTENT_MAP_DELALLOC;
+			em->block_len = found;
+		}
+	} else if (hole_em) {
+		return hole_em;
+	}
+out:
+
+	free_extent_map(hole_em);
+	if (err) {
+		free_extent_map(em);
+		return ERR_PTR(err);
+	}
+	return em;
+}
+
 static struct extent_map *btrfs_new_extent_direct(struct inode *inode,
 						  u64 start, u64 len)
 {
@@ -5934,6 +6054,7 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode,
 	if (!skip_sum) {
 		dip->csums = kmalloc(sizeof(u32) * bio->bi_vcnt, GFP_NOFS);
 		if (!dip->csums) {
+			kfree(dip);
 			ret = -ENOMEM;
 			goto free_ordered;
 		}
@@ -6102,7 +6223,7 @@ out:
 static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		__u64 start, __u64 len)
 {
-	return extent_fiemap(inode, fieinfo, start, len, btrfs_get_extent);
+	return extent_fiemap(inode, fieinfo, start, len, btrfs_get_extent_fiemap);
 }
 
 int btrfs_readpage(struct file *file, struct page *page)
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index be2d4f6aaa5..5fdb2abc4fa 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1071,12 +1071,15 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file,
 	if (copy_from_user(&flags, arg, sizeof(flags)))
 		return -EFAULT;
 
-	if (flags & ~BTRFS_SUBVOL_CREATE_ASYNC)
+	if (flags & BTRFS_SUBVOL_CREATE_ASYNC)
 		return -EINVAL;
 
 	if (flags & ~BTRFS_SUBVOL_RDONLY)
 		return -EOPNOTSUPP;
 
+	if (!is_owner_or_cap(inode))
+		return -EACCES;
+
 	down_write(&root->fs_info->subvol_sem);
 
 	/* nothing to do */
@@ -1097,7 +1100,7 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file,
 		goto out_reset;
 	}
 
-	ret = btrfs_update_root(trans, root,
+	ret = btrfs_update_root(trans, root->fs_info->tree_root,
 				&root->root_key, &root->root_item);
 
 	btrfs_commit_transaction(trans, root);
diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c
index cc9b450399d..a178f5ebea7 100644
--- a/fs/btrfs/lzo.c
+++ b/fs/btrfs/lzo.c
@@ -280,6 +280,7 @@ static int lzo_decompress_biovec(struct list_head *ws,
 	unsigned long tot_out;
 	unsigned long tot_len;
 	char *buf;
+	bool may_late_unmap, need_unmap;
 
 	data_in = kmap(pages_in[0]);
 	tot_len = read_compress_length(data_in);
@@ -300,11 +301,13 @@ static int lzo_decompress_biovec(struct list_head *ws,
 
 		tot_in += in_len;
 		working_bytes = in_len;
+		may_late_unmap = need_unmap = false;
 
 		/* fast path: avoid using the working buffer */
 		if (in_page_bytes_left >= in_len) {
 			buf = data_in + in_offset;
 			bytes = in_len;
+			may_late_unmap = true;
 			goto cont;
 		}
 
@@ -329,14 +332,17 @@ cont:
 				if (working_bytes == 0 && tot_in >= tot_len)
 					break;
 
-				kunmap(pages_in[page_in_index]);
-				page_in_index++;
-				if (page_in_index >= total_pages_in) {
+				if (page_in_index + 1 >= total_pages_in) {
 					ret = -1;
-					data_in = NULL;
 					goto done;
 				}
-				data_in = kmap(pages_in[page_in_index]);
+
+				if (may_late_unmap)
+					need_unmap = true;
+				else
+					kunmap(pages_in[page_in_index]);
+
+				data_in = kmap(pages_in[++page_in_index]);
 
 				in_page_bytes_left = PAGE_CACHE_SIZE;
 				in_offset = 0;
@@ -346,6 +352,8 @@ cont:
 		out_len = lzo1x_worst_compress(PAGE_CACHE_SIZE);
 		ret = lzo1x_decompress_safe(buf, in_len, workspace->buf,
 					    &out_len);
+		if (need_unmap)
+			kunmap(pages_in[page_in_index - 1]);
 		if (ret != LZO_E_OK) {
 			printk(KERN_WARNING "btrfs decompress failed\n");
 			ret = -1;
@@ -363,8 +371,7 @@ cont:
 			break;
 	}
 done:
-	if (data_in)
-		kunmap(pages_in[page_in_index]);
+	kunmap(pages_in[page_in_index]);
 	return ret;
 }
 
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 0825e4ed944..31ade5802ae 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3654,6 +3654,7 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
 	u32 item_size;
 	int ret;
 	int err = 0;
+	int progress = 0;
 
 	path = btrfs_alloc_path();
 	if (!path)
@@ -3666,9 +3667,10 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
 	}
 
 	while (1) {
+		progress++;
 		trans = btrfs_start_transaction(rc->extent_root, 0);
 		BUG_ON(IS_ERR(trans));
-
+restart:
 		if (update_backref_cache(trans, &rc->backref_cache)) {
 			btrfs_end_transaction(trans, rc->extent_root);
 			continue;
@@ -3781,6 +3783,15 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
 			}
 		}
 	}
+	if (trans && progress && err == -ENOSPC) {
+		ret = btrfs_force_chunk_alloc(trans, rc->extent_root,
+					      rc->block_group->flags);
+		if (ret == 0) {
+			err = 0;
+			progress = 0;
+			goto restart;
+		}
+	}
 
 	btrfs_release_path(rc->extent_root, path);
 	clear_extent_bits(&rc->processed_blocks, 0, (u64)-1, EXTENT_DIRTY,
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index a004008f7d2..d39a9895d93 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -155,7 +155,8 @@ enum {
 	Opt_nossd, Opt_ssd_spread, Opt_thread_pool, Opt_noacl, Opt_compress,
 	Opt_compress_type, Opt_compress_force, Opt_compress_force_type,
 	Opt_notreelog, Opt_ratio, Opt_flushoncommit, Opt_discard,
-	Opt_space_cache, Opt_clear_cache, Opt_user_subvol_rm_allowed, Opt_err,
+	Opt_space_cache, Opt_clear_cache, Opt_user_subvol_rm_allowed,
+	Opt_enospc_debug, Opt_err,
 };
 
 static match_table_t tokens = {
@@ -184,6 +185,7 @@ static match_table_t tokens = {
 	{Opt_space_cache, "space_cache"},
 	{Opt_clear_cache, "clear_cache"},
 	{Opt_user_subvol_rm_allowed, "user_subvol_rm_allowed"},
+	{Opt_enospc_debug, "enospc_debug"},
 	{Opt_err, NULL},
 };
 
@@ -358,6 +360,9 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 		case Opt_user_subvol_rm_allowed:
 			btrfs_set_opt(info->mount_opt, USER_SUBVOL_RM_ALLOWED);
 			break;
+		case Opt_enospc_debug:
+			btrfs_set_opt(info->mount_opt, ENOSPC_DEBUG);
+			break;
 		case Opt_err:
 			printk(KERN_INFO "btrfs: unrecognized mount option "
 			       "'%s'\n", p);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index af7dbca1527..dd13eb81ee4 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1338,11 +1338,11 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
 
 	ret = btrfs_shrink_device(device, 0);
 	if (ret)
-		goto error_brelse;
+		goto error_undo;
 
 	ret = btrfs_rm_dev_item(root->fs_info->chunk_root, device);
 	if (ret)
-		goto error_brelse;
+		goto error_undo;
 
 	device->in_fs_metadata = 0;
 
@@ -1416,6 +1416,13 @@ out:
 	mutex_unlock(&root->fs_info->volume_mutex);
 	mutex_unlock(&uuid_mutex);
 	return ret;
+error_undo:
+	if (device->writeable) {
+		list_add(&device->dev_alloc_list,
+			 &root->fs_info->fs_devices->alloc_list);
+		root->fs_info->fs_devices->rw_devices++;
+	}
+	goto error_brelse;
 }
 
 /*
@@ -1633,7 +1640,7 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path)
 	device->dev_root = root->fs_info->dev_root;
 	device->bdev = bdev;
 	device->in_fs_metadata = 1;
-	device->mode = 0;
+	device->mode = FMODE_EXCL;
 	set_blocksize(device->bdev, 4096);
 
 	if (seeding_dev) {
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 0bc68de8edd..ebafa65a29b 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -409,7 +409,7 @@ more:
 	spin_lock(&inode->i_lock);
 	if (ci->i_release_count == fi->dir_release_count) {
 		dout(" marking %p complete\n", inode);
-		ci->i_ceph_flags |= CEPH_I_COMPLETE;
+		/* ci->i_ceph_flags |= CEPH_I_COMPLETE; */
 		ci->i_max_offset = filp->f_pos;
 	}
 	spin_unlock(&inode->i_lock);
@@ -496,6 +496,7 @@ struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
 
 	/* .snap dir? */
 	if (err == -ENOENT &&
+	    ceph_snap(parent) == CEPH_NOSNAP &&
 	    strcmp(dentry->d_name.name,
 		   fsc->mount_options->snapdir_name) == 0) {
 		struct inode *inode = ceph_get_snapdir(parent);
@@ -992,7 +993,7 @@ static int ceph_d_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
 	struct inode *dir;
 
-	if (nd->flags & LOOKUP_RCU)
+	if (nd && nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 
 	dir = dentry->d_parent->d_inode;
@@ -1029,28 +1030,8 @@ out_touch:
 static void ceph_dentry_release(struct dentry *dentry)
 {
 	struct ceph_dentry_info *di = ceph_dentry(dentry);
-	struct inode *parent_inode = NULL;
-	u64 snapid = CEPH_NOSNAP;
 
-	if (!IS_ROOT(dentry)) {
-		parent_inode = dentry->d_parent->d_inode;
-		if (parent_inode)
-			snapid = ceph_snap(parent_inode);
-	}
-	dout("dentry_release %p parent %p\n", dentry, parent_inode);
-	if (parent_inode && snapid != CEPH_SNAPDIR) {
-		struct ceph_inode_info *ci = ceph_inode(parent_inode);
-
-		spin_lock(&parent_inode->i_lock);
-		if (ci->i_shared_gen == di->lease_shared_gen ||
-		    snapid <= CEPH_MAXSNAP) {
-			dout(" clearing %p complete (d_release)\n",
-			     parent_inode);
-			ci->i_ceph_flags &= ~CEPH_I_COMPLETE;
-			ci->i_release_count++;
-		}
-		spin_unlock(&parent_inode->i_lock);
-	}
+	dout("dentry_release %p\n", dentry);
 	if (di) {
 		ceph_dentry_lru_del(dentry);
 		if (di->lease_session)
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 5625463aa47..193bfa5e9cb 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -707,7 +707,7 @@ static int fill_inode(struct inode *inode,
 		    (issued & CEPH_CAP_FILE_EXCL) == 0 &&
 		    (ci->i_ceph_flags & CEPH_I_COMPLETE) == 0) {
 			dout(" marking %p complete (empty)\n", inode);
-			ci->i_ceph_flags |= CEPH_I_COMPLETE;
+			/* ci->i_ceph_flags |= CEPH_I_COMPLETE; */
 			ci->i_max_offset = 2;
 		}
 		break;
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index 39c243acd06..f40b9139e43 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -584,10 +584,14 @@ static void queue_realm_cap_snaps(struct ceph_snap_realm *realm)
 	if (lastinode)
 		iput(lastinode);
 
-	dout("queue_realm_cap_snaps %p %llx children\n", realm, realm->ino);
-	list_for_each_entry(child, &realm->children, child_item)
-		queue_realm_cap_snaps(child);
+	list_for_each_entry(child, &realm->children, child_item) {
+		dout("queue_realm_cap_snaps %p %llx queue child %p %llx\n",
+		     realm, realm->ino, child, child->ino);
+		list_del_init(&child->dirty_item);
+		list_add(&child->dirty_item, &realm->dirty_item);
+	}
 
+	list_del_init(&realm->dirty_item);
 	dout("queue_realm_cap_snaps %p %llx done\n", realm, realm->ino);
 }
 
@@ -683,7 +687,9 @@ more:
 	 * queue cap snaps _after_ we've built the new snap contexts,
 	 * so that i_head_snapc can be set appropriately.
 	 */
-	list_for_each_entry(realm, &dirty_realms, dirty_item) {
+	while (!list_empty(&dirty_realms)) {
+		realm = list_first_entry(&dirty_realms, struct ceph_snap_realm,
+					 dirty_item);
 		queue_realm_cap_snaps(realm);
 	}
 
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 4a3330235d5..a9371b6578c 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -127,5 +127,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
 extern const struct export_operations cifs_export_ops;
 #endif /* EXPERIMENTAL */
 
-#define CIFS_VERSION   "1.70"
+#define CIFS_VERSION   "1.71"
 #endif				/* _CIFSFS_H */
diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c
index 8d9189f6447..79f641eeda3 100644
--- a/fs/cifs/netmisc.c
+++ b/fs/cifs/netmisc.c
@@ -170,7 +170,7 @@ cifs_convert_address(struct sockaddr *dst, const char *src, int len)
 {
 	int rc, alen, slen;
 	const char *pct;
-	char *endp, scope_id[13];
+	char scope_id[13];
 	struct sockaddr_in *s4 = (struct sockaddr_in *) dst;
 	struct sockaddr_in6 *s6 = (struct sockaddr_in6 *) dst;
 
@@ -197,9 +197,9 @@ cifs_convert_address(struct sockaddr *dst, const char *src, int len)
 		memcpy(scope_id, pct + 1, slen);
 		scope_id[slen] = '\0';
 
-		s6->sin6_scope_id = (u32) simple_strtoul(pct, &endp, 0);
-		if (endp != scope_id + slen)
-			return 0;
+		rc = strict_strtoul(scope_id, 0,
+					(unsigned long *)&s6->sin6_scope_id);
+		rc = (rc == 0) ? 1 : 0;
 	}
 
 	return rc;
diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c
index 1adc9625a34..16765703131 100644
--- a/fs/cifs/sess.c
+++ b/fs/cifs/sess.c
@@ -656,13 +656,13 @@ ssetup_ntlmssp_authenticate:
 
 	if (type == LANMAN) {
 #ifdef CONFIG_CIFS_WEAK_PW_HASH
-		char lnm_session_key[CIFS_SESS_KEY_SIZE];
+		char lnm_session_key[CIFS_AUTH_RESP_SIZE];
 
 		pSMB->req.hdr.Flags2 &= ~SMBFLG2_UNICODE;
 
 		/* no capabilities flags in old lanman negotiation */
 
-		pSMB->old_req.PasswordLength = cpu_to_le16(CIFS_SESS_KEY_SIZE);
+		pSMB->old_req.PasswordLength = cpu_to_le16(CIFS_AUTH_RESP_SIZE);
 
 		/* Calculate hash with password and copy into bcc_ptr.
 		 * Encryption Key (stored as in cryptkey) gets used if the
@@ -675,8 +675,8 @@ ssetup_ntlmssp_authenticate:
 					true : false, lnm_session_key);
 
 		ses->flags |= CIFS_SES_LANMAN;
-		memcpy(bcc_ptr, (char *)lnm_session_key, CIFS_SESS_KEY_SIZE);
-		bcc_ptr += CIFS_SESS_KEY_SIZE;
+		memcpy(bcc_ptr, (char *)lnm_session_key, CIFS_AUTH_RESP_SIZE);
+		bcc_ptr += CIFS_AUTH_RESP_SIZE;
 
 		/* can not sign if LANMAN negotiated so no need
 		to calculate signing key? but what if server
diff --git a/fs/compat.c b/fs/compat.c
index f6fd0a00e6c..c6d31a3bab8 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -262,35 +262,19 @@ static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs *
  */
 asmlinkage long compat_sys_statfs(const char __user *pathname, struct compat_statfs __user *buf)
 {
-	struct path path;
-	int error;
-
-	error = user_path(pathname, &path);
-	if (!error) {
-		struct kstatfs tmp;
-		error = vfs_statfs(&path, &tmp);
-		if (!error)
-			error = put_compat_statfs(buf, &tmp);
-		path_put(&path);
-	}
+	struct kstatfs tmp;
+	int error = user_statfs(pathname, &tmp);
+	if (!error)
+		error = put_compat_statfs(buf, &tmp);
 	return error;
 }
 
 asmlinkage long compat_sys_fstatfs(unsigned int fd, struct compat_statfs __user *buf)
 {
-	struct file * file;
 	struct kstatfs tmp;
-	int error;
-
-	error = -EBADF;
-	file = fget(fd);
-	if (!file)
-		goto out;
-	error = vfs_statfs(&file->f_path, &tmp);
+	int error = fd_statfs(fd, &tmp);
 	if (!error)
 		error = put_compat_statfs(buf, &tmp);
-	fput(file);
-out:
 	return error;
 }
 
@@ -329,41 +313,29 @@ static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstat
 
 asmlinkage long compat_sys_statfs64(const char __user *pathname, compat_size_t sz, struct compat_statfs64 __user *buf)
 {
-	struct path path;
+	struct kstatfs tmp;
 	int error;
 
 	if (sz != sizeof(*buf))
 		return -EINVAL;
 
-	error = user_path(pathname, &path);
-	if (!error) {
-		struct kstatfs tmp;
-		error = vfs_statfs(&path, &tmp);
-		if (!error)
-			error = put_compat_statfs64(buf, &tmp);
-		path_put(&path);
-	}
+	error = user_statfs(pathname, &tmp);
+	if (!error)
+		error = put_compat_statfs64(buf, &tmp);
 	return error;
 }
 
 asmlinkage long compat_sys_fstatfs64(unsigned int fd, compat_size_t sz, struct compat_statfs64 __user *buf)
 {
-	struct file * file;
 	struct kstatfs tmp;
 	int error;
 
 	if (sz != sizeof(*buf))
 		return -EINVAL;
 
-	error = -EBADF;
-	file = fget(fd);
-	if (!file)
-		goto out;
-	error = vfs_statfs(&file->f_path, &tmp);
+	error = fd_statfs(fd, &tmp);
 	if (!error)
 		error = put_compat_statfs64(buf, &tmp);
-	fput(file);
-out:
 	return error;
 }
 
@@ -1228,7 +1200,9 @@ compat_sys_preadv(unsigned long fd, const struct compat_iovec __user *vec,
 	file = fget_light(fd, &fput_needed);
 	if (!file)
 		return -EBADF;
-	ret = compat_readv(file, vec, vlen, &pos);
+	ret = -ESPIPE;
+	if (file->f_mode & FMODE_PREAD)
+		ret = compat_readv(file, vec, vlen, &pos);
 	fput_light(file, fput_needed);
 	return ret;
 }
@@ -1285,7 +1259,9 @@ compat_sys_pwritev(unsigned long fd, const struct compat_iovec __user *vec,
 	file = fget_light(fd, &fput_needed);
 	if (!file)
 		return -EBADF;
-	ret = compat_writev(file, vec, vlen, &pos);
+	ret = -ESPIPE;
+	if (file->f_mode & FMODE_PWRITE)
+		ret = compat_writev(file, vec, vlen, &pos);
 	fput_light(file, fput_needed);
 	return ret;
 }
@@ -2308,3 +2284,16 @@ asmlinkage long compat_sys_timerfd_gettime(int ufd,
 }
 
 #endif /* CONFIG_TIMERFD */
+
+#ifdef CONFIG_FHANDLE
+/*
+ * Exactly like fs/open.c:sys_open_by_handle_at(), except that it
+ * doesn't set the O_LARGEFILE flag.
+ */
+asmlinkage long
+compat_sys_open_by_handle_at(int mountdirfd,
+			     struct file_handle __user *handle, int flags)
+{
+	return do_handle_open(mountdirfd, handle, flags);
+}
+#endif
diff --git a/fs/dcache.c b/fs/dcache.c
index 2a6bd9a4ae9..a39fe47c466 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -296,8 +296,12 @@ static struct dentry *d_kill(struct dentry *dentry, struct dentry *parent)
 	__releases(parent->d_lock)
 	__releases(dentry->d_inode->i_lock)
 {
-	dentry->d_parent = NULL;
 	list_del(&dentry->d_u.d_child);
+	/*
+	 * Inform try_to_ascend() that we are no longer attached to the
+	 * dentry tree
+	 */
+	dentry->d_flags |= DCACHE_DISCONNECTED;
 	if (parent)
 		spin_unlock(&parent->d_lock);
 	dentry_iput(dentry);
@@ -1012,6 +1016,35 @@ void shrink_dcache_for_umount(struct super_block *sb)
 }
 
 /*
+ * This tries to ascend one level of parenthood, but
+ * we can race with renaming, so we need to re-check
+ * the parenthood after dropping the lock and check
+ * that the sequence number still matches.
+ */
+static struct dentry *try_to_ascend(struct dentry *old, int locked, unsigned seq)
+{
+	struct dentry *new = old->d_parent;
+
+	rcu_read_lock();
+	spin_unlock(&old->d_lock);
+	spin_lock(&new->d_lock);
+
+	/*
+	 * might go back up the wrong parent if we have had a rename
+	 * or deletion
+	 */
+	if (new != old->d_parent ||
+		 (old->d_flags & DCACHE_DISCONNECTED) ||
+		 (!locked && read_seqretry(&rename_lock, seq))) {
+		spin_unlock(&new->d_lock);
+		new = NULL;
+	}
+	rcu_read_unlock();
+	return new;
+}
+
+
+/*
  * Search for at least 1 mount point in the dentry's subdirs.
  * We descend to the next level whenever the d_subdirs
  * list is non-empty and continue searching.
@@ -1066,24 +1099,10 @@ resume:
 	 * All done at this level ... ascend and resume the search.
 	 */
 	if (this_parent != parent) {
-		struct dentry *tmp;
-		struct dentry *child;
-
-		tmp = this_parent->d_parent;
-		rcu_read_lock();
-		spin_unlock(&this_parent->d_lock);
-		child = this_parent;
-		this_parent = tmp;
-		spin_lock(&this_parent->d_lock);
-		/* might go back up the wrong parent if we have had a rename
-		 * or deletion */
-		if (this_parent != child->d_parent ||
-			 (!locked && read_seqretry(&rename_lock, seq))) {
-			spin_unlock(&this_parent->d_lock);
-			rcu_read_unlock();
+		struct dentry *child = this_parent;
+		this_parent = try_to_ascend(this_parent, locked, seq);
+		if (!this_parent)
 			goto rename_retry;
-		}
-		rcu_read_unlock();
 		next = child->d_u.d_child.next;
 		goto resume;
 	}
@@ -1181,24 +1200,10 @@ resume:
 	 * All done at this level ... ascend and resume the search.
 	 */
 	if (this_parent != parent) {
-		struct dentry *tmp;
-		struct dentry *child;
-
-		tmp = this_parent->d_parent;
-		rcu_read_lock();
-		spin_unlock(&this_parent->d_lock);
-		child = this_parent;
-		this_parent = tmp;
-		spin_lock(&this_parent->d_lock);
-		/* might go back up the wrong parent if we have had a rename
-		 * or deletion */
-		if (this_parent != child->d_parent ||
-			(!locked && read_seqretry(&rename_lock, seq))) {
-			spin_unlock(&this_parent->d_lock);
-			rcu_read_unlock();
+		struct dentry *child = this_parent;
+		this_parent = try_to_ascend(this_parent, locked, seq);
+		if (!this_parent)
 			goto rename_retry;
-		}
-		rcu_read_unlock();
 		next = child->d_u.d_child.next;
 		goto resume;
 	}
@@ -1523,6 +1528,28 @@ struct dentry * d_alloc_root(struct inode * root_inode)
 }
 EXPORT_SYMBOL(d_alloc_root);
 
+static struct dentry * __d_find_any_alias(struct inode *inode)
+{
+	struct dentry *alias;
+
+	if (list_empty(&inode->i_dentry))
+		return NULL;
+	alias = list_first_entry(&inode->i_dentry, struct dentry, d_alias);
+	__dget(alias);
+	return alias;
+}
+
+static struct dentry * d_find_any_alias(struct inode *inode)
+{
+	struct dentry *de;
+
+	spin_lock(&inode->i_lock);
+	de = __d_find_any_alias(inode);
+	spin_unlock(&inode->i_lock);
+	return de;
+}
+
+
 /**
  * d_obtain_alias - find or allocate a dentry for a given inode
  * @inode: inode to allocate the dentry for
@@ -1552,7 +1579,7 @@ struct dentry *d_obtain_alias(struct inode *inode)
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
 
-	res = d_find_alias(inode);
+	res = d_find_any_alias(inode);
 	if (res)
 		goto out_iput;
 
@@ -1565,7 +1592,7 @@ struct dentry *d_obtain_alias(struct inode *inode)
 
 
 	spin_lock(&inode->i_lock);
-	res = __d_find_alias(inode, 0);
+	res = __d_find_any_alias(inode);
 	if (res) {
 		spin_unlock(&inode->i_lock);
 		dput(tmp);
@@ -2920,28 +2947,14 @@ resume:
 		spin_unlock(&dentry->d_lock);
 	}
 	if (this_parent != root) {
-		struct dentry *tmp;
-		struct dentry *child;
-
-		tmp = this_parent->d_parent;
+		struct dentry *child = this_parent;
 		if (!(this_parent->d_flags & DCACHE_GENOCIDE)) {
 			this_parent->d_flags |= DCACHE_GENOCIDE;
 			this_parent->d_count--;
 		}
-		rcu_read_lock();
-		spin_unlock(&this_parent->d_lock);
-		child = this_parent;
-		this_parent = tmp;
-		spin_lock(&this_parent->d_lock);
-		/* might go back up the wrong parent if we have had a rename
-		 * or deletion */
-		if (this_parent != child->d_parent ||
-			 (!locked && read_seqretry(&rename_lock, seq))) {
-			spin_unlock(&this_parent->d_lock);
-			rcu_read_unlock();
+		this_parent = try_to_ascend(this_parent, locked, seq);
+		if (!this_parent)
 			goto rename_retry;
-		}
-		rcu_read_unlock();
 		next = child->d_u.d_child.next;
 		goto resume;
 	}
diff --git a/fs/ecryptfs/dentry.c b/fs/ecryptfs/dentry.c
index 6fc4f319b55..534c1d46e69 100644
--- a/fs/ecryptfs/dentry.c
+++ b/fs/ecryptfs/dentry.c
@@ -46,24 +46,28 @@ static int ecryptfs_d_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
 	struct dentry *lower_dentry;
 	struct vfsmount *lower_mnt;
-	struct dentry *dentry_save;
-	struct vfsmount *vfsmount_save;
+	struct dentry *dentry_save = NULL;
+	struct vfsmount *vfsmount_save = NULL;
 	int rc = 1;
 
-	if (nd->flags & LOOKUP_RCU)
+	if (nd && nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 
 	lower_dentry = ecryptfs_dentry_to_lower(dentry);
 	lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry);
 	if (!lower_dentry->d_op || !lower_dentry->d_op->d_revalidate)
 		goto out;
-	dentry_save = nd->path.dentry;
-	vfsmount_save = nd->path.mnt;
-	nd->path.dentry = lower_dentry;
-	nd->path.mnt = lower_mnt;
+	if (nd) {
+		dentry_save = nd->path.dentry;
+		vfsmount_save = nd->path.mnt;
+		nd->path.dentry = lower_dentry;
+		nd->path.mnt = lower_mnt;
+	}
 	rc = lower_dentry->d_op->d_revalidate(lower_dentry, nd);
-	nd->path.dentry = dentry_save;
-	nd->path.mnt = vfsmount_save;
+	if (nd) {
+		nd->path.dentry = dentry_save;
+		nd->path.mnt = vfsmount_save;
+	}
 	if (dentry->d_inode) {
 		struct inode *lower_inode =
 			ecryptfs_inode_to_lower(dentry->d_inode);
diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h
index dbc84ed9633..e00753496e3 100644
--- a/fs/ecryptfs/ecryptfs_kernel.h
+++ b/fs/ecryptfs/ecryptfs_kernel.h
@@ -632,8 +632,7 @@ int ecryptfs_interpose(struct dentry *hidden_dentry,
 		       u32 flags);
 int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 					struct dentry *lower_dentry,
-					struct inode *ecryptfs_dir_inode,
-					struct nameidata *ecryptfs_nd);
+					struct inode *ecryptfs_dir_inode);
 int ecryptfs_decode_and_decrypt_filename(char **decrypted_name,
 					 size_t *decrypted_name_size,
 					 struct dentry *ecryptfs_dentry,
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index 81e10e6a944..7d1050e254f 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -317,6 +317,7 @@ ecryptfs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 
 const struct file_operations ecryptfs_dir_fops = {
 	.readdir = ecryptfs_readdir,
+	.read = generic_read_dir,
 	.unlocked_ioctl = ecryptfs_unlocked_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl = ecryptfs_compat_ioctl,
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index bd33f87a190..b592938a84b 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -74,16 +74,20 @@ ecryptfs_create_underlying_file(struct inode *lower_dir_inode,
 	unsigned int flags_save;
 	int rc;
 
-	dentry_save = nd->path.dentry;
-	vfsmount_save = nd->path.mnt;
-	flags_save = nd->flags;
-	nd->path.dentry = lower_dentry;
-	nd->path.mnt = lower_mnt;
-	nd->flags &= ~LOOKUP_OPEN;
+	if (nd) {
+		dentry_save = nd->path.dentry;
+		vfsmount_save = nd->path.mnt;
+		flags_save = nd->flags;
+		nd->path.dentry = lower_dentry;
+		nd->path.mnt = lower_mnt;
+		nd->flags &= ~LOOKUP_OPEN;
+	}
 	rc = vfs_create(lower_dir_inode, lower_dentry, mode, nd);
-	nd->path.dentry = dentry_save;
-	nd->path.mnt = vfsmount_save;
-	nd->flags = flags_save;
+	if (nd) {
+		nd->path.dentry = dentry_save;
+		nd->path.mnt = vfsmount_save;
+		nd->flags = flags_save;
+	}
 	return rc;
 }
 
@@ -241,8 +245,7 @@ out:
  */
 int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 					struct dentry *lower_dentry,
-					struct inode *ecryptfs_dir_inode,
-					struct nameidata *ecryptfs_nd)
+					struct inode *ecryptfs_dir_inode)
 {
 	struct dentry *lower_dir_dentry;
 	struct vfsmount *lower_mnt;
@@ -290,8 +293,6 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 		goto out;
 	if (special_file(lower_inode->i_mode))
 		goto out;
-	if (!ecryptfs_nd)
-		goto out;
 	/* Released in this function */
 	page_virt = kmem_cache_zalloc(ecryptfs_header_cache_2, GFP_USER);
 	if (!page_virt) {
@@ -349,75 +350,6 @@ out:
 }
 
 /**
- * ecryptfs_new_lower_dentry
- * @name: The name of the new dentry.
- * @lower_dir_dentry: Parent directory of the new dentry.
- * @nd: nameidata from last lookup.
- *
- * Create a new dentry or get it from lower parent dir.
- */
-static struct dentry *
-ecryptfs_new_lower_dentry(struct qstr *name, struct dentry *lower_dir_dentry,
-			  struct nameidata *nd)
-{
-	struct dentry *new_dentry;
-	struct dentry *tmp;
-	struct inode *lower_dir_inode;
-
-	lower_dir_inode = lower_dir_dentry->d_inode;
-
-	tmp = d_alloc(lower_dir_dentry, name);
-	if (!tmp)
-		return ERR_PTR(-ENOMEM);
-
-	mutex_lock(&lower_dir_inode->i_mutex);
-	new_dentry = lower_dir_inode->i_op->lookup(lower_dir_inode, tmp, nd);
-	mutex_unlock(&lower_dir_inode->i_mutex);
-
-	if (!new_dentry)
-		new_dentry = tmp;
-	else
-		dput(tmp);
-
-	return new_dentry;
-}
-
-
-/**
- * ecryptfs_lookup_one_lower
- * @ecryptfs_dentry: The eCryptfs dentry that we are looking up
- * @lower_dir_dentry: lower parent directory
- * @name: lower file name
- *
- * Get the lower dentry from vfs. If lower dentry does not exist yet,
- * create it.
- */
-static struct dentry *
-ecryptfs_lookup_one_lower(struct dentry *ecryptfs_dentry,
-			  struct dentry *lower_dir_dentry, struct qstr *name)
-{
-	struct nameidata nd;
-	struct vfsmount *lower_mnt;
-	int err;
-
-	lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(
-				    ecryptfs_dentry->d_parent));
-	err = vfs_path_lookup(lower_dir_dentry, lower_mnt, name->name , 0, &nd);
-	mntput(lower_mnt);
-
-	if (!err) {
-		/* we dont need the mount */
-		mntput(nd.path.mnt);
-		return nd.path.dentry;
-	}
-	if (err != -ENOENT)
-		return ERR_PTR(err);
-
-	/* create a new lower dentry */
-	return ecryptfs_new_lower_dentry(name, lower_dir_dentry, &nd);
-}
-
-/**
  * ecryptfs_lookup
  * @ecryptfs_dir_inode: The eCryptfs directory inode
  * @ecryptfs_dentry: The eCryptfs dentry that we are looking up
@@ -434,7 +366,6 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode,
 	size_t encrypted_and_encoded_name_size;
 	struct ecryptfs_mount_crypt_stat *mount_crypt_stat = NULL;
 	struct dentry *lower_dir_dentry, *lower_dentry;
-	struct qstr lower_name;
 	int rc = 0;
 
 	if ((ecryptfs_dentry->d_name.len == 1
@@ -444,20 +375,14 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode,
 		goto out_d_drop;
 	}
 	lower_dir_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry->d_parent);
-	lower_name.name = ecryptfs_dentry->d_name.name;
-	lower_name.len = ecryptfs_dentry->d_name.len;
-	lower_name.hash = ecryptfs_dentry->d_name.hash;
-	if (lower_dir_dentry->d_op && lower_dir_dentry->d_op->d_hash) {
-		rc = lower_dir_dentry->d_op->d_hash(lower_dir_dentry,
-				lower_dir_dentry->d_inode, &lower_name);
-		if (rc < 0)
-			goto out_d_drop;
-	}
-	lower_dentry = ecryptfs_lookup_one_lower(ecryptfs_dentry,
-						 lower_dir_dentry, &lower_name);
+	mutex_lock(&lower_dir_dentry->d_inode->i_mutex);
+	lower_dentry = lookup_one_len(ecryptfs_dentry->d_name.name,
+				      lower_dir_dentry,
+				      ecryptfs_dentry->d_name.len);
+	mutex_unlock(&lower_dir_dentry->d_inode->i_mutex);
 	if (IS_ERR(lower_dentry)) {
 		rc = PTR_ERR(lower_dentry);
-		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_lower() returned "
+		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_len() returned "
 				"[%d] on lower_dentry = [%s]\n", __func__, rc,
 				encrypted_and_encoded_name);
 		goto out_d_drop;
@@ -479,28 +404,21 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode,
 		       "filename; rc = [%d]\n", __func__, rc);
 		goto out_d_drop;
 	}
-	lower_name.name = encrypted_and_encoded_name;
-	lower_name.len = encrypted_and_encoded_name_size;
-	lower_name.hash = full_name_hash(lower_name.name, lower_name.len);
-	if (lower_dir_dentry->d_op && lower_dir_dentry->d_op->d_hash) {
-		rc = lower_dir_dentry->d_op->d_hash(lower_dir_dentry,
-				lower_dir_dentry->d_inode, &lower_name);
-		if (rc < 0)
-			goto out_d_drop;
-	}
-	lower_dentry = ecryptfs_lookup_one_lower(ecryptfs_dentry,
-						 lower_dir_dentry, &lower_name);
+	mutex_lock(&lower_dir_dentry->d_inode->i_mutex);
+	lower_dentry = lookup_one_len(encrypted_and_encoded_name,
+				      lower_dir_dentry,
+				      encrypted_and_encoded_name_size);
+	mutex_unlock(&lower_dir_dentry->d_inode->i_mutex);
 	if (IS_ERR(lower_dentry)) {
 		rc = PTR_ERR(lower_dentry);
-		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_lower() returned "
+		ecryptfs_printk(KERN_DEBUG, "%s: lookup_one_len() returned "
 				"[%d] on lower_dentry = [%s]\n", __func__, rc,
 				encrypted_and_encoded_name);
 		goto out_d_drop;
 	}
 lookup_and_interpose:
 	rc = ecryptfs_lookup_and_interpose_lower(ecryptfs_dentry, lower_dentry,
-						 ecryptfs_dir_inode,
-						 ecryptfs_nd);
+						 ecryptfs_dir_inode);
 	goto out;
 out_d_drop:
 	d_drop(ecryptfs_dentry);
@@ -1092,6 +1010,8 @@ int ecryptfs_getattr(struct vfsmount *mnt, struct dentry *dentry,
 	rc = vfs_getattr(ecryptfs_dentry_to_lower_mnt(dentry),
 			 ecryptfs_dentry_to_lower(dentry), &lower_stat);
 	if (!rc) {
+		fsstack_copy_attr_all(dentry->d_inode,
+				      ecryptfs_inode_to_lower(dentry->d_inode));
 		generic_fillattr(dentry->d_inode, stat);
 		stat->blocks = lower_stat.blocks;
 	}
diff --git a/fs/eventfd.c b/fs/eventfd.c
index e0194b3e14d..d9a59177391 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -99,7 +99,7 @@ EXPORT_SYMBOL_GPL(eventfd_ctx_get);
  * @ctx: [in] Pointer to eventfd context.
  *
  * The eventfd context reference must have been previously acquired either
- * with eventfd_ctx_get() or eventfd_ctx_fdget()).
+ * with eventfd_ctx_get() or eventfd_ctx_fdget().
  */
 void eventfd_ctx_put(struct eventfd_ctx *ctx)
 {
@@ -146,9 +146,9 @@ static void eventfd_ctx_do_read(struct eventfd_ctx *ctx, __u64 *cnt)
  * eventfd_ctx_remove_wait_queue - Read the current counter and removes wait queue.
  * @ctx: [in] Pointer to eventfd context.
  * @wait: [in] Wait queue to be removed.
- * @cnt: [out] Pointer to the 64bit conter value.
+ * @cnt: [out] Pointer to the 64-bit counter value.
  *
- * Returns zero if successful, or the following error codes:
+ * Returns %0 if successful, or the following error codes:
  *
  * -EAGAIN      : The operation would have blocked.
  *
@@ -175,11 +175,11 @@ EXPORT_SYMBOL_GPL(eventfd_ctx_remove_wait_queue);
  * eventfd_ctx_read - Reads the eventfd counter or wait if it is zero.
  * @ctx: [in] Pointer to eventfd context.
  * @no_wait: [in] Different from zero if the operation should not block.
- * @cnt: [out] Pointer to the 64bit conter value.
+ * @cnt: [out] Pointer to the 64-bit counter value.
  *
- * Returns zero if successful, or the following error codes:
+ * Returns %0 if successful, or the following error codes:
  *
- * -EAGAIN      : The operation would have blocked but @no_wait was nonzero.
+ * -EAGAIN      : The operation would have blocked but @no_wait was non-zero.
  * -ERESTARTSYS : A signal interrupted the wait operation.
  *
  * If @no_wait is zero, the function might sleep until the eventfd internal
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 267d0ada454..4a09af9e9a6 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -63,6 +63,13 @@
  * cleanup path and it is also acquired by eventpoll_release_file()
  * if a file has been pushed inside an epoll set and it is then
  * close()d without a previous call toepoll_ctl(EPOLL_CTL_DEL).
+ * It is also acquired when inserting an epoll fd onto another epoll
+ * fd. We do this so that we walk the epoll tree and ensure that this
+ * insertion does not create a cycle of epoll file descriptors, which
+ * could lead to deadlock. We need a global mutex to prevent two
+ * simultaneous inserts (A into B and B into A) from racing and
+ * constructing a cycle without either insert observing that it is
+ * going to.
  * It is possible to drop the "ep->mtx" and to use the global
  * mutex "epmutex" (together with "ep->lock") to have it working,
  * but having "ep->mtx" will make the interface more scalable.
@@ -224,6 +231,9 @@ static long max_user_watches __read_mostly;
  */
 static DEFINE_MUTEX(epmutex);
 
+/* Used to check for epoll file descriptor inclusion loops */
+static struct nested_calls poll_loop_ncalls;
+
 /* Used for safe wake up implementation */
 static struct nested_calls poll_safewake_ncalls;
 
@@ -1198,6 +1208,62 @@ retry:
 	return res;
 }
 
+/**
+ * ep_loop_check_proc - Callback function to be passed to the @ep_call_nested()
+ *                      API, to verify that adding an epoll file inside another
+ *                      epoll structure, does not violate the constraints, in
+ *                      terms of closed loops, or too deep chains (which can
+ *                      result in excessive stack usage).
+ *
+ * @priv: Pointer to the epoll file to be currently checked.
+ * @cookie: Original cookie for this call. This is the top-of-the-chain epoll
+ *          data structure pointer.
+ * @call_nests: Current dept of the @ep_call_nested() call stack.
+ *
+ * Returns: Returns zero if adding the epoll @file inside current epoll
+ *          structure @ep does not violate the constraints, or -1 otherwise.
+ */
+static int ep_loop_check_proc(void *priv, void *cookie, int call_nests)
+{
+	int error = 0;
+	struct file *file = priv;
+	struct eventpoll *ep = file->private_data;
+	struct rb_node *rbp;
+	struct epitem *epi;
+
+	mutex_lock(&ep->mtx);
+	for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) {
+		epi = rb_entry(rbp, struct epitem, rbn);
+		if (unlikely(is_file_epoll(epi->ffd.file))) {
+			error = ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS,
+					       ep_loop_check_proc, epi->ffd.file,
+					       epi->ffd.file->private_data, current);
+			if (error != 0)
+				break;
+		}
+	}
+	mutex_unlock(&ep->mtx);
+
+	return error;
+}
+
+/**
+ * ep_loop_check - Performs a check to verify that adding an epoll file (@file)
+ *                 another epoll file (represented by @ep) does not create
+ *                 closed loops or too deep chains.
+ *
+ * @ep: Pointer to the epoll private data structure.
+ * @file: Pointer to the epoll file to be checked.
+ *
+ * Returns: Returns zero if adding the epoll @file inside current epoll
+ *          structure @ep does not violate the constraints, or -1 otherwise.
+ */
+static int ep_loop_check(struct eventpoll *ep, struct file *file)
+{
+	return ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS,
+			      ep_loop_check_proc, file, ep, current);
+}
+
 /*
  * Open an eventpoll file descriptor.
  */
@@ -1246,6 +1312,7 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 		struct epoll_event __user *, event)
 {
 	int error;
+	int did_lock_epmutex = 0;
 	struct file *file, *tfile;
 	struct eventpoll *ep;
 	struct epitem *epi;
@@ -1287,6 +1354,25 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 	 */
 	ep = file->private_data;
 
+	/*
+	 * When we insert an epoll file descriptor, inside another epoll file
+	 * descriptor, there is the change of creating closed loops, which are
+	 * better be handled here, than in more critical paths.
+	 *
+	 * We hold epmutex across the loop check and the insert in this case, in
+	 * order to prevent two separate inserts from racing and each doing the
+	 * insert "at the same time" such that ep_loop_check passes on both
+	 * before either one does the insert, thereby creating a cycle.
+	 */
+	if (unlikely(is_file_epoll(tfile) && op == EPOLL_CTL_ADD)) {
+		mutex_lock(&epmutex);
+		did_lock_epmutex = 1;
+		error = -ELOOP;
+		if (ep_loop_check(ep, tfile) != 0)
+			goto error_tgt_fput;
+	}
+
+
 	mutex_lock(&ep->mtx);
 
 	/*
@@ -1322,6 +1408,9 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
 	mutex_unlock(&ep->mtx);
 
 error_tgt_fput:
+	if (unlikely(did_lock_epmutex))
+		mutex_unlock(&epmutex);
+
 	fput(tfile);
 error_fput:
 	fput(file);
@@ -1441,6 +1530,12 @@ static int __init eventpoll_init(void)
 		EP_ITEM_COST;
 	BUG_ON(max_user_watches < 0);
 
+	/*
+	 * Initialize the structure used to perform epoll file descriptor
+	 * inclusion loops checks.
+	 */
+	ep_nested_calls_init(&poll_loop_ncalls);
+
 	/* Initialize the structure used to perform safe poll wait head wake ups */
 	ep_nested_calls_init(&poll_safewake_ncalls);
 
diff --git a/fs/exec.c b/fs/exec.c
index 52a447d9b6a..ba99e1abb1a 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -115,13 +115,16 @@ SYSCALL_DEFINE1(uselib, const char __user *, library)
 	struct file *file;
 	char *tmp = getname(library);
 	int error = PTR_ERR(tmp);
+	static const struct open_flags uselib_flags = {
+		.open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC,
+		.acc_mode = MAY_READ | MAY_EXEC | MAY_OPEN,
+		.intent = LOOKUP_OPEN
+	};
 
 	if (IS_ERR(tmp))
 		goto out;
 
-	file = do_filp_open(AT_FDCWD, tmp,
-				O_LARGEFILE | O_RDONLY | __FMODE_EXEC, 0,
-				MAY_READ | MAY_EXEC | MAY_OPEN);
+	file = do_filp_open(AT_FDCWD, tmp, &uselib_flags, LOOKUP_FOLLOW);
 	putname(tmp);
 	error = PTR_ERR(file);
 	if (IS_ERR(file))
@@ -721,10 +724,13 @@ struct file *open_exec(const char *name)
 {
 	struct file *file;
 	int err;
+	static const struct open_flags open_exec_flags = {
+		.open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC,
+		.acc_mode = MAY_EXEC | MAY_OPEN,
+		.intent = LOOKUP_OPEN
+	};
 
-	file = do_filp_open(AT_FDCWD, name,
-				O_LARGEFILE | O_RDONLY | __FMODE_EXEC, 0,
-				MAY_EXEC | MAY_OPEN);
+	file = do_filp_open(AT_FDCWD, name, &open_exec_flags, LOOKUP_FOLLOW);
 	if (IS_ERR(file))
 		goto out;
 
diff --git a/fs/exofs/namei.c b/fs/exofs/namei.c
index 264e95d0283..4d70db110cf 100644
--- a/fs/exofs/namei.c
+++ b/fs/exofs/namei.c
@@ -272,7 +272,6 @@ static int exofs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		new_de = exofs_find_entry(new_dir, new_dentry, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inode_inc_link_count(old_inode);
 		err = exofs_set_link(new_dir, new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME;
 		if (dir_de)
@@ -286,12 +285,9 @@ static int exofs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			if (new_dir->i_nlink >= EXOFS_LINK_MAX)
 				goto out_dir;
 		}
-		inode_inc_link_count(old_inode);
 		err = exofs_add_link(new_dentry, old_inode);
-		if (err) {
-			inode_dec_link_count(old_inode);
+		if (err)
 			goto out_dir;
-		}
 		if (dir_de)
 			inode_inc_link_count(new_dir);
 	}
@@ -299,7 +295,7 @@ static int exofs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	old_inode->i_ctime = CURRENT_TIME;
 
 	exofs_delete_entry(old_de, old_page);
-	inode_dec_link_count(old_inode);
+	mark_inode_dirty(old_inode);
 
 	if (dir_de) {
 		err = exofs_set_link(old_inode, dir_de, dir_page, new_dir);
diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c
index 4b6825740dd..b05acb79613 100644
--- a/fs/exportfs/expfs.c
+++ b/fs/exportfs/expfs.c
@@ -320,9 +320,14 @@ static int export_encode_fh(struct dentry *dentry, struct fid *fid,
 	struct inode * inode = dentry->d_inode;
 	int len = *max_len;
 	int type = FILEID_INO32_GEN;
-	
-	if (len < 2 || (connectable && len < 4))
+
+	if (connectable && (len < 4)) {
+		*max_len = 4;
+		return 255;
+	} else if (len < 2) {
+		*max_len = 2;
 		return 255;
+	}
 
 	len = 2;
 	fid->i32.ino = inode->i_ino;
@@ -369,6 +374,8 @@ struct dentry *exportfs_decode_fh(struct vfsmount *mnt, struct fid *fid,
 	/*
 	 * Try to get any dentry for the given file handle from the filesystem.
 	 */
+	if (!nop || !nop->fh_to_dentry)
+		return ERR_PTR(-ESTALE);
 	result = nop->fh_to_dentry(mnt->mnt_sb, fid, fh_len, fileid_type);
 	if (!result)
 		result = ERR_PTR(-ESTALE);
diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c
index 2e1d8341d82..adb91855ccd 100644
--- a/fs/ext2/namei.c
+++ b/fs/ext2/namei.c
@@ -344,7 +344,6 @@ static int ext2_rename (struct inode * old_dir, struct dentry * old_dentry,
 		new_de = ext2_find_entry (new_dir, &new_dentry->d_name, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inode_inc_link_count(old_inode);
 		ext2_set_link(new_dir, new_de, new_page, old_inode, 1);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
@@ -356,12 +355,9 @@ static int ext2_rename (struct inode * old_dir, struct dentry * old_dentry,
 			if (new_dir->i_nlink >= EXT2_LINK_MAX)
 				goto out_dir;
 		}
-		inode_inc_link_count(old_inode);
 		err = ext2_add_link(new_dentry, old_inode);
-		if (err) {
-			inode_dec_link_count(old_inode);
+		if (err)
 			goto out_dir;
-		}
 		if (dir_de)
 			inode_inc_link_count(new_dir);
 	}
@@ -369,12 +365,11 @@ static int ext2_rename (struct inode * old_dir, struct dentry * old_dentry,
 	/*
 	 * Like most other Unix systems, set the ctime for inodes on a
  	 * rename.
-	 * inode_dec_link_count() will mark the inode dirty.
 	 */
 	old_inode->i_ctime = CURRENT_TIME_SEC;
+	mark_inode_dirty(old_inode);
 
 	ext2_delete_entry (old_de, old_page);
-	inode_dec_link_count(old_inode);
 
 	if (dir_de) {
 		if (old_dir != new_dir)
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index b27ba71810e..561f6925626 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -2253,13 +2253,6 @@ static int ext3_link (struct dentry * old_dentry,
 
 	dquot_initialize(dir);
 
-	/*
-	 * Return -ENOENT if we've raced with unlink and i_nlink is 0.  Doing
-	 * otherwise has the potential to corrupt the orphan inode list.
-	 */
-	if (inode->i_nlink == 0)
-		return -ENOENT;
-
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT3_INDEX_EXTRA_TRANS_BLOCKS);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 85c8cc8f247..9cc19a1dea8 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -1936,6 +1936,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	sb->s_qcop = &ext3_qctl_operations;
 	sb->dq_op = &ext3_quota_operations;
 #endif
+	memcpy(sb->s_uuid, es->s_uuid, sizeof(es->s_uuid));
 	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
 	mutex_init(&sbi->s_orphan_lock);
 	mutex_init(&sbi->s_resize_lock);
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 5485390d32c..e781b7ea563 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -2304,13 +2304,6 @@ static int ext4_link(struct dentry *old_dentry,
 
 	dquot_initialize(dir);
 
-	/*
-	 * Return -ENOENT if we've raced with unlink and i_nlink is 0.  Doing
-	 * otherwise has the potential to corrupt the orphan inode list.
-	 */
-	if (inode->i_nlink == 0)
-		return -ENOENT;
-
 retry:
 	handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT4_INDEX_EXTRA_TRANS_BLOCKS);
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index f6a318f836b..5977b356a43 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -3415,6 +3415,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_qcop = &ext4_qctl_operations;
 	sb->dq_op = &ext4_quota_operations;
 #endif
+	memcpy(sb->s_uuid, es->s_uuid, sizeof(es->s_uuid));
+
 	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
 	mutex_init(&sbi->s_orphan_lock);
 	mutex_init(&sbi->s_resize_lock);
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 86753fe10bd..0e277ec4b61 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -757,8 +757,10 @@ fat_encode_fh(struct dentry *de, __u32 *fh, int *lenp, int connectable)
 	struct inode *inode =  de->d_inode;
 	u32 ipos_h, ipos_m, ipos_l;
 
-	if (len < 5)
+	if (len < 5) {
+		*lenp = 5;
 		return 255; /* no room */
+	}
 
 	ipos_h = MSDOS_I(inode)->i_pos >> 8;
 	ipos_m = (MSDOS_I(inode)->i_pos & 0xf0) << 24;
diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c
index f88f752babd..adae3fb7451 100644
--- a/fs/fat/namei_vfat.c
+++ b/fs/fat/namei_vfat.c
@@ -43,7 +43,7 @@ static int vfat_revalidate_shortname(struct dentry *dentry)
 
 static int vfat_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
-	if (nd->flags & LOOKUP_RCU)
+	if (nd && nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 
 	/* This is not negative dentry. Always valid. */
@@ -54,7 +54,7 @@ static int vfat_revalidate(struct dentry *dentry, struct nameidata *nd)
 
 static int vfat_revalidate_ci(struct dentry *dentry, struct nameidata *nd)
 {
-	if (nd->flags & LOOKUP_RCU)
+	if (nd && nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 
 	/*
diff --git a/fs/fcntl.c b/fs/fcntl.c
index cb1026181bd..6c82e5bac03 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -131,7 +131,7 @@ SYSCALL_DEFINE2(dup2, unsigned int, oldfd, unsigned int, newfd)
 SYSCALL_DEFINE1(dup, unsigned int, fildes)
 {
 	int ret = -EBADF;
-	struct file *file = fget(fildes);
+	struct file *file = fget_raw(fildes);
 
 	if (file) {
 		ret = get_unused_fd();
@@ -426,15 +426,35 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 	return err;
 }
 
+static int check_fcntl_cmd(unsigned cmd)
+{
+	switch (cmd) {
+	case F_DUPFD:
+	case F_DUPFD_CLOEXEC:
+	case F_GETFD:
+	case F_SETFD:
+	case F_GETFL:
+		return 1;
+	}
+	return 0;
+}
+
 SYSCALL_DEFINE3(fcntl, unsigned int, fd, unsigned int, cmd, unsigned long, arg)
 {	
 	struct file *filp;
 	long err = -EBADF;
 
-	filp = fget(fd);
+	filp = fget_raw(fd);
 	if (!filp)
 		goto out;
 
+	if (unlikely(filp->f_mode & FMODE_PATH)) {
+		if (!check_fcntl_cmd(cmd)) {
+			fput(filp);
+			goto out;
+		}
+	}
+
 	err = security_file_fcntl(filp, cmd, arg);
 	if (err) {
 		fput(filp);
@@ -456,10 +476,17 @@ SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd,
 	long err;
 
 	err = -EBADF;
-	filp = fget(fd);
+	filp = fget_raw(fd);
 	if (!filp)
 		goto out;
 
+	if (unlikely(filp->f_mode & FMODE_PATH)) {
+		if (!check_fcntl_cmd(cmd)) {
+			fput(filp);
+			goto out;
+		}
+	}
+
 	err = security_file_fcntl(filp, cmd, arg);
 	if (err) {
 		fput(filp);
@@ -808,14 +835,14 @@ static int __init fcntl_init(void)
 	 * Exceptions: O_NONBLOCK is a two bit define on parisc; O_NDELAY
 	 * is defined as O_NONBLOCK on some platforms and not on others.
 	 */
-	BUILD_BUG_ON(18 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32(
+	BUILD_BUG_ON(19 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32(
 		O_RDONLY	| O_WRONLY	| O_RDWR	|
 		O_CREAT		| O_EXCL	| O_NOCTTY	|
 		O_TRUNC		| O_APPEND	| /* O_NONBLOCK	| */
 		__O_SYNC	| O_DSYNC	| FASYNC	|
 		O_DIRECT	| O_LARGEFILE	| O_DIRECTORY	|
 		O_NOFOLLOW	| O_NOATIME	| O_CLOEXEC	|
-		__FMODE_EXEC
+		__FMODE_EXEC	| O_PATH
 		));
 
 	fasync_cache = kmem_cache_create("fasync_cache",
diff --git a/fs/fhandle.c b/fs/fhandle.c
new file mode 100644
index 00000000000..bf93ad2bee0
--- /dev/null
+++ b/fs/fhandle.c
@@ -0,0 +1,265 @@
+#include <linux/syscalls.h>
+#include <linux/slab.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/exportfs.h>
+#include <linux/fs_struct.h>
+#include <linux/fsnotify.h>
+#include <asm/uaccess.h>
+#include "internal.h"
+
+static long do_sys_name_to_handle(struct path *path,
+				  struct file_handle __user *ufh,
+				  int __user *mnt_id)
+{
+	long retval;
+	struct file_handle f_handle;
+	int handle_dwords, handle_bytes;
+	struct file_handle *handle = NULL;
+
+	/*
+	 * We need t make sure wether the file system
+	 * support decoding of the file handle
+	 */
+	if (!path->mnt->mnt_sb->s_export_op ||
+	    !path->mnt->mnt_sb->s_export_op->fh_to_dentry)
+		return -EOPNOTSUPP;
+
+	if (copy_from_user(&f_handle, ufh, sizeof(struct file_handle)))
+		return -EFAULT;
+
+	if (f_handle.handle_bytes > MAX_HANDLE_SZ)
+		return -EINVAL;
+
+	handle = kmalloc(sizeof(struct file_handle) + f_handle.handle_bytes,
+			 GFP_KERNEL);
+	if (!handle)
+		return -ENOMEM;
+
+	/* convert handle size to  multiple of sizeof(u32) */
+	handle_dwords = f_handle.handle_bytes >> 2;
+
+	/* we ask for a non connected handle */
+	retval = exportfs_encode_fh(path->dentry,
+				    (struct fid *)handle->f_handle,
+				    &handle_dwords,  0);
+	handle->handle_type = retval;
+	/* convert handle size to bytes */
+	handle_bytes = handle_dwords * sizeof(u32);
+	handle->handle_bytes = handle_bytes;
+	if ((handle->handle_bytes > f_handle.handle_bytes) ||
+	    (retval == 255) || (retval == -ENOSPC)) {
+		/* As per old exportfs_encode_fh documentation
+		 * we could return ENOSPC to indicate overflow
+		 * But file system returned 255 always. So handle
+		 * both the values
+		 */
+		/*
+		 * set the handle size to zero so we copy only
+		 * non variable part of the file_handle
+		 */
+		handle_bytes = 0;
+		retval = -EOVERFLOW;
+	} else
+		retval = 0;
+	/* copy the mount id */
+	if (copy_to_user(mnt_id, &path->mnt->mnt_id, sizeof(*mnt_id)) ||
+	    copy_to_user(ufh, handle,
+			 sizeof(struct file_handle) + handle_bytes))
+		retval = -EFAULT;
+	kfree(handle);
+	return retval;
+}
+
+/**
+ * sys_name_to_handle_at: convert name to handle
+ * @dfd: directory relative to which name is interpreted if not absolute
+ * @name: name that should be converted to handle.
+ * @handle: resulting file handle
+ * @mnt_id: mount id of the file system containing the file
+ * @flag: flag value to indicate whether to follow symlink or not
+ *
+ * @handle->handle_size indicate the space available to store the
+ * variable part of the file handle in bytes. If there is not
+ * enough space, the field is updated to return the minimum
+ * value required.
+ */
+SYSCALL_DEFINE5(name_to_handle_at, int, dfd, const char __user *, name,
+		struct file_handle __user *, handle, int __user *, mnt_id,
+		int, flag)
+{
+	struct path path;
+	int lookup_flags;
+	int err;
+
+	if ((flag & ~(AT_SYMLINK_FOLLOW | AT_EMPTY_PATH)) != 0)
+		return -EINVAL;
+
+	lookup_flags = (flag & AT_SYMLINK_FOLLOW) ? LOOKUP_FOLLOW : 0;
+	if (flag & AT_EMPTY_PATH)
+		lookup_flags |= LOOKUP_EMPTY;
+	err = user_path_at(dfd, name, lookup_flags, &path);
+	if (!err) {
+		err = do_sys_name_to_handle(&path, handle, mnt_id);
+		path_put(&path);
+	}
+	return err;
+}
+
+static struct vfsmount *get_vfsmount_from_fd(int fd)
+{
+	struct path path;
+
+	if (fd == AT_FDCWD) {
+		struct fs_struct *fs = current->fs;
+		spin_lock(&fs->lock);
+		path = fs->pwd;
+		mntget(path.mnt);
+		spin_unlock(&fs->lock);
+	} else {
+		int fput_needed;
+		struct file *file = fget_light(fd, &fput_needed);
+		if (!file)
+			return ERR_PTR(-EBADF);
+		path = file->f_path;
+		mntget(path.mnt);
+		fput_light(file, fput_needed);
+	}
+	return path.mnt;
+}
+
+static int vfs_dentry_acceptable(void *context, struct dentry *dentry)
+{
+	return 1;
+}
+
+static int do_handle_to_path(int mountdirfd, struct file_handle *handle,
+			     struct path *path)
+{
+	int retval = 0;
+	int handle_dwords;
+
+	path->mnt = get_vfsmount_from_fd(mountdirfd);
+	if (IS_ERR(path->mnt)) {
+		retval = PTR_ERR(path->mnt);
+		goto out_err;
+	}
+	/* change the handle size to multiple of sizeof(u32) */
+	handle_dwords = handle->handle_bytes >> 2;
+	path->dentry = exportfs_decode_fh(path->mnt,
+					  (struct fid *)handle->f_handle,
+					  handle_dwords, handle->handle_type,
+					  vfs_dentry_acceptable, NULL);
+	if (IS_ERR(path->dentry)) {
+		retval = PTR_ERR(path->dentry);
+		goto out_mnt;
+	}
+	return 0;
+out_mnt:
+	mntput(path->mnt);
+out_err:
+	return retval;
+}
+
+static int handle_to_path(int mountdirfd, struct file_handle __user *ufh,
+		   struct path *path)
+{
+	int retval = 0;
+	struct file_handle f_handle;
+	struct file_handle *handle = NULL;
+
+	/*
+	 * With handle we don't look at the execute bit on the
+	 * the directory. Ideally we would like CAP_DAC_SEARCH.
+	 * But we don't have that
+	 */
+	if (!capable(CAP_DAC_READ_SEARCH)) {
+		retval = -EPERM;
+		goto out_err;
+	}
+	if (copy_from_user(&f_handle, ufh, sizeof(struct file_handle))) {
+		retval = -EFAULT;
+		goto out_err;
+	}
+	if ((f_handle.handle_bytes > MAX_HANDLE_SZ) ||
+	    (f_handle.handle_bytes == 0)) {
+		retval = -EINVAL;
+		goto out_err;
+	}
+	handle = kmalloc(sizeof(struct file_handle) + f_handle.handle_bytes,
+			 GFP_KERNEL);
+	if (!handle) {
+		retval = -ENOMEM;
+		goto out_err;
+	}
+	/* copy the full handle */
+	if (copy_from_user(handle, ufh,
+			   sizeof(struct file_handle) +
+			   f_handle.handle_bytes)) {
+		retval = -EFAULT;
+		goto out_handle;
+	}
+
+	retval = do_handle_to_path(mountdirfd, handle, path);
+
+out_handle:
+	kfree(handle);
+out_err:
+	return retval;
+}
+
+long do_handle_open(int mountdirfd,
+		    struct file_handle __user *ufh, int open_flag)
+{
+	long retval = 0;
+	struct path path;
+	struct file *file;
+	int fd;
+
+	retval = handle_to_path(mountdirfd, ufh, &path);
+	if (retval)
+		return retval;
+
+	fd = get_unused_fd_flags(open_flag);
+	if (fd < 0) {
+		path_put(&path);
+		return fd;
+	}
+	file = file_open_root(path.dentry, path.mnt, "", open_flag);
+	if (IS_ERR(file)) {
+		put_unused_fd(fd);
+		retval =  PTR_ERR(file);
+	} else {
+		retval = fd;
+		fsnotify_open(file);
+		fd_install(fd, file);
+	}
+	path_put(&path);
+	return retval;
+}
+
+/**
+ * sys_open_by_handle_at: Open the file handle
+ * @mountdirfd: directory file descriptor
+ * @handle: file handle to be opened
+ * @flag: open flags.
+ *
+ * @mountdirfd indicate the directory file descriptor
+ * of the mount point. file handle is decoded relative
+ * to the vfsmount pointed by the @mountdirfd. @flags
+ * value is same as the open(2) flags.
+ */
+SYSCALL_DEFINE3(open_by_handle_at, int, mountdirfd,
+		struct file_handle __user *, handle,
+		int, flags)
+{
+	long ret;
+
+	if (force_o_largefile())
+		flags |= O_LARGEFILE;
+
+	ret = do_handle_open(mountdirfd, handle, flags);
+	return ret;
+}
diff --git a/fs/file_table.c b/fs/file_table.c
index eb36b6b17e2..74a9544ac77 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -276,11 +276,10 @@ struct file *fget(unsigned int fd)
 	rcu_read_lock();
 	file = fcheck_files(files, fd);
 	if (file) {
-		if (!atomic_long_inc_not_zero(&file->f_count)) {
-			/* File object ref couldn't be taken */
-			rcu_read_unlock();
-			return NULL;
-		}
+		/* File object ref couldn't be taken */
+		if (file->f_mode & FMODE_PATH ||
+		    !atomic_long_inc_not_zero(&file->f_count))
+			file = NULL;
 	}
 	rcu_read_unlock();
 
@@ -289,6 +288,25 @@ struct file *fget(unsigned int fd)
 
 EXPORT_SYMBOL(fget);
 
+struct file *fget_raw(unsigned int fd)
+{
+	struct file *file;
+	struct files_struct *files = current->files;
+
+	rcu_read_lock();
+	file = fcheck_files(files, fd);
+	if (file) {
+		/* File object ref couldn't be taken */
+		if (!atomic_long_inc_not_zero(&file->f_count))
+			file = NULL;
+	}
+	rcu_read_unlock();
+
+	return file;
+}
+
+EXPORT_SYMBOL(fget_raw);
+
 /*
  * Lightweight file lookup - no refcnt increment if fd table isn't shared.
  *
@@ -313,6 +331,33 @@ struct file *fget_light(unsigned int fd, int *fput_needed)
 	*fput_needed = 0;
 	if (atomic_read(&files->count) == 1) {
 		file = fcheck_files(files, fd);
+		if (file && (file->f_mode & FMODE_PATH))
+			file = NULL;
+	} else {
+		rcu_read_lock();
+		file = fcheck_files(files, fd);
+		if (file) {
+			if (!(file->f_mode & FMODE_PATH) &&
+			    atomic_long_inc_not_zero(&file->f_count))
+				*fput_needed = 1;
+			else
+				/* Didn't get the reference, someone's freed */
+				file = NULL;
+		}
+		rcu_read_unlock();
+	}
+
+	return file;
+}
+
+struct file *fget_raw_light(unsigned int fd, int *fput_needed)
+{
+	struct file *file;
+	struct files_struct *files = current->files;
+
+	*fput_needed = 0;
+	if (atomic_read(&files->count) == 1) {
+		file = fcheck_files(files, fd);
 	} else {
 		rcu_read_lock();
 		file = fcheck_files(files, fd);
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index bfed8447ed8..8bd0ef9286c 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -158,7 +158,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd)
 {
 	struct inode *inode;
 
-	if (nd->flags & LOOKUP_RCU)
+	if (nd && nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 
 	inode = entry->d_inode;
@@ -1283,8 +1283,11 @@ static int fuse_do_setattr(struct dentry *entry, struct iattr *attr,
 	if (err)
 		return err;
 
-	if ((attr->ia_valid & ATTR_OPEN) && fc->atomic_o_trunc)
-		return 0;
+	if (attr->ia_valid & ATTR_OPEN) {
+		if (fc->atomic_o_trunc)
+			return 0;
+		file = NULL;
+	}
 
 	if (attr->ia_valid & ATTR_SIZE)
 		is_truncate = true;
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 95da1bc1c82..9e0832dbb1e 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -86,18 +86,52 @@ struct fuse_file *fuse_file_get(struct fuse_file *ff)
 	return ff;
 }
 
+static void fuse_release_async(struct work_struct *work)
+{
+	struct fuse_req *req;
+	struct fuse_conn *fc;
+	struct path path;
+
+	req = container_of(work, struct fuse_req, misc.release.work);
+	path = req->misc.release.path;
+	fc = get_fuse_conn(path.dentry->d_inode);
+
+	fuse_put_request(fc, req);
+	path_put(&path);
+}
+
 static void fuse_release_end(struct fuse_conn *fc, struct fuse_req *req)
 {
-	path_put(&req->misc.release.path);
+	if (fc->destroy_req) {
+		/*
+		 * If this is a fuseblk mount, then it's possible that
+		 * releasing the path will result in releasing the
+		 * super block and sending the DESTROY request.  If
+		 * the server is single threaded, this would hang.
+		 * For this reason do the path_put() in a separate
+		 * thread.
+		 */
+		atomic_inc(&req->count);
+		INIT_WORK(&req->misc.release.work, fuse_release_async);
+		schedule_work(&req->misc.release.work);
+	} else {
+		path_put(&req->misc.release.path);
+	}
 }
 
-static void fuse_file_put(struct fuse_file *ff)
+static void fuse_file_put(struct fuse_file *ff, bool sync)
 {
 	if (atomic_dec_and_test(&ff->count)) {
 		struct fuse_req *req = ff->reserved_req;
 
-		req->end = fuse_release_end;
-		fuse_request_send_background(ff->fc, req);
+		if (sync) {
+			fuse_request_send(ff->fc, req);
+			path_put(&req->misc.release.path);
+			fuse_put_request(ff->fc, req);
+		} else {
+			req->end = fuse_release_end;
+			fuse_request_send_background(ff->fc, req);
+		}
 		kfree(ff);
 	}
 }
@@ -219,8 +253,12 @@ void fuse_release_common(struct file *file, int opcode)
 	 * Normally this will send the RELEASE request, however if
 	 * some asynchronous READ or WRITE requests are outstanding,
 	 * the sending will be delayed.
+	 *
+	 * Make the release synchronous if this is a fuseblk mount,
+	 * synchronous RELEASE is allowed (and desirable) in this case
+	 * because the server can be trusted not to screw up.
 	 */
-	fuse_file_put(ff);
+	fuse_file_put(ff, ff->fc->destroy_req != NULL);
 }
 
 static int fuse_open(struct inode *inode, struct file *file)
@@ -558,7 +596,7 @@ static void fuse_readpages_end(struct fuse_conn *fc, struct fuse_req *req)
 		page_cache_release(page);
 	}
 	if (req->ff)
-		fuse_file_put(req->ff);
+		fuse_file_put(req->ff, false);
 }
 
 static void fuse_send_readpages(struct fuse_req *req, struct file *file)
@@ -1137,7 +1175,7 @@ static ssize_t fuse_direct_write(struct file *file, const char __user *buf,
 static void fuse_writepage_free(struct fuse_conn *fc, struct fuse_req *req)
 {
 	__free_page(req->pages[0]);
-	fuse_file_put(req->ff);
+	fuse_file_put(req->ff, false);
 }
 
 static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req)
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index ae5744a2f9e..d4286947bc2 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -21,6 +21,7 @@
 #include <linux/rwsem.h>
 #include <linux/rbtree.h>
 #include <linux/poll.h>
+#include <linux/workqueue.h>
 
 /** Max number of pages that can be used in a single read request */
 #define FUSE_MAX_PAGES_PER_REQ 32
@@ -262,7 +263,10 @@ struct fuse_req {
 	/** Data for asynchronous requests */
 	union {
 		struct {
-			struct fuse_release_in in;
+			union {
+				struct fuse_release_in in;
+				struct work_struct work;
+			};
 			struct path path;
 		} release;
 		struct fuse_init_in init_in;
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 9e3f68cc1bd..051b1a08452 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -637,8 +637,10 @@ static int fuse_encode_fh(struct dentry *dentry, u32 *fh, int *max_len,
 	u64 nodeid;
 	u32 generation;
 
-	if (*max_len < len)
+	if (*max_len < len) {
+		*max_len = len;
 		return  255;
+	}
 
 	nodeid = get_fuse_inode(inode)->nodeid;
 	generation = inode->i_generation;
diff --git a/fs/gfs2/dentry.c b/fs/gfs2/dentry.c
index 4a456338b87..0da8da2c991 100644
--- a/fs/gfs2/dentry.c
+++ b/fs/gfs2/dentry.c
@@ -44,7 +44,7 @@ static int gfs2_drevalidate(struct dentry *dentry, struct nameidata *nd)
 	int error;
 	int had_lock = 0;
 
-	if (nd->flags & LOOKUP_RCU)
+	if (nd && nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 
 	parent = dget_parent(dentry);
diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c
index 9023db8184f..b5a5e60df0d 100644
--- a/fs/gfs2/export.c
+++ b/fs/gfs2/export.c
@@ -36,9 +36,13 @@ static int gfs2_encode_fh(struct dentry *dentry, __u32 *p, int *len,
 	struct super_block *sb = inode->i_sb;
 	struct gfs2_inode *ip = GFS2_I(inode);
 
-	if (*len < GFS2_SMALL_FH_SIZE ||
-	    (connectable && *len < GFS2_LARGE_FH_SIZE))
+	if (connectable && (*len < GFS2_LARGE_FH_SIZE)) {
+		*len = GFS2_LARGE_FH_SIZE;
 		return 255;
+	} else if (*len < GFS2_SMALL_FH_SIZE) {
+		*len = GFS2_SMALL_FH_SIZE;
+		return 255;
+	}
 
 	fh[0] = cpu_to_be32(ip->i_no_formal_ino >> 32);
 	fh[1] = cpu_to_be32(ip->i_no_formal_ino & 0xFFFFFFFF);
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index 85ba027d1c4..72c31a315d9 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -59,14 +59,7 @@ static void gfs2_init_gl_aspace_once(void *foo)
 	struct address_space *mapping = (struct address_space *)(gl + 1);
 
 	gfs2_init_glock_once(gl);
-	memset(mapping, 0, sizeof(*mapping));
-	INIT_RADIX_TREE(&mapping->page_tree, GFP_ATOMIC);
-	spin_lock_init(&mapping->tree_lock);
-	spin_lock_init(&mapping->i_mmap_lock);
-	INIT_LIST_HEAD(&mapping->private_list);
-	spin_lock_init(&mapping->private_lock);
-	INIT_RAW_PRIO_TREE_ROOT(&mapping->i_mmap);
-	INIT_LIST_HEAD(&mapping->i_mmap_nonlinear);
+	address_space_init_once(mapping);
 }
 
 /**
diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c
index afa66aaa223..b4d70b13be9 100644
--- a/fs/hfs/dir.c
+++ b/fs/hfs/dir.c
@@ -238,46 +238,22 @@ static int hfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 }
 
 /*
- * hfs_unlink()
+ * hfs_remove()
  *
- * This is the unlink() entry in the inode_operations structure for
- * regular HFS directories.  The purpose is to delete an existing
- * file, given the inode for the parent directory and the name
- * (and its length) of the existing file.
- */
-static int hfs_unlink(struct inode *dir, struct dentry *dentry)
-{
-	struct inode *inode;
-	int res;
-
-	inode = dentry->d_inode;
-	res = hfs_cat_delete(inode->i_ino, dir, &dentry->d_name);
-	if (res)
-		return res;
-
-	drop_nlink(inode);
-	hfs_delete_inode(inode);
-	inode->i_ctime = CURRENT_TIME_SEC;
-	mark_inode_dirty(inode);
-
-	return res;
-}
-
-/*
- * hfs_rmdir()
+ * This serves as both unlink() and rmdir() in the inode_operations
+ * structure for regular HFS directories.  The purpose is to delete
+ * an existing child, given the inode for the parent directory and
+ * the name (and its length) of the existing directory.
  *
- * This is the rmdir() entry in the inode_operations structure for
- * regular HFS directories.  The purpose is to delete an existing
- * directory, given the inode for the parent directory and the name
- * (and its length) of the existing directory.
+ * HFS does not have hardlinks, so both rmdir and unlink set the
+ * link count to 0.  The only difference is the emptiness check.
  */
-static int hfs_rmdir(struct inode *dir, struct dentry *dentry)
+static int hfs_remove(struct inode *dir, struct dentry *dentry)
 {
-	struct inode *inode;
+	struct inode *inode = dentry->d_inode;
 	int res;
 
-	inode = dentry->d_inode;
-	if (inode->i_size != 2)
+	if (S_ISDIR(inode->i_mode) && inode->i_size != 2)
 		return -ENOTEMPTY;
 	res = hfs_cat_delete(inode->i_ino, dir, &dentry->d_name);
 	if (res)
@@ -307,7 +283,7 @@ static int hfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	/* Unlink destination if it already exists */
 	if (new_dentry->d_inode) {
-		res = hfs_unlink(new_dir, new_dentry);
+		res = hfs_remove(new_dir, new_dentry);
 		if (res)
 			return res;
 	}
@@ -332,9 +308,9 @@ const struct file_operations hfs_dir_operations = {
 const struct inode_operations hfs_dir_inode_operations = {
 	.create		= hfs_create,
 	.lookup		= hfs_lookup,
-	.unlink		= hfs_unlink,
+	.unlink		= hfs_remove,
 	.mkdir		= hfs_mkdir,
-	.rmdir		= hfs_rmdir,
+	.rmdir		= hfs_remove,
 	.rename		= hfs_rename,
 	.setattr	= hfs_inode_setattr,
 };
diff --git a/fs/inode.c b/fs/inode.c
index da85e56378f..0647d80accf 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -295,6 +295,20 @@ static void destroy_inode(struct inode *inode)
 		call_rcu(&inode->i_rcu, i_callback);
 }
 
+void address_space_init_once(struct address_space *mapping)
+{
+	memset(mapping, 0, sizeof(*mapping));
+	INIT_RADIX_TREE(&mapping->page_tree, GFP_ATOMIC);
+	spin_lock_init(&mapping->tree_lock);
+	spin_lock_init(&mapping->i_mmap_lock);
+	INIT_LIST_HEAD(&mapping->private_list);
+	spin_lock_init(&mapping->private_lock);
+	INIT_RAW_PRIO_TREE_ROOT(&mapping->i_mmap);
+	INIT_LIST_HEAD(&mapping->i_mmap_nonlinear);
+	mutex_init(&mapping->unmap_mutex);
+}
+EXPORT_SYMBOL(address_space_init_once);
+
 /*
  * These are initializations that only need to be done
  * once, because the fields are idempotent across use
@@ -308,13 +322,7 @@ void inode_init_once(struct inode *inode)
 	INIT_LIST_HEAD(&inode->i_devices);
 	INIT_LIST_HEAD(&inode->i_wb_list);
 	INIT_LIST_HEAD(&inode->i_lru);
-	INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC);
-	spin_lock_init(&inode->i_data.tree_lock);
-	spin_lock_init(&inode->i_data.i_mmap_lock);
-	INIT_LIST_HEAD(&inode->i_data.private_list);
-	spin_lock_init(&inode->i_data.private_lock);
-	INIT_RAW_PRIO_TREE_ROOT(&inode->i_data.i_mmap);
-	INIT_LIST_HEAD(&inode->i_data.i_mmap_nonlinear);
+	address_space_init_once(&inode->i_data);
 	i_size_ordered_init(inode);
 #ifdef CONFIG_FSNOTIFY
 	INIT_HLIST_HEAD(&inode->i_fsnotify_marks);
@@ -540,11 +548,14 @@ void evict_inodes(struct super_block *sb)
 /**
  * invalidate_inodes	- attempt to free all inodes on a superblock
  * @sb:		superblock to operate on
+ * @kill_dirty: flag to guide handling of dirty inodes
  *
  * Attempts to free all inodes for a given superblock.  If there were any
  * busy inodes return a non-zero value, else zero.
+ * If @kill_dirty is set, discard dirty inodes too, otherwise treat
+ * them as busy.
  */
-int invalidate_inodes(struct super_block *sb)
+int invalidate_inodes(struct super_block *sb, bool kill_dirty)
 {
 	int busy = 0;
 	struct inode *inode, *next;
@@ -556,6 +567,10 @@ int invalidate_inodes(struct super_block *sb)
 	list_for_each_entry_safe(inode, next, &sb->s_inodes, i_sb_list) {
 		if (inode->i_state & (I_NEW | I_FREEING | I_WILL_FREE))
 			continue;
+		if (inode->i_state & I_DIRTY && !kill_dirty) {
+			busy = 1;
+			continue;
+		}
 		if (atomic_read(&inode->i_count)) {
 			busy = 1;
 			continue;
diff --git a/fs/internal.h b/fs/internal.h
index 0663568b124..f3d15de44b1 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -106,10 +106,23 @@ extern void put_super(struct super_block *sb);
 struct nameidata;
 extern struct file *nameidata_to_filp(struct nameidata *);
 extern void release_open_intent(struct nameidata *);
+struct open_flags {
+	int open_flag;
+	int mode;
+	int acc_mode;
+	int intent;
+};
+extern struct file *do_filp_open(int dfd, const char *pathname,
+		const struct open_flags *op, int lookup_flags);
+extern struct file *do_file_open_root(struct dentry *, struct vfsmount *,
+		const char *, const struct open_flags *, int lookup_flags);
+
+extern long do_handle_open(int mountdirfd,
+			   struct file_handle __user *ufh, int open_flag);
 
 /*
  * inode.c
  */
 extern int get_nr_dirty_inodes(void);
 extern void evict_inodes(struct super_block *);
-extern int invalidate_inodes(struct super_block *);
+extern int invalidate_inodes(struct super_block *, bool);
diff --git a/fs/isofs/export.c b/fs/isofs/export.c
index ed752cb3847..dd4687ff30d 100644
--- a/fs/isofs/export.c
+++ b/fs/isofs/export.c
@@ -124,9 +124,13 @@ isofs_export_encode_fh(struct dentry *dentry,
 	 * offset of the inode and the upper 16 bits of fh32[1] to
 	 * hold the offset of the parent.
 	 */
-
-	if (len < 3 || (connectable && len < 5))
+	if (connectable && (len < 5)) {
+		*max_len = 5;
+		return 255;
+	} else if (len < 3) {
+		*max_len = 3;
 		return 255;
+	}
 
 	len = 3;
 	fh32[0] = ei->i_iget5_block;
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index 81ead850ddb..3f04a180493 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -809,9 +809,6 @@ static int jfs_link(struct dentry *old_dentry,
 	if (ip->i_nlink == JFS_LINK_MAX)
 		return -EMLINK;
 
-	if (ip->i_nlink == 0)
-		return -ENOENT;
-
 	dquot_initialize(dir);
 
 	tid = txBegin(ip->i_sb, 0);
@@ -1600,7 +1597,7 @@ out:
 
 static int jfs_ci_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
-	if (nd->flags & LOOKUP_RCU)
+	if (nd && nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 	/*
 	 * This is not negative dentry. Always valid.
diff --git a/fs/minix/namei.c b/fs/minix/namei.c
index ce7337ddfdb..6e6777f1b4b 100644
--- a/fs/minix/namei.c
+++ b/fs/minix/namei.c
@@ -213,7 +213,6 @@ static int minix_rename(struct inode * old_dir, struct dentry *old_dentry,
 		new_de = minix_find_entry(new_dentry, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inode_inc_link_count(old_inode);
 		minix_set_link(new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
@@ -225,18 +224,15 @@ static int minix_rename(struct inode * old_dir, struct dentry *old_dentry,
 			if (new_dir->i_nlink >= info->s_link_max)
 				goto out_dir;
 		}
-		inode_inc_link_count(old_inode);
 		err = minix_add_link(new_dentry, old_inode);
-		if (err) {
-			inode_dec_link_count(old_inode);
+		if (err)
 			goto out_dir;
-		}
 		if (dir_de)
 			inode_inc_link_count(new_dir);
 	}
 
 	minix_delete_entry(old_de, old_page);
-	inode_dec_link_count(old_inode);
+	mark_inode_dirty(old_inode);
 
 	if (dir_de) {
 		minix_set_link(dir_de, dir_page, new_dir);
diff --git a/fs/namei.c b/fs/namei.c
index 0087cf9c2c6..0a601cae23d 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -136,7 +136,7 @@ static int do_getname(const char __user *filename, char *page)
 	return retval;
 }
 
-char * getname(const char __user * filename)
+static char *getname_flags(const char __user * filename, int flags)
 {
 	char *tmp, *result;
 
@@ -147,14 +147,21 @@ char * getname(const char __user * filename)
 
 		result = tmp;
 		if (retval < 0) {
-			__putname(tmp);
-			result = ERR_PTR(retval);
+			if (retval != -ENOENT || !(flags & LOOKUP_EMPTY)) {
+				__putname(tmp);
+				result = ERR_PTR(retval);
+			}
 		}
 	}
 	audit_getname(result);
 	return result;
 }
 
+char *getname(const char __user * filename)
+{
+	return getname_flags(filename, 0);
+}
+
 #ifdef CONFIG_AUDITSYSCALL
 void putname(const char *name)
 {
@@ -401,9 +408,11 @@ static int nameidata_drop_rcu(struct nameidata *nd)
 {
 	struct fs_struct *fs = current->fs;
 	struct dentry *dentry = nd->path.dentry;
+	int want_root = 0;
 
 	BUG_ON(!(nd->flags & LOOKUP_RCU));
-	if (nd->root.mnt) {
+	if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) {
+		want_root = 1;
 		spin_lock(&fs->lock);
 		if (nd->root.mnt != fs->root.mnt ||
 				nd->root.dentry != fs->root.dentry)
@@ -414,7 +423,7 @@ static int nameidata_drop_rcu(struct nameidata *nd)
 		goto err;
 	BUG_ON(nd->inode != dentry->d_inode);
 	spin_unlock(&dentry->d_lock);
-	if (nd->root.mnt) {
+	if (want_root) {
 		path_get(&nd->root);
 		spin_unlock(&fs->lock);
 	}
@@ -427,7 +436,7 @@ static int nameidata_drop_rcu(struct nameidata *nd)
 err:
 	spin_unlock(&dentry->d_lock);
 err_root:
-	if (nd->root.mnt)
+	if (want_root)
 		spin_unlock(&fs->lock);
 	return -ECHILD;
 }
@@ -454,9 +463,11 @@ static int nameidata_dentry_drop_rcu(struct nameidata *nd, struct dentry *dentry
 {
 	struct fs_struct *fs = current->fs;
 	struct dentry *parent = nd->path.dentry;
+	int want_root = 0;
 
 	BUG_ON(!(nd->flags & LOOKUP_RCU));
-	if (nd->root.mnt) {
+	if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) {
+		want_root = 1;
 		spin_lock(&fs->lock);
 		if (nd->root.mnt != fs->root.mnt ||
 				nd->root.dentry != fs->root.dentry)
@@ -476,7 +487,7 @@ static int nameidata_dentry_drop_rcu(struct nameidata *nd, struct dentry *dentry
 	parent->d_count++;
 	spin_unlock(&dentry->d_lock);
 	spin_unlock(&parent->d_lock);
-	if (nd->root.mnt) {
+	if (want_root) {
 		path_get(&nd->root);
 		spin_unlock(&fs->lock);
 	}
@@ -490,7 +501,7 @@ err:
 	spin_unlock(&dentry->d_lock);
 	spin_unlock(&parent->d_lock);
 err_root:
-	if (nd->root.mnt)
+	if (want_root)
 		spin_unlock(&fs->lock);
 	return -ECHILD;
 }
@@ -498,8 +509,16 @@ err_root:
 /* Try to drop out of rcu-walk mode if we were in it, otherwise do nothing.  */
 static inline int nameidata_dentry_drop_rcu_maybe(struct nameidata *nd, struct dentry *dentry)
 {
-	if (nd->flags & LOOKUP_RCU)
-		return nameidata_dentry_drop_rcu(nd, dentry);
+	if (nd->flags & LOOKUP_RCU) {
+		if (unlikely(nameidata_dentry_drop_rcu(nd, dentry))) {
+			nd->flags &= ~LOOKUP_RCU;
+			if (!(nd->flags & LOOKUP_ROOT))
+				nd->root.mnt = NULL;
+			rcu_read_unlock();
+			br_read_unlock(vfsmount_lock);
+			return -ECHILD;
+		}
+	}
 	return 0;
 }
 
@@ -518,7 +537,8 @@ static int nameidata_drop_rcu_last(struct nameidata *nd)
 
 	BUG_ON(!(nd->flags & LOOKUP_RCU));
 	nd->flags &= ~LOOKUP_RCU;
-	nd->root.mnt = NULL;
+	if (!(nd->flags & LOOKUP_ROOT))
+		nd->root.mnt = NULL;
 	spin_lock(&dentry->d_lock);
 	if (!__d_rcu_to_refcount(dentry, nd->seq))
 		goto err_unlock;
@@ -539,14 +559,6 @@ err_unlock:
 	return -ECHILD;
 }
 
-/* Try to drop out of rcu-walk mode if we were in it, otherwise do nothing.  */
-static inline int nameidata_drop_rcu_last_maybe(struct nameidata *nd)
-{
-	if (likely(nd->flags & LOOKUP_RCU))
-		return nameidata_drop_rcu_last(nd);
-	return 0;
-}
-
 /**
  * release_open_intent - free up open intent resources
  * @nd: pointer to nameidata
@@ -590,42 +602,8 @@ do_revalidate(struct dentry *dentry, struct nameidata *nd)
 	return dentry;
 }
 
-static inline struct dentry *
-do_revalidate_rcu(struct dentry *dentry, struct nameidata *nd)
-{
-	int status = d_revalidate(dentry, nd);
-	if (likely(status > 0))
-		return dentry;
-	if (status == -ECHILD) {
-		if (nameidata_dentry_drop_rcu(nd, dentry))
-			return ERR_PTR(-ECHILD);
-		return do_revalidate(dentry, nd);
-	}
-	if (status < 0)
-		return ERR_PTR(status);
-	/* Don't d_invalidate in rcu-walk mode */
-	if (nameidata_dentry_drop_rcu(nd, dentry))
-		return ERR_PTR(-ECHILD);
-	if (!d_invalidate(dentry)) {
-		dput(dentry);
-		dentry = NULL;
-	}
-	return dentry;
-}
-
-static inline int need_reval_dot(struct dentry *dentry)
-{
-	if (likely(!(dentry->d_flags & DCACHE_OP_REVALIDATE)))
-		return 0;
-
-	if (likely(!(dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT)))
-		return 0;
-
-	return 1;
-}
-
 /*
- * force_reval_path - force revalidation of a dentry
+ * handle_reval_path - force revalidation of a dentry
  *
  * In some situations the path walking code will trust dentries without
  * revalidating them. This causes problems for filesystems that depend on
@@ -639,27 +617,28 @@ static inline int need_reval_dot(struct dentry *dentry)
  * invalidate the dentry. It's up to the caller to handle putting references
  * to the path if necessary.
  */
-static int
-force_reval_path(struct path *path, struct nameidata *nd)
+static inline int handle_reval_path(struct nameidata *nd)
 {
+	struct dentry *dentry = nd->path.dentry;
 	int status;
-	struct dentry *dentry = path->dentry;
 
-	/*
-	 * only check on filesystems where it's possible for the dentry to
-	 * become stale.
-	 */
-	if (!need_reval_dot(dentry))
+	if (likely(!(nd->flags & LOOKUP_JUMPED)))
+		return 0;
+
+	if (likely(!(dentry->d_flags & DCACHE_OP_REVALIDATE)))
+		return 0;
+
+	if (likely(!(dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT)))
 		return 0;
 
+	/* Note: we do not d_invalidate() */
 	status = d_revalidate(dentry, nd);
 	if (status > 0)
 		return 0;
 
-	if (!status) {
-		d_invalidate(dentry);
+	if (!status)
 		status = -ESTALE;
-	}
+
 	return status;
 }
 
@@ -728,6 +707,7 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l
 		path_put(&nd->path);
 		nd->path = nd->root;
 		path_get(&nd->root);
+		nd->flags |= LOOKUP_JUMPED;
 	}
 	nd->inode = nd->path.dentry->d_inode;
 
@@ -757,20 +737,44 @@ static inline void path_to_nameidata(const struct path *path,
 	nd->path.dentry = path->dentry;
 }
 
+static inline void put_link(struct nameidata *nd, struct path *link, void *cookie)
+{
+	struct inode *inode = link->dentry->d_inode;
+	if (!IS_ERR(cookie) && inode->i_op->put_link)
+		inode->i_op->put_link(link->dentry, nd, cookie);
+	path_put(link);
+}
+
 static __always_inline int
-__do_follow_link(const struct path *link, struct nameidata *nd, void **p)
+follow_link(struct path *link, struct nameidata *nd, void **p)
 {
 	int error;
 	struct dentry *dentry = link->dentry;
 
 	BUG_ON(nd->flags & LOOKUP_RCU);
 
+	if (unlikely(current->total_link_count >= 40)) {
+		*p = ERR_PTR(-ELOOP); /* no ->put_link(), please */
+		path_put_conditional(link, nd);
+		path_put(&nd->path);
+		return -ELOOP;
+	}
+	cond_resched();
+	current->total_link_count++;
+
 	touch_atime(link->mnt, dentry);
 	nd_set_link(nd, NULL);
 
 	if (link->mnt == nd->path.mnt)
 		mntget(link->mnt);
 
+	error = security_inode_follow_link(link->dentry, nd);
+	if (error) {
+		*p = ERR_PTR(error); /* no ->put_link(), please */
+		path_put(&nd->path);
+		return error;
+	}
+
 	nd->last_type = LAST_BIND;
 	*p = dentry->d_inode->i_op->follow_link(dentry, nd);
 	error = PTR_ERR(*p);
@@ -780,56 +784,18 @@ __do_follow_link(const struct path *link, struct nameidata *nd, void **p)
 		if (s)
 			error = __vfs_follow_link(nd, s);
 		else if (nd->last_type == LAST_BIND) {
-			error = force_reval_path(&nd->path, nd);
-			if (error)
+			nd->flags |= LOOKUP_JUMPED;
+			nd->inode = nd->path.dentry->d_inode;
+			if (nd->inode->i_op->follow_link) {
+				/* stepped on a _really_ weird one */
 				path_put(&nd->path);
+				error = -ELOOP;
+			}
 		}
 	}
 	return error;
 }
 
-/*
- * This limits recursive symlink follows to 8, while
- * limiting consecutive symlinks to 40.
- *
- * Without that kind of total limit, nasty chains of consecutive
- * symlinks can cause almost arbitrarily long lookups. 
- */
-static inline int do_follow_link(struct inode *inode, struct path *path, struct nameidata *nd)
-{
-	void *cookie;
-	int err = -ELOOP;
-
-	/* We drop rcu-walk here */
-	if (nameidata_dentry_drop_rcu_maybe(nd, path->dentry))
-		return -ECHILD;
-	BUG_ON(inode != path->dentry->d_inode);
-
-	if (current->link_count >= MAX_NESTED_LINKS)
-		goto loop;
-	if (current->total_link_count >= 40)
-		goto loop;
-	BUG_ON(nd->depth >= MAX_NESTED_LINKS);
-	cond_resched();
-	err = security_inode_follow_link(path->dentry, nd);
-	if (err)
-		goto loop;
-	current->link_count++;
-	current->total_link_count++;
-	nd->depth++;
-	err = __do_follow_link(path, nd, &cookie);
-	if (!IS_ERR(cookie) && path->dentry->d_inode->i_op->put_link)
-		path->dentry->d_inode->i_op->put_link(path->dentry, nd, cookie);
-	path_put(path);
-	current->link_count--;
-	nd->depth--;
-	return err;
-loop:
-	path_put_conditional(path, nd);
-	path_put(&nd->path);
-	return err;
-}
-
 static int follow_up_rcu(struct path *path)
 {
 	struct vfsmount *parent;
@@ -1068,7 +1034,7 @@ static int follow_dotdot_rcu(struct nameidata *nd)
 
 			seq = read_seqcount_begin(&parent->d_seq);
 			if (read_seqcount_retry(&old->d_seq, nd->seq))
-				return -ECHILD;
+				goto failed;
 			inode = parent->d_inode;
 			nd->path.dentry = parent;
 			nd->seq = seq;
@@ -1081,8 +1047,15 @@ static int follow_dotdot_rcu(struct nameidata *nd)
 	}
 	__follow_mount_rcu(nd, &nd->path, &inode, true);
 	nd->inode = inode;
-
 	return 0;
+
+failed:
+	nd->flags &= ~LOOKUP_RCU;
+	if (!(nd->flags & LOOKUP_ROOT))
+		nd->root.mnt = NULL;
+	rcu_read_unlock();
+	br_read_unlock(vfsmount_lock);
+	return -ECHILD;
 }
 
 /*
@@ -1216,68 +1189,85 @@ static int do_lookup(struct nameidata *nd, struct qstr *name,
 {
 	struct vfsmount *mnt = nd->path.mnt;
 	struct dentry *dentry, *parent = nd->path.dentry;
-	struct inode *dir;
+	int need_reval = 1;
+	int status = 1;
 	int err;
 
 	/*
-	 * See if the low-level filesystem might want
-	 * to use its own hash..
-	 */
-	if (unlikely(parent->d_flags & DCACHE_OP_HASH)) {
-		err = parent->d_op->d_hash(parent, nd->inode, name);
-		if (err < 0)
-			return err;
-	}
-
-	/*
 	 * Rename seqlock is not required here because in the off chance
 	 * of a false negative due to a concurrent rename, we're going to
 	 * do the non-racy lookup, below.
 	 */
 	if (nd->flags & LOOKUP_RCU) {
 		unsigned seq;
-
 		*inode = nd->inode;
 		dentry = __d_lookup_rcu(parent, name, &seq, inode);
-		if (!dentry) {
-			if (nameidata_drop_rcu(nd))
-				return -ECHILD;
-			goto need_lookup;
-		}
+		if (!dentry)
+			goto unlazy;
+
 		/* Memory barrier in read_seqcount_begin of child is enough */
 		if (__read_seqcount_retry(&parent->d_seq, nd->seq))
 			return -ECHILD;
-
 		nd->seq = seq;
+
 		if (unlikely(dentry->d_flags & DCACHE_OP_REVALIDATE)) {
-			dentry = do_revalidate_rcu(dentry, nd);
-			if (!dentry)
-				goto need_lookup;
-			if (IS_ERR(dentry))
-				goto fail;
-			if (!(nd->flags & LOOKUP_RCU))
-				goto done;
+			status = d_revalidate(dentry, nd);
+			if (unlikely(status <= 0)) {
+				if (status != -ECHILD)
+					need_reval = 0;
+				goto unlazy;
+			}
 		}
 		path->mnt = mnt;
 		path->dentry = dentry;
 		if (likely(__follow_mount_rcu(nd, path, inode, false)))
 			return 0;
-		if (nameidata_drop_rcu(nd))
-			return -ECHILD;
-		/* fallthru */
+unlazy:
+		if (dentry) {
+			if (nameidata_dentry_drop_rcu(nd, dentry))
+				return -ECHILD;
+		} else {
+			if (nameidata_drop_rcu(nd))
+				return -ECHILD;
+		}
+	} else {
+		dentry = __d_lookup(parent, name);
 	}
-	dentry = __d_lookup(parent, name);
-	if (!dentry)
-		goto need_lookup;
-found:
-	if (unlikely(dentry->d_flags & DCACHE_OP_REVALIDATE)) {
-		dentry = do_revalidate(dentry, nd);
-		if (!dentry)
-			goto need_lookup;
-		if (IS_ERR(dentry))
-			goto fail;
+
+retry:
+	if (unlikely(!dentry)) {
+		struct inode *dir = parent->d_inode;
+		BUG_ON(nd->inode != dir);
+
+		mutex_lock(&dir->i_mutex);
+		dentry = d_lookup(parent, name);
+		if (likely(!dentry)) {
+			dentry = d_alloc_and_lookup(parent, name, nd);
+			if (IS_ERR(dentry)) {
+				mutex_unlock(&dir->i_mutex);
+				return PTR_ERR(dentry);
+			}
+			/* known good */
+			need_reval = 0;
+			status = 1;
+		}
+		mutex_unlock(&dir->i_mutex);
 	}
-done:
+	if (unlikely(dentry->d_flags & DCACHE_OP_REVALIDATE) && need_reval)
+		status = d_revalidate(dentry, nd);
+	if (unlikely(status <= 0)) {
+		if (status < 0) {
+			dput(dentry);
+			return status;
+		}
+		if (!d_invalidate(dentry)) {
+			dput(dentry);
+			dentry = NULL;
+			need_reval = 1;
+			goto retry;
+		}
+	}
+
 	path->mnt = mnt;
 	path->dentry = dentry;
 	err = follow_managed(path, nd->flags);
@@ -1287,39 +1277,113 @@ done:
 	}
 	*inode = path->dentry->d_inode;
 	return 0;
+}
 
-need_lookup:
-	dir = parent->d_inode;
-	BUG_ON(nd->inode != dir);
+static inline int may_lookup(struct nameidata *nd)
+{
+	if (nd->flags & LOOKUP_RCU) {
+		int err = exec_permission(nd->inode, IPERM_FLAG_RCU);
+		if (err != -ECHILD)
+			return err;
+		if (nameidata_drop_rcu(nd))
+			return -ECHILD;
+	}
+	return exec_permission(nd->inode, 0);
+}
 
-	mutex_lock(&dir->i_mutex);
-	/*
-	 * First re-do the cached lookup just in case it was created
-	 * while we waited for the directory semaphore, or the first
-	 * lookup failed due to an unrelated rename.
-	 *
-	 * This could use version numbering or similar to avoid unnecessary
-	 * cache lookups, but then we'd have to do the first lookup in the
-	 * non-racy way. However in the common case here, everything should
-	 * be hot in cache, so would it be a big win?
-	 */
-	dentry = d_lookup(parent, name);
-	if (likely(!dentry)) {
-		dentry = d_alloc_and_lookup(parent, name, nd);
-		mutex_unlock(&dir->i_mutex);
-		if (IS_ERR(dentry))
-			goto fail;
-		goto done;
+static inline int handle_dots(struct nameidata *nd, int type)
+{
+	if (type == LAST_DOTDOT) {
+		if (nd->flags & LOOKUP_RCU) {
+			if (follow_dotdot_rcu(nd))
+				return -ECHILD;
+		} else
+			follow_dotdot(nd);
+	}
+	return 0;
+}
+
+static void terminate_walk(struct nameidata *nd)
+{
+	if (!(nd->flags & LOOKUP_RCU)) {
+		path_put(&nd->path);
+	} else {
+		nd->flags &= ~LOOKUP_RCU;
+		if (!(nd->flags & LOOKUP_ROOT))
+			nd->root.mnt = NULL;
+		rcu_read_unlock();
+		br_read_unlock(vfsmount_lock);
 	}
+}
+
+static inline int walk_component(struct nameidata *nd, struct path *path,
+		struct qstr *name, int type, int follow)
+{
+	struct inode *inode;
+	int err;
 	/*
-	 * Uhhuh! Nasty case: the cache was re-populated while
-	 * we waited on the semaphore. Need to revalidate.
+	 * "." and ".." are special - ".." especially so because it has
+	 * to be able to know about the current root directory and
+	 * parent relationships.
 	 */
-	mutex_unlock(&dir->i_mutex);
-	goto found;
+	if (unlikely(type != LAST_NORM))
+		return handle_dots(nd, type);
+	err = do_lookup(nd, name, path, &inode);
+	if (unlikely(err)) {
+		terminate_walk(nd);
+		return err;
+	}
+	if (!inode) {
+		path_to_nameidata(path, nd);
+		terminate_walk(nd);
+		return -ENOENT;
+	}
+	if (unlikely(inode->i_op->follow_link) && follow) {
+		if (nameidata_dentry_drop_rcu_maybe(nd, path->dentry))
+			return -ECHILD;
+		BUG_ON(inode != path->dentry->d_inode);
+		return 1;
+	}
+	path_to_nameidata(path, nd);
+	nd->inode = inode;
+	return 0;
+}
 
-fail:
-	return PTR_ERR(dentry);
+/*
+ * This limits recursive symlink follows to 8, while
+ * limiting consecutive symlinks to 40.
+ *
+ * Without that kind of total limit, nasty chains of consecutive
+ * symlinks can cause almost arbitrarily long lookups.
+ */
+static inline int nested_symlink(struct path *path, struct nameidata *nd)
+{
+	int res;
+
+	BUG_ON(nd->depth >= MAX_NESTED_LINKS);
+	if (unlikely(current->link_count >= MAX_NESTED_LINKS)) {
+		path_put_conditional(path, nd);
+		path_put(&nd->path);
+		return -ELOOP;
+	}
+
+	nd->depth++;
+	current->link_count++;
+
+	do {
+		struct path link = *path;
+		void *cookie;
+
+		res = follow_link(&link, nd, &cookie);
+		if (!res)
+			res = walk_component(nd, path, &nd->last,
+					     nd->last_type, LOOKUP_FOLLOW);
+		put_link(nd, &link, cookie);
+	} while (res > 0);
+
+	current->link_count--;
+	nd->depth--;
+	return res;
 }
 
 /*
@@ -1339,30 +1403,18 @@ static int link_path_walk(const char *name, struct nameidata *nd)
 	while (*name=='/')
 		name++;
 	if (!*name)
-		goto return_reval;
-
-	if (nd->depth)
-		lookup_flags = LOOKUP_FOLLOW | (nd->flags & LOOKUP_CONTINUE);
+		return 0;
 
 	/* At this point we know we have a real path component. */
 	for(;;) {
-		struct inode *inode;
 		unsigned long hash;
 		struct qstr this;
 		unsigned int c;
+		int type;
 
 		nd->flags |= LOOKUP_CONTINUE;
-		if (nd->flags & LOOKUP_RCU) {
-			err = exec_permission(nd->inode, IPERM_FLAG_RCU);
-			if (err == -ECHILD) {
-				if (nameidata_drop_rcu(nd))
-					return -ECHILD;
-				goto exec_again;
-			}
-		} else {
-exec_again:
-			err = exec_permission(nd->inode, 0);
-		}
+
+		err = may_lookup(nd);
  		if (err)
 			break;
 
@@ -1378,52 +1430,43 @@ exec_again:
 		this.len = name - (const char *) this.name;
 		this.hash = end_name_hash(hash);
 
+		type = LAST_NORM;
+		if (this.name[0] == '.') switch (this.len) {
+			case 2:
+				if (this.name[1] == '.') {
+					type = LAST_DOTDOT;
+					nd->flags |= LOOKUP_JUMPED;
+				}
+				break;
+			case 1:
+				type = LAST_DOT;
+		}
+		if (likely(type == LAST_NORM)) {
+			struct dentry *parent = nd->path.dentry;
+			nd->flags &= ~LOOKUP_JUMPED;
+			if (unlikely(parent->d_flags & DCACHE_OP_HASH)) {
+				err = parent->d_op->d_hash(parent, nd->inode,
+							   &this);
+				if (err < 0)
+					break;
+			}
+		}
+
 		/* remove trailing slashes? */
 		if (!c)
 			goto last_component;
 		while (*++name == '/');
 		if (!*name)
-			goto last_with_slashes;
+			goto last_component;
 
-		/*
-		 * "." and ".." are special - ".." especially so because it has
-		 * to be able to know about the current root directory and
-		 * parent relationships.
-		 */
-		if (this.name[0] == '.') switch (this.len) {
-			default:
-				break;
-			case 2:
-				if (this.name[1] != '.')
-					break;
-				if (nd->flags & LOOKUP_RCU) {
-					if (follow_dotdot_rcu(nd))
-						return -ECHILD;
-				} else
-					follow_dotdot(nd);
-				/* fallthrough */
-			case 1:
-				continue;
-		}
-		/* This does the actual lookups.. */
-		err = do_lookup(nd, &this, &next, &inode);
-		if (err)
-			break;
-		err = -ENOENT;
-		if (!inode)
-			goto out_dput;
+		err = walk_component(nd, &next, &this, type, LOOKUP_FOLLOW);
+		if (err < 0)
+			return err;
 
-		if (inode->i_op->follow_link) {
-			err = do_follow_link(inode, &next, nd);
+		if (err) {
+			err = nested_symlink(&next, nd);
 			if (err)
-				goto return_err;
-			nd->inode = nd->path.dentry->d_inode;
-			err = -ENOENT;
-			if (!nd->inode)
-				break;
-		} else {
-			path_to_nameidata(&next, nd);
-			nd->inode = inode;
+				return err;
 		}
 		err = -ENOTDIR; 
 		if (!nd->inode->i_op->lookup)
@@ -1431,209 +1474,109 @@ exec_again:
 		continue;
 		/* here ends the main loop */
 
-last_with_slashes:
-		lookup_flags |= LOOKUP_FOLLOW | LOOKUP_DIRECTORY;
 last_component:
 		/* Clear LOOKUP_CONTINUE iff it was previously unset */
 		nd->flags &= lookup_flags | ~LOOKUP_CONTINUE;
-		if (lookup_flags & LOOKUP_PARENT)
-			goto lookup_parent;
-		if (this.name[0] == '.') switch (this.len) {
-			default:
-				break;
-			case 2:
-				if (this.name[1] != '.')
-					break;
-				if (nd->flags & LOOKUP_RCU) {
-					if (follow_dotdot_rcu(nd))
-						return -ECHILD;
-				} else
-					follow_dotdot(nd);
-				/* fallthrough */
-			case 1:
-				goto return_reval;
-		}
-		err = do_lookup(nd, &this, &next, &inode);
-		if (err)
-			break;
-		if (inode && unlikely(inode->i_op->follow_link) &&
-		    (lookup_flags & LOOKUP_FOLLOW)) {
-			err = do_follow_link(inode, &next, nd);
-			if (err)
-				goto return_err;
-			nd->inode = nd->path.dentry->d_inode;
-		} else {
-			path_to_nameidata(&next, nd);
-			nd->inode = inode;
-		}
-		err = -ENOENT;
-		if (!nd->inode)
-			break;
-		if (lookup_flags & LOOKUP_DIRECTORY) {
-			err = -ENOTDIR; 
-			if (!nd->inode->i_op->lookup)
-				break;
-		}
-		goto return_base;
-lookup_parent:
 		nd->last = this;
-		nd->last_type = LAST_NORM;
-		if (this.name[0] != '.')
-			goto return_base;
-		if (this.len == 1)
-			nd->last_type = LAST_DOT;
-		else if (this.len == 2 && this.name[1] == '.')
-			nd->last_type = LAST_DOTDOT;
-		else
-			goto return_base;
-return_reval:
-		/*
-		 * We bypassed the ordinary revalidation routines.
-		 * We may need to check the cached dentry for staleness.
-		 */
-		if (need_reval_dot(nd->path.dentry)) {
-			if (nameidata_drop_rcu_last_maybe(nd))
-				return -ECHILD;
-			/* Note: we do not d_invalidate() */
-			err = d_revalidate(nd->path.dentry, nd);
-			if (!err)
-				err = -ESTALE;
-			if (err < 0)
-				break;
-			return 0;
-		}
-return_base:
-		if (nameidata_drop_rcu_last_maybe(nd))
-			return -ECHILD;
+		nd->last_type = type;
 		return 0;
-out_dput:
-		if (!(nd->flags & LOOKUP_RCU))
-			path_put_conditional(&next, nd);
-		break;
 	}
-	if (!(nd->flags & LOOKUP_RCU))
-		path_put(&nd->path);
-return_err:
+	terminate_walk(nd);
 	return err;
 }
 
-static inline int path_walk_rcu(const char *name, struct nameidata *nd)
-{
-	current->total_link_count = 0;
-
-	return link_path_walk(name, nd);
-}
-
-static inline int path_walk_simple(const char *name, struct nameidata *nd)
-{
-	current->total_link_count = 0;
-
-	return link_path_walk(name, nd);
-}
-
-static int path_walk(const char *name, struct nameidata *nd)
-{
-	struct path save = nd->path;
-	int result;
-
-	current->total_link_count = 0;
-
-	/* make sure the stuff we saved doesn't go away */
-	path_get(&save);
-
-	result = link_path_walk(name, nd);
-	if (result == -ESTALE) {
-		/* nd->path had been dropped */
-		current->total_link_count = 0;
-		nd->path = save;
-		path_get(&nd->path);
-		nd->flags |= LOOKUP_REVAL;
-		result = link_path_walk(name, nd);
-	}
-
-	path_put(&save);
-
-	return result;
-}
-
-static void path_finish_rcu(struct nameidata *nd)
-{
-	if (nd->flags & LOOKUP_RCU) {
-		/* RCU dangling. Cancel it. */
-		nd->flags &= ~LOOKUP_RCU;
-		nd->root.mnt = NULL;
-		rcu_read_unlock();
-		br_read_unlock(vfsmount_lock);
-	}
-	if (nd->file)
-		fput(nd->file);
-}
-
-static int path_init_rcu(int dfd, const char *name, unsigned int flags, struct nameidata *nd)
+static int path_init(int dfd, const char *name, unsigned int flags,
+		     struct nameidata *nd, struct file **fp)
 {
 	int retval = 0;
 	int fput_needed;
 	struct file *file;
 
 	nd->last_type = LAST_ROOT; /* if there are only slashes... */
-	nd->flags = flags | LOOKUP_RCU;
+	nd->flags = flags | LOOKUP_JUMPED;
 	nd->depth = 0;
+	if (flags & LOOKUP_ROOT) {
+		struct inode *inode = nd->root.dentry->d_inode;
+		if (*name) {
+			if (!inode->i_op->lookup)
+				return -ENOTDIR;
+			retval = inode_permission(inode, MAY_EXEC);
+			if (retval)
+				return retval;
+		}
+		nd->path = nd->root;
+		nd->inode = inode;
+		if (flags & LOOKUP_RCU) {
+			br_read_lock(vfsmount_lock);
+			rcu_read_lock();
+			nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
+		} else {
+			path_get(&nd->path);
+		}
+		return 0;
+	}
+
 	nd->root.mnt = NULL;
-	nd->file = NULL;
 
 	if (*name=='/') {
-		struct fs_struct *fs = current->fs;
-		unsigned seq;
-
-		br_read_lock(vfsmount_lock);
-		rcu_read_lock();
-
-		do {
-			seq = read_seqcount_begin(&fs->seq);
-			nd->root = fs->root;
-			nd->path = nd->root;
-			nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
-		} while (read_seqcount_retry(&fs->seq, seq));
-
+		if (flags & LOOKUP_RCU) {
+			br_read_lock(vfsmount_lock);
+			rcu_read_lock();
+			set_root_rcu(nd);
+		} else {
+			set_root(nd);
+			path_get(&nd->root);
+		}
+		nd->path = nd->root;
 	} else if (dfd == AT_FDCWD) {
-		struct fs_struct *fs = current->fs;
-		unsigned seq;
-
-		br_read_lock(vfsmount_lock);
-		rcu_read_lock();
+		if (flags & LOOKUP_RCU) {
+			struct fs_struct *fs = current->fs;
+			unsigned seq;
 
-		do {
-			seq = read_seqcount_begin(&fs->seq);
-			nd->path = fs->pwd;
-			nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
-		} while (read_seqcount_retry(&fs->seq, seq));
+			br_read_lock(vfsmount_lock);
+			rcu_read_lock();
 
+			do {
+				seq = read_seqcount_begin(&fs->seq);
+				nd->path = fs->pwd;
+				nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
+			} while (read_seqcount_retry(&fs->seq, seq));
+		} else {
+			get_fs_pwd(current->fs, &nd->path);
+		}
 	} else {
 		struct dentry *dentry;
 
-		file = fget_light(dfd, &fput_needed);
+		file = fget_raw_light(dfd, &fput_needed);
 		retval = -EBADF;
 		if (!file)
 			goto out_fail;
 
 		dentry = file->f_path.dentry;
 
-		retval = -ENOTDIR;
-		if (!S_ISDIR(dentry->d_inode->i_mode))
-			goto fput_fail;
+		if (*name) {
+			retval = -ENOTDIR;
+			if (!S_ISDIR(dentry->d_inode->i_mode))
+				goto fput_fail;
 
-		retval = file_permission(file, MAY_EXEC);
-		if (retval)
-			goto fput_fail;
+			retval = file_permission(file, MAY_EXEC);
+			if (retval)
+				goto fput_fail;
+		}
 
 		nd->path = file->f_path;
-		if (fput_needed)
-			nd->file = file;
-
-		nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
-		br_read_lock(vfsmount_lock);
-		rcu_read_lock();
+		if (flags & LOOKUP_RCU) {
+			if (fput_needed)
+				*fp = file;
+			nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
+			br_read_lock(vfsmount_lock);
+			rcu_read_lock();
+		} else {
+			path_get(&file->f_path);
+			fput_light(file, fput_needed);
+		}
 	}
+
 	nd->inode = nd->path.dentry->d_inode;
 	return 0;
 
@@ -1643,60 +1586,23 @@ out_fail:
 	return retval;
 }
 
-static int path_init(int dfd, const char *name, unsigned int flags, struct nameidata *nd)
+static inline int lookup_last(struct nameidata *nd, struct path *path)
 {
-	int retval = 0;
-	int fput_needed;
-	struct file *file;
+	if (nd->last_type == LAST_NORM && nd->last.name[nd->last.len])
+		nd->flags |= LOOKUP_FOLLOW | LOOKUP_DIRECTORY;
 
-	nd->last_type = LAST_ROOT; /* if there are only slashes... */
-	nd->flags = flags;
-	nd->depth = 0;
-	nd->root.mnt = NULL;
-
-	if (*name=='/') {
-		set_root(nd);
-		nd->path = nd->root;
-		path_get(&nd->root);
-	} else if (dfd == AT_FDCWD) {
-		get_fs_pwd(current->fs, &nd->path);
-	} else {
-		struct dentry *dentry;
-
-		file = fget_light(dfd, &fput_needed);
-		retval = -EBADF;
-		if (!file)
-			goto out_fail;
-
-		dentry = file->f_path.dentry;
-
-		retval = -ENOTDIR;
-		if (!S_ISDIR(dentry->d_inode->i_mode))
-			goto fput_fail;
-
-		retval = file_permission(file, MAY_EXEC);
-		if (retval)
-			goto fput_fail;
-
-		nd->path = file->f_path;
-		path_get(&file->f_path);
-
-		fput_light(file, fput_needed);
-	}
-	nd->inode = nd->path.dentry->d_inode;
-	return 0;
-
-fput_fail:
-	fput_light(file, fput_needed);
-out_fail:
-	return retval;
+	nd->flags &= ~LOOKUP_PARENT;
+	return walk_component(nd, path, &nd->last, nd->last_type,
+					nd->flags & LOOKUP_FOLLOW);
 }
 
 /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */
-static int do_path_lookup(int dfd, const char *name,
+static int path_lookupat(int dfd, const char *name,
 				unsigned int flags, struct nameidata *nd)
 {
-	int retval;
+	struct file *base = NULL;
+	struct path path;
+	int err;
 
 	/*
 	 * Path walking is largely split up into 2 different synchronisation
@@ -1712,44 +1618,75 @@ static int do_path_lookup(int dfd, const char *name,
 	 * be handled by restarting a traditional ref-walk (which will always
 	 * be able to complete).
 	 */
-	retval = path_init_rcu(dfd, name, flags, nd);
-	if (unlikely(retval))
-		return retval;
-	retval = path_walk_rcu(name, nd);
-	path_finish_rcu(nd);
-	if (nd->root.mnt) {
-		path_put(&nd->root);
-		nd->root.mnt = NULL;
+	err = path_init(dfd, name, flags | LOOKUP_PARENT, nd, &base);
+
+	if (unlikely(err))
+		return err;
+
+	current->total_link_count = 0;
+	err = link_path_walk(name, nd);
+
+	if (!err && !(flags & LOOKUP_PARENT)) {
+		err = lookup_last(nd, &path);
+		while (err > 0) {
+			void *cookie;
+			struct path link = path;
+			nd->flags |= LOOKUP_PARENT;
+			err = follow_link(&link, nd, &cookie);
+			if (!err)
+				err = lookup_last(nd, &path);
+			put_link(nd, &link, cookie);
+		}
 	}
 
-	if (unlikely(retval == -ECHILD || retval == -ESTALE)) {
-		/* slower, locked walk */
-		if (retval == -ESTALE)
-			flags |= LOOKUP_REVAL;
-		retval = path_init(dfd, name, flags, nd);
-		if (unlikely(retval))
-			return retval;
-		retval = path_walk(name, nd);
-		if (nd->root.mnt) {
-			path_put(&nd->root);
-			nd->root.mnt = NULL;
+	if (nd->flags & LOOKUP_RCU) {
+		/* went all way through without dropping RCU */
+		BUG_ON(err);
+		if (nameidata_drop_rcu_last(nd))
+			err = -ECHILD;
+	}
+
+	if (!err)
+		err = handle_reval_path(nd);
+
+	if (!err && nd->flags & LOOKUP_DIRECTORY) {
+		if (!nd->inode->i_op->lookup) {
+			path_put(&nd->path);
+			return -ENOTDIR;
 		}
 	}
 
+	if (base)
+		fput(base);
+
+	if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT)) {
+		path_put(&nd->root);
+		nd->root.mnt = NULL;
+	}
+	return err;
+}
+
+static int do_path_lookup(int dfd, const char *name,
+				unsigned int flags, struct nameidata *nd)
+{
+	int retval = path_lookupat(dfd, name, flags | LOOKUP_RCU, nd);
+	if (unlikely(retval == -ECHILD))
+		retval = path_lookupat(dfd, name, flags, nd);
+	if (unlikely(retval == -ESTALE))
+		retval = path_lookupat(dfd, name, flags | LOOKUP_REVAL, nd);
+
 	if (likely(!retval)) {
 		if (unlikely(!audit_dummy_context())) {
 			if (nd->path.dentry && nd->inode)
 				audit_inode(name, nd->path.dentry);
 		}
 	}
-
 	return retval;
 }
 
-int path_lookup(const char *name, unsigned int flags,
-			struct nameidata *nd)
+int kern_path_parent(const char *name, struct nameidata *nd)
 {
-	return do_path_lookup(AT_FDCWD, name, flags, nd);
+	return do_path_lookup(AT_FDCWD, name, LOOKUP_PARENT, nd);
 }
 
 int kern_path(const char *name, unsigned int flags, struct path *path)
@@ -1773,29 +1710,10 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt,
 		    const char *name, unsigned int flags,
 		    struct nameidata *nd)
 {
-	int retval;
-
-	/* same as do_path_lookup */
-	nd->last_type = LAST_ROOT;
-	nd->flags = flags;
-	nd->depth = 0;
-
-	nd->path.dentry = dentry;
-	nd->path.mnt = mnt;
-	path_get(&nd->path);
-	nd->root = nd->path;
-	path_get(&nd->root);
-	nd->inode = nd->path.dentry->d_inode;
-
-	retval = path_walk(name, nd);
-	if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry &&
-				nd->inode))
-		audit_inode(name, nd->path.dentry);
-
-	path_put(&nd->root);
-	nd->root.mnt = NULL;
-
-	return retval;
+	nd->root.dentry = dentry;
+	nd->root.mnt = mnt;
+	/* the first argument of do_path_lookup() is ignored with LOOKUP_ROOT */
+	return do_path_lookup(AT_FDCWD, name, flags | LOOKUP_ROOT, nd);
 }
 
 static struct dentry *__lookup_hash(struct qstr *name,
@@ -1810,17 +1728,6 @@ static struct dentry *__lookup_hash(struct qstr *name,
 		return ERR_PTR(err);
 
 	/*
-	 * See if the low-level filesystem might want
-	 * to use its own hash..
-	 */
-	if (base->d_flags & DCACHE_OP_HASH) {
-		err = base->d_op->d_hash(base, inode, name);
-		dentry = ERR_PTR(err);
-		if (err < 0)
-			goto out;
-	}
-
-	/*
 	 * Don't bother with __d_lookup: callers are for creat as
 	 * well as unlink, so a lot of the time it would cost
 	 * a double lookup.
@@ -1832,7 +1739,7 @@ static struct dentry *__lookup_hash(struct qstr *name,
 
 	if (!dentry)
 		dentry = d_alloc_and_lookup(base, name, nd);
-out:
+
 	return dentry;
 }
 
@@ -1846,28 +1753,6 @@ static struct dentry *lookup_hash(struct nameidata *nd)
 	return __lookup_hash(&nd->last, nd->path.dentry, nd);
 }
 
-static int __lookup_one_len(const char *name, struct qstr *this,
-		struct dentry *base, int len)
-{
-	unsigned long hash;
-	unsigned int c;
-
-	this->name = name;
-	this->len = len;
-	if (!len)
-		return -EACCES;
-
-	hash = init_name_hash();
-	while (len--) {
-		c = *(const unsigned char *)name++;
-		if (c == '/' || c == '\0')
-			return -EACCES;
-		hash = partial_name_hash(c, hash);
-	}
-	this->hash = end_name_hash(hash);
-	return 0;
-}
-
 /**
  * lookup_one_len - filesystem helper to lookup single pathname component
  * @name:	pathname component to lookup
@@ -1881,14 +1766,34 @@ static int __lookup_one_len(const char *name, struct qstr *this,
  */
 struct dentry *lookup_one_len(const char *name, struct dentry *base, int len)
 {
-	int err;
 	struct qstr this;
+	unsigned long hash;
+	unsigned int c;
 
 	WARN_ON_ONCE(!mutex_is_locked(&base->d_inode->i_mutex));
 
-	err = __lookup_one_len(name, &this, base, len);
-	if (err)
-		return ERR_PTR(err);
+	this.name = name;
+	this.len = len;
+	if (!len)
+		return ERR_PTR(-EACCES);
+
+	hash = init_name_hash();
+	while (len--) {
+		c = *(const unsigned char *)name++;
+		if (c == '/' || c == '\0')
+			return ERR_PTR(-EACCES);
+		hash = partial_name_hash(c, hash);
+	}
+	this.hash = end_name_hash(hash);
+	/*
+	 * See if the low-level filesystem might want
+	 * to use its own hash..
+	 */
+	if (base->d_flags & DCACHE_OP_HASH) {
+		int err = base->d_op->d_hash(base, base->d_inode, &this);
+		if (err < 0)
+			return ERR_PTR(err);
+	}
 
 	return __lookup_hash(&this, base, NULL);
 }
@@ -1897,7 +1802,7 @@ int user_path_at(int dfd, const char __user *name, unsigned flags,
 		 struct path *path)
 {
 	struct nameidata nd;
-	char *tmp = getname(name);
+	char *tmp = getname_flags(name, flags);
 	int err = PTR_ERR(tmp);
 	if (!IS_ERR(tmp)) {
 
@@ -2077,12 +1982,16 @@ int vfs_create(struct inode *dir, struct dentry *dentry, int mode,
 	return error;
 }
 
-int may_open(struct path *path, int acc_mode, int flag)
+static int may_open(struct path *path, int acc_mode, int flag)
 {
 	struct dentry *dentry = path->dentry;
 	struct inode *inode = dentry->d_inode;
 	int error;
 
+	/* O_PATH? */
+	if (!acc_mode)
+		return 0;
+
 	if (!inode)
 		return -ENOENT;
 
@@ -2151,34 +2060,6 @@ static int handle_truncate(struct file *filp)
 }
 
 /*
- * Be careful about ever adding any more callers of this
- * function.  Its flags must be in the namei format, not
- * what get passed to sys_open().
- */
-static int __open_namei_create(struct nameidata *nd, struct path *path,
-				int open_flag, int mode)
-{
-	int error;
-	struct dentry *dir = nd->path.dentry;
-
-	if (!IS_POSIXACL(dir->d_inode))
-		mode &= ~current_umask();
-	error = security_path_mknod(&nd->path, path->dentry, mode, 0);
-	if (error)
-		goto out_unlock;
-	error = vfs_create(dir->d_inode, path->dentry, mode, nd);
-out_unlock:
-	mutex_unlock(&dir->d_inode->i_mutex);
-	dput(nd->path.dentry);
-	nd->path.dentry = path->dentry;
-
-	if (error)
-		return error;
-	/* Don't check for write permission, don't truncate */
-	return may_open(&nd->path, 0, open_flag & ~O_TRUNC);
-}
-
-/*
  * Note that while the flag value (low two bits) for sys_open means:
  *	00 - read-only
  *	01 - write-only
@@ -2202,126 +2083,115 @@ static inline int open_to_namei_flags(int flag)
 	return flag;
 }
 
-static int open_will_truncate(int flag, struct inode *inode)
-{
-	/*
-	 * We'll never write to the fs underlying
-	 * a device file.
-	 */
-	if (special_file(inode->i_mode))
-		return 0;
-	return (flag & O_TRUNC);
-}
-
-static struct file *finish_open(struct nameidata *nd,
-				int open_flag, int acc_mode)
-{
-	struct file *filp;
-	int will_truncate;
-	int error;
-
-	will_truncate = open_will_truncate(open_flag, nd->path.dentry->d_inode);
-	if (will_truncate) {
-		error = mnt_want_write(nd->path.mnt);
-		if (error)
-			goto exit;
-	}
-	error = may_open(&nd->path, acc_mode, open_flag);
-	if (error) {
-		if (will_truncate)
-			mnt_drop_write(nd->path.mnt);
-		goto exit;
-	}
-	filp = nameidata_to_filp(nd);
-	if (!IS_ERR(filp)) {
-		error = ima_file_check(filp, acc_mode);
-		if (error) {
-			fput(filp);
-			filp = ERR_PTR(error);
-		}
-	}
-	if (!IS_ERR(filp)) {
-		if (will_truncate) {
-			error = handle_truncate(filp);
-			if (error) {
-				fput(filp);
-				filp = ERR_PTR(error);
-			}
-		}
-	}
-	/*
-	 * It is now safe to drop the mnt write
-	 * because the filp has had a write taken
-	 * on its behalf.
-	 */
-	if (will_truncate)
-		mnt_drop_write(nd->path.mnt);
-	path_put(&nd->path);
-	return filp;
-
-exit:
-	path_put(&nd->path);
-	return ERR_PTR(error);
-}
-
 /*
- * Handle O_CREAT case for do_filp_open
+ * Handle the last step of open()
  */
 static struct file *do_last(struct nameidata *nd, struct path *path,
-			    int open_flag, int acc_mode,
-			    int mode, const char *pathname)
+			    const struct open_flags *op, const char *pathname)
 {
 	struct dentry *dir = nd->path.dentry;
+	struct dentry *dentry;
+	int open_flag = op->open_flag;
+	int will_truncate = open_flag & O_TRUNC;
+	int want_write = 0;
+	int acc_mode = op->acc_mode;
 	struct file *filp;
-	int error = -EISDIR;
+	int error;
+
+	nd->flags &= ~LOOKUP_PARENT;
+	nd->flags |= op->intent;
 
 	switch (nd->last_type) {
 	case LAST_DOTDOT:
-		follow_dotdot(nd);
-		dir = nd->path.dentry;
 	case LAST_DOT:
-		if (need_reval_dot(dir)) {
-			int status = d_revalidate(nd->path.dentry, nd);
-			if (!status)
-				status = -ESTALE;
-			if (status < 0) {
-				error = status;
-				goto exit;
-			}
-		}
+		error = handle_dots(nd, nd->last_type);
+		if (error)
+			return ERR_PTR(error);
 		/* fallthrough */
 	case LAST_ROOT:
-		goto exit;
+		if (nd->flags & LOOKUP_RCU) {
+			if (nameidata_drop_rcu_last(nd))
+				return ERR_PTR(-ECHILD);
+		}
+		error = handle_reval_path(nd);
+		if (error)
+			goto exit;
+		audit_inode(pathname, nd->path.dentry);
+		if (open_flag & O_CREAT) {
+			error = -EISDIR;
+			goto exit;
+		}
+		goto ok;
 	case LAST_BIND:
+		/* can't be RCU mode here */
+		error = handle_reval_path(nd);
+		if (error)
+			goto exit;
 		audit_inode(pathname, dir);
 		goto ok;
 	}
 
+	if (!(open_flag & O_CREAT)) {
+		int symlink_ok = 0;
+		if (nd->last.name[nd->last.len])
+			nd->flags |= LOOKUP_FOLLOW | LOOKUP_DIRECTORY;
+		if (open_flag & O_PATH && !(nd->flags & LOOKUP_FOLLOW))
+			symlink_ok = 1;
+		/* we _can_ be in RCU mode here */
+		error = walk_component(nd, path, &nd->last, LAST_NORM,
+					!symlink_ok);
+		if (error < 0)
+			return ERR_PTR(error);
+		if (error) /* symlink */
+			return NULL;
+		/* sayonara */
+		if (nd->flags & LOOKUP_RCU) {
+			if (nameidata_drop_rcu_last(nd))
+				return ERR_PTR(-ECHILD);
+		}
+
+		error = -ENOTDIR;
+		if (nd->flags & LOOKUP_DIRECTORY) {
+			if (!nd->inode->i_op->lookup)
+				goto exit;
+		}
+		audit_inode(pathname, nd->path.dentry);
+		goto ok;
+	}
+
+	/* create side of things */
+
+	if (nd->flags & LOOKUP_RCU) {
+		if (nameidata_drop_rcu_last(nd))
+			return ERR_PTR(-ECHILD);
+	}
+
+	audit_inode(pathname, dir);
+	error = -EISDIR;
 	/* trailing slashes? */
 	if (nd->last.name[nd->last.len])
 		goto exit;
 
 	mutex_lock(&dir->d_inode->i_mutex);
 
-	path->dentry = lookup_hash(nd);
-	path->mnt = nd->path.mnt;
-
-	error = PTR_ERR(path->dentry);
-	if (IS_ERR(path->dentry)) {
+	dentry = lookup_hash(nd);
+	error = PTR_ERR(dentry);
+	if (IS_ERR(dentry)) {
 		mutex_unlock(&dir->d_inode->i_mutex);
 		goto exit;
 	}
 
-	if (IS_ERR(nd->intent.open.file)) {
-		error = PTR_ERR(nd->intent.open.file);
-		goto exit_mutex_unlock;
-	}
+	path->dentry = dentry;
+	path->mnt = nd->path.mnt;
 
 	/* Negative dentry, just create the file */
-	if (!path->dentry->d_inode) {
+	if (!dentry->d_inode) {
+		int mode = op->mode;
+		if (!IS_POSIXACL(dir->d_inode))
+			mode &= ~current_umask();
 		/*
 		 * This write is needed to ensure that a
-		 * ro->rw transition does not occur between
+		 * rw->ro transition does not occur between
 		 * the time when the file is created and when
 		 * a permanent write count is taken through
 		 * the 'struct file' in nameidata_to_filp().
@@ -2329,22 +2199,21 @@ static struct file *do_last(struct nameidata *nd, struct path *path,
 		error = mnt_want_write(nd->path.mnt);
 		if (error)
 			goto exit_mutex_unlock;
-		error = __open_namei_create(nd, path, open_flag, mode);
-		if (error) {
-			mnt_drop_write(nd->path.mnt);
-			goto exit;
-		}
-		filp = nameidata_to_filp(nd);
-		mnt_drop_write(nd->path.mnt);
-		path_put(&nd->path);
-		if (!IS_ERR(filp)) {
-			error = ima_file_check(filp, acc_mode);
-			if (error) {
-				fput(filp);
-				filp = ERR_PTR(error);
-			}
-		}
-		return filp;
+		want_write = 1;
+		/* Don't check for write permission, don't truncate */
+		open_flag &= ~O_TRUNC;
+		will_truncate = 0;
+		acc_mode = MAY_OPEN;
+		error = security_path_mknod(&nd->path, dentry, mode, 0);
+		if (error)
+			goto exit_mutex_unlock;
+		error = vfs_create(dir->d_inode, dentry, mode, nd);
+		if (error)
+			goto exit_mutex_unlock;
+		mutex_unlock(&dir->d_inode->i_mutex);
+		dput(nd->path.dentry);
+		nd->path.dentry = dentry;
+		goto common;
 	}
 
 	/*
@@ -2374,7 +2243,40 @@ static struct file *do_last(struct nameidata *nd, struct path *path,
 	if (S_ISDIR(nd->inode->i_mode))
 		goto exit;
 ok:
-	filp = finish_open(nd, open_flag, acc_mode);
+	if (!S_ISREG(nd->inode->i_mode))
+		will_truncate = 0;
+
+	if (will_truncate) {
+		error = mnt_want_write(nd->path.mnt);
+		if (error)
+			goto exit;
+		want_write = 1;
+	}
+common:
+	error = may_open(&nd->path, acc_mode, open_flag);
+	if (error)
+		goto exit;
+	filp = nameidata_to_filp(nd);
+	if (!IS_ERR(filp)) {
+		error = ima_file_check(filp, op->acc_mode);
+		if (error) {
+			fput(filp);
+			filp = ERR_PTR(error);
+		}
+	}
+	if (!IS_ERR(filp)) {
+		if (will_truncate) {
+			error = handle_truncate(filp);
+			if (error) {
+				fput(filp);
+				filp = ERR_PTR(error);
+			}
+		}
+	}
+out:
+	if (want_write)
+		mnt_drop_write(nd->path.mnt);
+	path_put(&nd->path);
 	return filp;
 
 exit_mutex_unlock:
@@ -2382,197 +2284,103 @@ exit_mutex_unlock:
 exit_dput:
 	path_put_conditional(path, nd);
 exit:
-	path_put(&nd->path);
-	return ERR_PTR(error);
+	filp = ERR_PTR(error);
+	goto out;
 }
 
-/*
- * Note that the low bits of the passed in "open_flag"
- * are not the same as in the local variable "flag". See
- * open_to_namei_flags() for more details.
- */
-struct file *do_filp_open(int dfd, const char *pathname,
-		int open_flag, int mode, int acc_mode)
+static struct file *path_openat(int dfd, const char *pathname,
+		struct nameidata *nd, const struct open_flags *op, int flags)
 {
+	struct file *base = NULL;
 	struct file *filp;
-	struct nameidata nd;
-	int error;
 	struct path path;
-	int count = 0;
-	int flag = open_to_namei_flags(open_flag);
-	int flags;
-
-	if (!(open_flag & O_CREAT))
-		mode = 0;
-
-	/* Must never be set by userspace */
-	open_flag &= ~FMODE_NONOTIFY;
-
-	/*
-	 * O_SYNC is implemented as __O_SYNC|O_DSYNC.  As many places only
-	 * check for O_DSYNC if the need any syncing at all we enforce it's
-	 * always set instead of having to deal with possibly weird behaviour
-	 * for malicious applications setting only __O_SYNC.
-	 */
-	if (open_flag & __O_SYNC)
-		open_flag |= O_DSYNC;
-
-	if (!acc_mode)
-		acc_mode = MAY_OPEN | ACC_MODE(open_flag);
-
-	/* O_TRUNC implies we need access checks for write permissions */
-	if (open_flag & O_TRUNC)
-		acc_mode |= MAY_WRITE;
-
-	/* Allow the LSM permission hook to distinguish append 
-	   access from general write access. */
-	if (open_flag & O_APPEND)
-		acc_mode |= MAY_APPEND;
-
-	flags = LOOKUP_OPEN;
-	if (open_flag & O_CREAT) {
-		flags |= LOOKUP_CREATE;
-		if (open_flag & O_EXCL)
-			flags |= LOOKUP_EXCL;
-	}
-	if (open_flag & O_DIRECTORY)
-		flags |= LOOKUP_DIRECTORY;
-	if (!(open_flag & O_NOFOLLOW))
-		flags |= LOOKUP_FOLLOW;
+	int error;
 
 	filp = get_empty_filp();
 	if (!filp)
 		return ERR_PTR(-ENFILE);
 
-	filp->f_flags = open_flag;
-	nd.intent.open.file = filp;
-	nd.intent.open.flags = flag;
-	nd.intent.open.create_mode = mode;
-
-	if (open_flag & O_CREAT)
-		goto creat;
+	filp->f_flags = op->open_flag;
+	nd->intent.open.file = filp;
+	nd->intent.open.flags = open_to_namei_flags(op->open_flag);
+	nd->intent.open.create_mode = op->mode;
 
-	/* !O_CREAT, simple open */
-	error = do_path_lookup(dfd, pathname, flags, &nd);
+	error = path_init(dfd, pathname, flags | LOOKUP_PARENT, nd, &base);
 	if (unlikely(error))
 		goto out_filp;
-	error = -ELOOP;
-	if (!(nd.flags & LOOKUP_FOLLOW)) {
-		if (nd.inode->i_op->follow_link)
-			goto out_path;
-	}
-	error = -ENOTDIR;
-	if (nd.flags & LOOKUP_DIRECTORY) {
-		if (!nd.inode->i_op->lookup)
-			goto out_path;
-	}
-	audit_inode(pathname, nd.path.dentry);
-	filp = finish_open(&nd, open_flag, acc_mode);
-	release_open_intent(&nd);
-	return filp;
 
-creat:
-	/* OK, have to create the file. Find the parent. */
-	error = path_init_rcu(dfd, pathname,
-			LOOKUP_PARENT | (flags & LOOKUP_REVAL), &nd);
-	if (error)
-		goto out_filp;
-	error = path_walk_rcu(pathname, &nd);
-	path_finish_rcu(&nd);
-	if (unlikely(error == -ECHILD || error == -ESTALE)) {
-		/* slower, locked walk */
-		if (error == -ESTALE) {
-reval:
-			flags |= LOOKUP_REVAL;
-		}
-		error = path_init(dfd, pathname,
-				LOOKUP_PARENT | (flags & LOOKUP_REVAL), &nd);
-		if (error)
-			goto out_filp;
-
-		error = path_walk_simple(pathname, &nd);
-	}
+	current->total_link_count = 0;
+	error = link_path_walk(pathname, nd);
 	if (unlikely(error))
 		goto out_filp;
-	if (unlikely(!audit_dummy_context()))
-		audit_inode(pathname, nd.path.dentry);
 
-	/*
-	 * We have the parent and last component.
-	 */
-	nd.flags = flags;
-	filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname);
+	filp = do_last(nd, &path, op, pathname);
 	while (unlikely(!filp)) { /* trailing symlink */
 		struct path link = path;
-		struct inode *linki = link.dentry->d_inode;
 		void *cookie;
-		error = -ELOOP;
-		if (!(nd.flags & LOOKUP_FOLLOW))
-			goto exit_dput;
-		if (count++ == 32)
-			goto exit_dput;
-		/*
-		 * This is subtle. Instead of calling do_follow_link() we do
-		 * the thing by hands. The reason is that this way we have zero
-		 * link_count and path_walk() (called from ->follow_link)
-		 * honoring LOOKUP_PARENT.  After that we have the parent and
-		 * last component, i.e. we are in the same situation as after
-		 * the first path_walk().  Well, almost - if the last component
-		 * is normal we get its copy stored in nd->last.name and we will
-		 * have to putname() it when we are done. Procfs-like symlinks
-		 * just set LAST_BIND.
-		 */
-		nd.flags |= LOOKUP_PARENT;
-		error = security_inode_follow_link(link.dentry, &nd);
-		if (error)
-			goto exit_dput;
-		error = __do_follow_link(&link, &nd, &cookie);
-		if (unlikely(error)) {
-			if (!IS_ERR(cookie) && linki->i_op->put_link)
-				linki->i_op->put_link(link.dentry, &nd, cookie);
-			/* nd.path had been dropped */
-			nd.path = link;
-			goto out_path;
+		if (!(nd->flags & LOOKUP_FOLLOW)) {
+			path_put_conditional(&path, nd);
+			path_put(&nd->path);
+			filp = ERR_PTR(-ELOOP);
+			break;
 		}
-		nd.flags &= ~LOOKUP_PARENT;
-		filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname);
-		if (linki->i_op->put_link)
-			linki->i_op->put_link(link.dentry, &nd, cookie);
-		path_put(&link);
+		nd->flags |= LOOKUP_PARENT;
+		nd->flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL);
+		error = follow_link(&link, nd, &cookie);
+		if (unlikely(error))
+			filp = ERR_PTR(error);
+		else
+			filp = do_last(nd, &path, op, pathname);
+		put_link(nd, &link, cookie);
 	}
 out:
-	if (nd.root.mnt)
-		path_put(&nd.root);
-	if (filp == ERR_PTR(-ESTALE) && !(flags & LOOKUP_REVAL))
-		goto reval;
-	release_open_intent(&nd);
+	if (nd->root.mnt && !(nd->flags & LOOKUP_ROOT))
+		path_put(&nd->root);
+	if (base)
+		fput(base);
+	release_open_intent(nd);
 	return filp;
 
-exit_dput:
-	path_put_conditional(&path, &nd);
-out_path:
-	path_put(&nd.path);
 out_filp:
 	filp = ERR_PTR(error);
 	goto out;
 }
 
-/**
- * filp_open - open file and return file pointer
- *
- * @filename:	path to open
- * @flags:	open flags as per the open(2) second argument
- * @mode:	mode for the new file if O_CREAT is set, else ignored
- *
- * This is the helper to open a file from kernelspace if you really
- * have to.  But in generally you should not do this, so please move
- * along, nothing to see here..
- */
-struct file *filp_open(const char *filename, int flags, int mode)
+struct file *do_filp_open(int dfd, const char *pathname,
+		const struct open_flags *op, int flags)
+{
+	struct nameidata nd;
+	struct file *filp;
+
+	filp = path_openat(dfd, pathname, &nd, op, flags | LOOKUP_RCU);
+	if (unlikely(filp == ERR_PTR(-ECHILD)))
+		filp = path_openat(dfd, pathname, &nd, op, flags);
+	if (unlikely(filp == ERR_PTR(-ESTALE)))
+		filp = path_openat(dfd, pathname, &nd, op, flags | LOOKUP_REVAL);
+	return filp;
+}
+
+struct file *do_file_open_root(struct dentry *dentry, struct vfsmount *mnt,
+		const char *name, const struct open_flags *op, int flags)
 {
-	return do_filp_open(AT_FDCWD, filename, flags, mode, 0);
+	struct nameidata nd;
+	struct file *file;
+
+	nd.root.mnt = mnt;
+	nd.root.dentry = dentry;
+
+	flags |= LOOKUP_ROOT;
+
+	if (dentry->d_inode->i_op->follow_link && op->intent & LOOKUP_OPEN)
+		return ERR_PTR(-ELOOP);
+
+	file = path_openat(-1, name, &nd, op, flags | LOOKUP_RCU);
+	if (unlikely(file == ERR_PTR(-ECHILD)))
+		file = path_openat(-1, name, &nd, op, flags);
+	if (unlikely(file == ERR_PTR(-ESTALE)))
+		file = path_openat(-1, name, &nd, op, flags | LOOKUP_REVAL);
+	return file;
 }
-EXPORT_SYMBOL(filp_open);
 
 /**
  * lookup_create - lookup a dentry, creating it if it doesn't exist
@@ -3111,7 +2919,11 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de
 		return error;
 
 	mutex_lock(&inode->i_mutex);
-	error = dir->i_op->link(old_dentry, dir, new_dentry);
+	/* Make sure we don't allow creating hardlink to an unlinked file */
+	if (inode->i_nlink == 0)
+		error =  -ENOENT;
+	else
+		error = dir->i_op->link(old_dentry, dir, new_dentry);
 	mutex_unlock(&inode->i_mutex);
 	if (!error)
 		fsnotify_link(dir, inode, new_dentry);
@@ -3133,15 +2945,27 @@ SYSCALL_DEFINE5(linkat, int, olddfd, const char __user *, oldname,
 	struct dentry *new_dentry;
 	struct nameidata nd;
 	struct path old_path;
+	int how = 0;
 	int error;
 	char *to;
 
-	if ((flags & ~AT_SYMLINK_FOLLOW) != 0)
+	if ((flags & ~(AT_SYMLINK_FOLLOW | AT_EMPTY_PATH)) != 0)
 		return -EINVAL;
+	/*
+	 * To use null names we require CAP_DAC_READ_SEARCH
+	 * This ensures that not everyone will be able to create
+	 * handlink using the passed filedescriptor.
+	 */
+	if (flags & AT_EMPTY_PATH) {
+		if (!capable(CAP_DAC_READ_SEARCH))
+			return -ENOENT;
+		how = LOOKUP_EMPTY;
+	}
+
+	if (flags & AT_SYMLINK_FOLLOW)
+		how |= LOOKUP_FOLLOW;
 
-	error = user_path_at(olddfd, oldname,
-			     flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0,
-			     &old_path);
+	error = user_path_at(olddfd, oldname, how, &old_path);
 	if (error)
 		return error;
 
@@ -3578,7 +3402,7 @@ EXPORT_SYMBOL(page_readlink);
 EXPORT_SYMBOL(__page_symlink);
 EXPORT_SYMBOL(page_symlink);
 EXPORT_SYMBOL(page_symlink_inode_operations);
-EXPORT_SYMBOL(path_lookup);
+EXPORT_SYMBOL(kern_path_parent);
 EXPORT_SYMBOL(kern_path);
 EXPORT_SYMBOL(vfs_path_lookup);
 EXPORT_SYMBOL(inode_permission);
diff --git a/fs/namespace.c b/fs/namespace.c
index 7b0b9537169..dffe6f49ab9 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1002,6 +1002,18 @@ const struct seq_operations mounts_op = {
 	.show	= show_vfsmnt
 };
 
+static int uuid_is_nil(u8 *uuid)
+{
+	int i;
+	u8  *cp = (u8 *)uuid;
+
+	for (i = 0; i < 16; i++) {
+		if (*cp++)
+			return 0;
+	}
+	return 1;
+}
+
 static int show_mountinfo(struct seq_file *m, void *v)
 {
 	struct proc_mounts *p = m->private;
@@ -1040,6 +1052,10 @@ static int show_mountinfo(struct seq_file *m, void *v)
 	if (IS_MNT_UNBINDABLE(mnt))
 		seq_puts(m, " unbindable");
 
+	if (!uuid_is_nil(mnt->mnt_sb->s_uuid))
+		/* print the uuid */
+		seq_printf(m, " uuid:%pU", mnt->mnt_sb->s_uuid);
+
 	/* Filesystem specific data */
 	seq_puts(m, " - ");
 	show_type(m, sb);
@@ -1244,7 +1260,7 @@ static int do_umount(struct vfsmount *mnt, int flags)
 		 */
 		br_write_lock(vfsmount_lock);
 		if (mnt_get_count(mnt) != 2) {
-			br_write_lock(vfsmount_lock);
+			br_write_unlock(vfsmount_lock);
 			return -EBUSY;
 		}
 		br_write_unlock(vfsmount_lock);
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 1cc600e77bb..2f8e61816d7 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -37,6 +37,7 @@
 #include <linux/inet.h>
 #include <linux/nfs_xdr.h>
 #include <linux/slab.h>
+#include <linux/compat.h>
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -89,7 +90,11 @@ int nfs_wait_bit_killable(void *word)
  */
 u64 nfs_compat_user_ino64(u64 fileid)
 {
-	int ino;
+#ifdef CONFIG_COMPAT
+	compat_ulong_t ino;
+#else	
+	unsigned long ino;
+#endif
 
 	if (enable_ino64)
 		return fileid;
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 7a747407314..1be36cf65bf 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -298,6 +298,11 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp);
 #if defined(CONFIG_NFS_V4_1)
 struct rpc_cred *nfs4_get_machine_cred_locked(struct nfs_client *clp);
 struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp);
+extern void nfs4_schedule_session_recovery(struct nfs4_session *);
+#else
+static inline void nfs4_schedule_session_recovery(struct nfs4_session *session)
+{
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *);
@@ -307,10 +312,9 @@ extern void nfs4_put_open_state(struct nfs4_state *);
 extern void nfs4_close_state(struct path *, struct nfs4_state *, fmode_t);
 extern void nfs4_close_sync(struct path *, struct nfs4_state *, fmode_t);
 extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t);
-extern void nfs4_schedule_state_recovery(struct nfs_client *);
+extern void nfs4_schedule_lease_recovery(struct nfs_client *);
 extern void nfs4_schedule_state_manager(struct nfs_client *);
-extern int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state);
-extern int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state);
+extern void nfs4_schedule_stateid_recovery(const struct nfs_server *, struct nfs4_state *);
 extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags);
 extern void nfs41_handle_recall_slot(struct nfs_client *clp);
 extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c
index f5c9b125e8c..b73c34375f6 100644
--- a/fs/nfs/nfs4filelayoutdev.c
+++ b/fs/nfs/nfs4filelayoutdev.c
@@ -219,6 +219,10 @@ decode_and_add_ds(__be32 **pp, struct inode *inode)
 		goto out_err;
 	}
 	buf = kmalloc(rlen + 1, GFP_KERNEL);
+	if (!buf) {
+		dprintk("%s: Not enough memory\n", __func__);
+		goto out_err;
+	}
 	buf[rlen] = '\0';
 	memcpy(buf, r_addr, rlen);
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 78936a8f40a..0a07e353a96 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -256,12 +256,13 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 		case -NFS4ERR_OPENMODE:
 			if (state == NULL)
 				break;
-			nfs4_state_mark_reclaim_nograce(clp, state);
-			goto do_state_recovery;
+			nfs4_schedule_stateid_recovery(server, state);
+			goto wait_on_recovery;
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_STALE_CLIENTID:
 		case -NFS4ERR_EXPIRED:
-			goto do_state_recovery;
+			nfs4_schedule_lease_recovery(clp);
+			goto wait_on_recovery;
 #if defined(CONFIG_NFS_V4_1)
 		case -NFS4ERR_BADSESSION:
 		case -NFS4ERR_BADSLOT:
@@ -272,7 +273,7 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 		case -NFS4ERR_SEQ_MISORDERED:
 			dprintk("%s ERROR: %d Reset session\n", __func__,
 				errorcode);
-			nfs4_schedule_state_recovery(clp);
+			nfs4_schedule_session_recovery(clp->cl_session);
 			exception->retry = 1;
 			break;
 #endif /* defined(CONFIG_NFS_V4_1) */
@@ -295,8 +296,7 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 	}
 	/* We failed to handle the error */
 	return nfs4_map_errors(ret);
-do_state_recovery:
-	nfs4_schedule_state_recovery(clp);
+wait_on_recovery:
 	ret = nfs4_wait_clnt_recover(clp);
 	if (ret == 0)
 		exception->retry = 1;
@@ -435,8 +435,8 @@ static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *
 		clp = res->sr_session->clp;
 		do_renew_lease(clp, timestamp);
 		/* Check sequence flags */
-		if (atomic_read(&clp->cl_count) > 1)
-			nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags);
+		if (res->sr_status_flags != 0)
+			nfs4_schedule_lease_recovery(clp);
 		break;
 	case -NFS4ERR_DELAY:
 		/* The server detected a resend of the RPC call and
@@ -1255,14 +1255,13 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state
 			case -NFS4ERR_BAD_HIGH_SLOT:
 			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 			case -NFS4ERR_DEADSESSION:
-				nfs4_schedule_state_recovery(
-					server->nfs_client);
+				nfs4_schedule_session_recovery(server->nfs_client->cl_session);
 				goto out;
 			case -NFS4ERR_STALE_CLIENTID:
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
 				/* Don't recall a delegation if it was lost */
-				nfs4_schedule_state_recovery(server->nfs_client);
+				nfs4_schedule_lease_recovery(server->nfs_client);
 				goto out;
 			case -ERESTARTSYS:
 				/*
@@ -1271,7 +1270,7 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state
 				 */
 			case -NFS4ERR_ADMIN_REVOKED:
 			case -NFS4ERR_BAD_STATEID:
-				nfs4_state_mark_reclaim_nograce(server->nfs_client, state);
+				nfs4_schedule_stateid_recovery(server, state);
 			case -EKEYEXPIRED:
 				/*
 				 * User RPCSEC_GSS context has expired.
@@ -1587,7 +1586,7 @@ static int nfs4_recover_expired_lease(struct nfs_server *server)
 		if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) &&
 		    !test_bit(NFS4CLNT_CHECK_LEASE,&clp->cl_state))
 			break;
-		nfs4_schedule_state_recovery(clp);
+		nfs4_schedule_state_manager(clp);
 		ret = -EIO;
 	}
 	return ret;
@@ -3178,7 +3177,7 @@ static void nfs4_renew_done(struct rpc_task *task, void *calldata)
 	if (task->tk_status < 0) {
 		/* Unless we're shutting down, schedule state recovery! */
 		if (test_bit(NFS_CS_RENEWD, &clp->cl_res_state) != 0)
-			nfs4_schedule_state_recovery(clp);
+			nfs4_schedule_lease_recovery(clp);
 		return;
 	}
 	do_renew_lease(clp, timestamp);
@@ -3252,6 +3251,35 @@ static void buf_to_pages(const void *buf, size_t buflen,
 	}
 }
 
+static int buf_to_pages_noslab(const void *buf, size_t buflen,
+		struct page **pages, unsigned int *pgbase)
+{
+	struct page *newpage, **spages;
+	int rc = 0;
+	size_t len;
+	spages = pages;
+
+	do {
+		len = min_t(size_t, PAGE_CACHE_SIZE, buflen);
+		newpage = alloc_page(GFP_KERNEL);
+
+		if (newpage == NULL)
+			goto unwind;
+		memcpy(page_address(newpage), buf, len);
+                buf += len;
+                buflen -= len;
+		*pages++ = newpage;
+		rc++;
+	} while (buflen != 0);
+
+	return rc;
+
+unwind:
+	for(; rc > 0; rc--)
+		__free_page(spages[rc-1]);
+	return -ENOMEM;
+}
+
 struct nfs4_cached_acl {
 	int cached;
 	size_t len;
@@ -3420,13 +3448,23 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl
 		.rpc_argp	= &arg,
 		.rpc_resp	= &res,
 	};
-	int ret;
+	int ret, i;
 
 	if (!nfs4_server_supports_acls(server))
 		return -EOPNOTSUPP;
+	i = buf_to_pages_noslab(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
+	if (i < 0)
+		return i;
 	nfs_inode_return_delegation(inode);
-	buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase);
 	ret = nfs4_call_sync(server, &msg, &arg, &res, 1);
+
+	/*
+	 * Free each page after tx, so the only ref left is
+	 * held by the network stack
+	 */
+	for (; i > 0; i--)
+		put_page(pages[i-1]);
+
 	/*
 	 * Acl update can result in inode attribute update.
 	 * so mark the attribute cache invalid.
@@ -3464,12 +3502,13 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 		case -NFS4ERR_OPENMODE:
 			if (state == NULL)
 				break;
-			nfs4_state_mark_reclaim_nograce(clp, state);
-			goto do_state_recovery;
+			nfs4_schedule_stateid_recovery(server, state);
+			goto wait_on_recovery;
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_STALE_CLIENTID:
 		case -NFS4ERR_EXPIRED:
-			goto do_state_recovery;
+			nfs4_schedule_lease_recovery(clp);
+			goto wait_on_recovery;
 #if defined(CONFIG_NFS_V4_1)
 		case -NFS4ERR_BADSESSION:
 		case -NFS4ERR_BADSLOT:
@@ -3480,7 +3519,7 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 		case -NFS4ERR_SEQ_MISORDERED:
 			dprintk("%s ERROR %d, Reset session\n", __func__,
 				task->tk_status);
-			nfs4_schedule_state_recovery(clp);
+			nfs4_schedule_session_recovery(clp->cl_session);
 			task->tk_status = 0;
 			return -EAGAIN;
 #endif /* CONFIG_NFS_V4_1 */
@@ -3497,9 +3536,8 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 	}
 	task->tk_status = nfs4_map_errors(task->tk_status);
 	return 0;
-do_state_recovery:
+wait_on_recovery:
 	rpc_sleep_on(&clp->cl_rpcwaitq, task, NULL);
-	nfs4_schedule_state_recovery(clp);
 	if (test_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) == 0)
 		rpc_wake_up_queued_task(&clp->cl_rpcwaitq, task);
 	task->tk_status = 0;
@@ -4110,7 +4148,7 @@ static void nfs4_lock_release(void *calldata)
 		task = nfs4_do_unlck(&data->fl, data->ctx, data->lsp,
 				data->arg.lock_seqid);
 		if (!IS_ERR(task))
-			rpc_put_task(task);
+			rpc_put_task_async(task);
 		dprintk("%s: cancelling lock!\n", __func__);
 	} else
 		nfs_free_seqid(data->arg.lock_seqid);
@@ -4134,23 +4172,18 @@ static const struct rpc_call_ops nfs4_recover_lock_ops = {
 
 static void nfs4_handle_setlk_error(struct nfs_server *server, struct nfs4_lock_state *lsp, int new_lock_owner, int error)
 {
-	struct nfs_client *clp = server->nfs_client;
-	struct nfs4_state *state = lsp->ls_state;
-
 	switch (error) {
 	case -NFS4ERR_ADMIN_REVOKED:
 	case -NFS4ERR_BAD_STATEID:
-	case -NFS4ERR_EXPIRED:
+		lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED;
 		if (new_lock_owner != 0 ||
 		   (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0)
-			nfs4_state_mark_reclaim_nograce(clp, state);
-		lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED;
+			nfs4_schedule_stateid_recovery(server, lsp->ls_state);
 		break;
 	case -NFS4ERR_STALE_STATEID:
-		if (new_lock_owner != 0 ||
-		    (lsp->ls_flags & NFS_LOCK_INITIALIZED) != 0)
-			nfs4_state_mark_reclaim_reboot(clp, state);
 		lsp->ls_seqid.flags &= ~NFS_SEQID_CONFIRMED;
+	case -NFS4ERR_EXPIRED:
+		nfs4_schedule_lease_recovery(server->nfs_client);
 	};
 }
 
@@ -4366,12 +4399,14 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
 			case -NFS4ERR_EXPIRED:
 			case -NFS4ERR_STALE_CLIENTID:
 			case -NFS4ERR_STALE_STATEID:
+				nfs4_schedule_lease_recovery(server->nfs_client);
+				goto out;
 			case -NFS4ERR_BADSESSION:
 			case -NFS4ERR_BADSLOT:
 			case -NFS4ERR_BAD_HIGH_SLOT:
 			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 			case -NFS4ERR_DEADSESSION:
-				nfs4_schedule_state_recovery(server->nfs_client);
+				nfs4_schedule_session_recovery(server->nfs_client->cl_session);
 				goto out;
 			case -ERESTARTSYS:
 				/*
@@ -4381,7 +4416,7 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
 			case -NFS4ERR_ADMIN_REVOKED:
 			case -NFS4ERR_BAD_STATEID:
 			case -NFS4ERR_OPENMODE:
-				nfs4_state_mark_reclaim_nograce(server->nfs_client, state);
+				nfs4_schedule_stateid_recovery(server, state);
 				err = 0;
 				goto out;
 			case -EKEYEXPIRED:
@@ -4988,10 +5023,20 @@ int nfs4_proc_create_session(struct nfs_client *clp)
 	int status;
 	unsigned *ptr;
 	struct nfs4_session *session = clp->cl_session;
+	long timeout = 0;
+	int err;
 
 	dprintk("--> %s clp=%p session=%p\n", __func__, clp, session);
 
-	status = _nfs4_proc_create_session(clp);
+	do {
+		status = _nfs4_proc_create_session(clp);
+		if (status == -NFS4ERR_DELAY) {
+			err = nfs4_delay(clp->cl_rpcclient, &timeout);
+			if (err)
+				status = err;
+		}
+	} while (status == -NFS4ERR_DELAY);
+
 	if (status)
 		goto out;
 
@@ -5100,7 +5145,7 @@ static int nfs41_sequence_handle_errors(struct rpc_task *task, struct nfs_client
 		rpc_delay(task, NFS4_POLL_RETRY_MAX);
 		return -EAGAIN;
 	default:
-		nfs4_schedule_state_recovery(clp);
+		nfs4_schedule_lease_recovery(clp);
 	}
 	return 0;
 }
@@ -5187,7 +5232,7 @@ static int nfs41_proc_async_sequence(struct nfs_client *clp, struct rpc_cred *cr
 	if (IS_ERR(task))
 		ret = PTR_ERR(task);
 	else
-		rpc_put_task(task);
+		rpc_put_task_async(task);
 	dprintk("<-- %s status=%d\n", __func__, ret);
 	return ret;
 }
@@ -5203,8 +5248,13 @@ static int nfs4_proc_sequence(struct nfs_client *clp, struct rpc_cred *cred)
 		goto out;
 	}
 	ret = rpc_wait_for_completion_task(task);
-	if (!ret)
+	if (!ret) {
+		struct nfs4_sequence_res *res = task->tk_msg.rpc_resp;
+
+		if (task->tk_status == 0)
+			nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags);
 		ret = task->tk_status;
+	}
 	rpc_put_task(task);
 out:
 	dprintk("<-- %s status=%d\n", __func__, ret);
@@ -5241,7 +5291,7 @@ static int nfs41_reclaim_complete_handle_errors(struct rpc_task *task, struct nf
 		rpc_delay(task, NFS4_POLL_RETRY_MAX);
 		return -EAGAIN;
 	default:
-		nfs4_schedule_state_recovery(clp);
+		nfs4_schedule_lease_recovery(clp);
 	}
 	return 0;
 }
@@ -5309,6 +5359,9 @@ static int nfs41_proc_reclaim_complete(struct nfs_client *clp)
 		status = PTR_ERR(task);
 		goto out;
 	}
+	status = nfs4_wait_for_completion_rpc_task(task);
+	if (status == 0)
+		status = task->tk_status;
 	rpc_put_task(task);
 	return 0;
 out:
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index e6742b57a04..0592288f9f0 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -1007,9 +1007,9 @@ void nfs4_schedule_state_manager(struct nfs_client *clp)
 }
 
 /*
- * Schedule a state recovery attempt
+ * Schedule a lease recovery attempt
  */
-void nfs4_schedule_state_recovery(struct nfs_client *clp)
+void nfs4_schedule_lease_recovery(struct nfs_client *clp)
 {
 	if (!clp)
 		return;
@@ -1018,7 +1018,7 @@ void nfs4_schedule_state_recovery(struct nfs_client *clp)
 	nfs4_schedule_state_manager(clp);
 }
 
-int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state)
+static int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *state)
 {
 
 	set_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags);
@@ -1032,7 +1032,7 @@ int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_state *st
 	return 1;
 }
 
-int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state)
+static int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state)
 {
 	set_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags);
 	clear_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags);
@@ -1041,6 +1041,14 @@ int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *s
 	return 1;
 }
 
+void nfs4_schedule_stateid_recovery(const struct nfs_server *server, struct nfs4_state *state)
+{
+	struct nfs_client *clp = server->nfs_client;
+
+	nfs4_state_mark_reclaim_nograce(clp, state);
+	nfs4_schedule_state_manager(clp);
+}
+
 static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_recovery_ops *ops)
 {
 	struct inode *inode = state->inode;
@@ -1436,10 +1444,15 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 }
 
 #ifdef CONFIG_NFS_V4_1
+void nfs4_schedule_session_recovery(struct nfs4_session *session)
+{
+	nfs4_schedule_lease_recovery(session->clp);
+}
+
 void nfs41_handle_recall_slot(struct nfs_client *clp)
 {
 	set_bit(NFS4CLNT_RECALL_SLOT, &clp->cl_state);
-	nfs4_schedule_state_recovery(clp);
+	nfs4_schedule_state_manager(clp);
 }
 
 static void nfs4_reset_all_state(struct nfs_client *clp)
@@ -1447,7 +1460,7 @@ static void nfs4_reset_all_state(struct nfs_client *clp)
 	if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) {
 		clp->cl_boot_time = CURRENT_TIME;
 		nfs4_state_start_reclaim_nograce(clp);
-		nfs4_schedule_state_recovery(clp);
+		nfs4_schedule_state_manager(clp);
 	}
 }
 
@@ -1455,7 +1468,7 @@ static void nfs41_handle_server_reboot(struct nfs_client *clp)
 {
 	if (test_and_set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) == 0) {
 		nfs4_state_start_reclaim_reboot(clp);
-		nfs4_schedule_state_recovery(clp);
+		nfs4_schedule_state_manager(clp);
 	}
 }
 
@@ -1475,7 +1488,7 @@ static void nfs41_handle_cb_path_down(struct nfs_client *clp)
 {
 	nfs_expire_all_delegations(clp);
 	if (test_and_set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state) == 0)
-		nfs4_schedule_state_recovery(clp);
+		nfs4_schedule_state_manager(clp);
 }
 
 void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags)
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 4e2c168b6ee..94d50e86a12 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1660,7 +1660,7 @@ static void encode_create_session(struct xdr_stream *xdr,
 
 	p = reserve_space(xdr, 20 + 2*28 + 20 + len + 12);
 	*p++ = cpu_to_be32(OP_CREATE_SESSION);
-	p = xdr_encode_hyper(p, clp->cl_ex_clid);
+	p = xdr_encode_hyper(p, clp->cl_clientid);
 	*p++ = cpu_to_be32(clp->cl_seqid);			/*Sequence id */
 	*p++ = cpu_to_be32(args->flags);			/*flags */
 
@@ -4694,7 +4694,7 @@ static int decode_exchange_id(struct xdr_stream *xdr,
 	p = xdr_inline_decode(xdr, 8);
 	if (unlikely(!p))
 		goto out_overflow;
-	xdr_decode_hyper(p, &clp->cl_ex_clid);
+	xdr_decode_hyper(p, &clp->cl_clientid);
 	p = xdr_inline_decode(xdr, 12);
 	if (unlikely(!p))
 		goto out_overflow;
diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index 903908a2002..c541093a5bf 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -86,11 +86,14 @@
 /* Default path we try to mount. "%s" gets replaced by our IP address */
 #define NFS_ROOT		"/tftpboot/%s"
 
+/* Default NFSROOT mount options. */
+#define NFS_DEF_OPTIONS		"udp"
+
 /* Parameters passed from the kernel command line */
 static char nfs_root_parms[256] __initdata = "";
 
 /* Text-based mount options passed to super.c */
-static char nfs_root_options[256] __initdata = "";
+static char nfs_root_options[256] __initdata = NFS_DEF_OPTIONS;
 
 /* Address of NFS server */
 static __be32 servaddr __initdata = htonl(INADDR_NONE);
@@ -160,8 +163,14 @@ static int __init root_nfs_copy(char *dest, const char *src,
 }
 
 static int __init root_nfs_cat(char *dest, const char *src,
-				  const size_t destlen)
+			       const size_t destlen)
 {
+	size_t len = strlen(dest);
+
+	if (len && dest[len - 1] != ',')
+		if (strlcat(dest, ",", destlen) > destlen)
+			return -1;
+
 	if (strlcat(dest, src, destlen) > destlen)
 		return -1;
 	return 0;
@@ -194,16 +203,6 @@ static int __init root_nfs_parse_options(char *incoming, char *exppath,
 		if (root_nfs_cat(nfs_root_options, incoming,
 						sizeof(nfs_root_options)))
 			return -1;
-
-	/*
-	 * Possibly prepare for more options to be appended
-	 */
-	if (nfs_root_options[0] != '\0' &&
-	    nfs_root_options[strlen(nfs_root_options)] != ',')
-		if (root_nfs_cat(nfs_root_options, ",",
-						sizeof(nfs_root_options)))
-			return -1;
-
 	return 0;
 }
 
@@ -217,7 +216,7 @@ static int __init root_nfs_parse_options(char *incoming, char *exppath,
  */
 static int __init root_nfs_data(char *cmdline)
 {
-	char addr_option[sizeof("nolock,addr=") + INET_ADDRSTRLEN + 1];
+	char mand_options[sizeof("nolock,addr=") + INET_ADDRSTRLEN + 1];
 	int len, retval = -1;
 	char *tmp = NULL;
 	const size_t tmplen = sizeof(nfs_export_path);
@@ -244,9 +243,9 @@ static int __init root_nfs_data(char *cmdline)
 	 * Append mandatory options for nfsroot so they override
 	 * what has come before
 	 */
-	snprintf(addr_option, sizeof(addr_option), "nolock,addr=%pI4",
+	snprintf(mand_options, sizeof(mand_options), "nolock,addr=%pI4",
 			&servaddr);
-	if (root_nfs_cat(nfs_root_options, addr_option,
+	if (root_nfs_cat(nfs_root_options, mand_options,
 						sizeof(nfs_root_options)))
 		goto out_optionstoolong;
 
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index e313a51acdd..6481d537d69 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -180,7 +180,7 @@ static int nfs_do_call_unlink(struct dentry *parent, struct inode *dir, struct n
 	task_setup_data.rpc_client = NFS_CLIENT(dir);
 	task = rpc_run_task(&task_setup_data);
 	if (!IS_ERR(task))
-		rpc_put_task(task);
+		rpc_put_task_async(task);
 	return 1;
 }
 
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index c8278f4046c..42b92d7a9cc 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1292,6 +1292,8 @@ static int nfs_commit_rpcsetup(struct list_head *head,
 	task = rpc_run_task(&task_setup_data);
 	if (IS_ERR(task))
 		return PTR_ERR(task);
+	if (how & FLUSH_SYNC)
+		rpc_wait_for_completion_task(task);
 	rpc_put_task(task);
 	return 0;
 }
diff --git a/fs/nfsctl.c b/fs/nfsctl.c
index bf9cbd242dd..124e8fcb0dd 100644
--- a/fs/nfsctl.c
+++ b/fs/nfsctl.c
@@ -22,30 +22,17 @@
 
 static struct file *do_open(char *name, int flags)
 {
-	struct nameidata nd;
 	struct vfsmount *mnt;
-	int error;
+	struct file *file;
 
 	mnt = do_kern_mount("nfsd", 0, "nfsd", NULL);
 	if (IS_ERR(mnt))
 		return (struct file *)mnt;
 
-	error = vfs_path_lookup(mnt->mnt_root, mnt, name, 0, &nd);
-	mntput(mnt);	/* drop do_kern_mount reference */
-	if (error)
-		return ERR_PTR(error);
-
-	if (flags == O_RDWR)
-		error = may_open(&nd.path, MAY_READ|MAY_WRITE, flags);
-	else
-		error = may_open(&nd.path, MAY_WRITE, flags);
+	file = file_open_root(mnt->mnt_root, mnt, name, flags);
 
-	if (!error)
-		return dentry_open(nd.path.dentry, nd.path.mnt, flags,
-				   current_cred());
-
-	path_put(&nd.path);
-	return ERR_PTR(error);
+	mntput(mnt);	/* drop do_kern_mount reference */
+	return file;
 }
 
 static struct {
diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index cde36cb0f34..02eb4edf0ec 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -432,7 +432,7 @@ static int decode_cb_sequence4resok(struct xdr_stream *xdr,
 	 * If the server returns different values for sessionID, slotID or
 	 * sequence number, the server is looney tunes.
 	 */
-	p = xdr_inline_decode(xdr, NFS4_MAX_SESSIONID_LEN + 4 + 4);
+	p = xdr_inline_decode(xdr, NFS4_MAX_SESSIONID_LEN + 4 + 4 + 4 + 4);
 	if (unlikely(p == NULL))
 		goto out_overflow;
 	memcpy(id.data, p, NFS4_MAX_SESSIONID_LEN);
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 54b60bfceb8..7b566ec14e1 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -2445,15 +2445,16 @@ nfs4_check_delegmode(struct nfs4_delegation *dp, int flags)
 static struct nfs4_delegation *
 find_delegation_file(struct nfs4_file *fp, stateid_t *stid)
 {
-	struct nfs4_delegation *dp = NULL;
+	struct nfs4_delegation *dp;
 
 	spin_lock(&recall_lock);
-	list_for_each_entry(dp, &fp->fi_delegations, dl_perfile) {
-		if (dp->dl_stateid.si_stateownerid == stid->si_stateownerid)
-			break;
-	}
+	list_for_each_entry(dp, &fp->fi_delegations, dl_perfile)
+		if (dp->dl_stateid.si_stateownerid == stid->si_stateownerid) {
+			spin_unlock(&recall_lock);
+			return dp;
+		}
 	spin_unlock(&recall_lock);
-	return dp;
+	return NULL;
 }
 
 int share_access_to_flags(u32 share_access)
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 1275b865507..615f0a9f060 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -1142,7 +1142,7 @@ nfsd4_decode_create_session(struct nfsd4_compoundargs *argp,
 
 	u32 dummy;
 	char *machine_name;
-	int i;
+	int i, j;
 	int nr_secflavs;
 
 	READ_BUF(16);
@@ -1215,7 +1215,7 @@ nfsd4_decode_create_session(struct nfsd4_compoundargs *argp,
 			READ_BUF(4);
 			READ32(dummy);
 			READ_BUF(dummy * 4);
-			for (i = 0; i < dummy; ++i)
+			for (j = 0; j < dummy; ++j)
 				READ32(dummy);
 			break;
 		case RPC_AUTH_GSS:
diff --git a/fs/nilfs2/btnode.c b/fs/nilfs2/btnode.c
index 388e9e8f528..85f7baa15f5 100644
--- a/fs/nilfs2/btnode.c
+++ b/fs/nilfs2/btnode.c
@@ -35,11 +35,6 @@
 #include "btnode.h"
 
 
-void nilfs_btnode_cache_init_once(struct address_space *btnc)
-{
-	nilfs_mapping_init_once(btnc);
-}
-
 static const struct address_space_operations def_btnode_aops = {
 	.sync_page		= block_sync_page,
 };
diff --git a/fs/nilfs2/btnode.h b/fs/nilfs2/btnode.h
index 79037494f1e..1b8ebd888c2 100644
--- a/fs/nilfs2/btnode.h
+++ b/fs/nilfs2/btnode.h
@@ -37,7 +37,6 @@ struct nilfs_btnode_chkey_ctxt {
 	struct buffer_head *newbh;
 };
 
-void nilfs_btnode_cache_init_once(struct address_space *);
 void nilfs_btnode_cache_init(struct address_space *, struct backing_dev_info *);
 void nilfs_btnode_cache_clear(struct address_space *);
 struct buffer_head *nilfs_btnode_create_block(struct address_space *btnc,
diff --git a/fs/nilfs2/mdt.c b/fs/nilfs2/mdt.c
index 6a0e2a189f6..a0babd2bff6 100644
--- a/fs/nilfs2/mdt.c
+++ b/fs/nilfs2/mdt.c
@@ -454,9 +454,9 @@ int nilfs_mdt_setup_shadow_map(struct inode *inode,
 	struct backing_dev_info *bdi = inode->i_sb->s_bdi;
 
 	INIT_LIST_HEAD(&shadow->frozen_buffers);
-	nilfs_mapping_init_once(&shadow->frozen_data);
+	address_space_init_once(&shadow->frozen_data);
 	nilfs_mapping_init(&shadow->frozen_data, bdi, &shadow_map_aops);
-	nilfs_mapping_init_once(&shadow->frozen_btnodes);
+	address_space_init_once(&shadow->frozen_btnodes);
 	nilfs_mapping_init(&shadow->frozen_btnodes, bdi, &shadow_map_aops);
 	mi->mi_shadow = shadow;
 	return 0;
diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c
index 98034271cd0..161791d2645 100644
--- a/fs/nilfs2/namei.c
+++ b/fs/nilfs2/namei.c
@@ -397,7 +397,6 @@ static int nilfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		new_de = nilfs_find_entry(new_dir, &new_dentry->d_name, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inc_nlink(old_inode);
 		nilfs_set_link(new_dir, new_de, new_page, old_inode);
 		nilfs_mark_inode_dirty(new_dir);
 		new_inode->i_ctime = CURRENT_TIME;
@@ -411,13 +410,9 @@ static int nilfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			if (new_dir->i_nlink >= NILFS_LINK_MAX)
 				goto out_dir;
 		}
-		inc_nlink(old_inode);
 		err = nilfs_add_link(new_dentry, old_inode);
-		if (err) {
-			drop_nlink(old_inode);
-			nilfs_mark_inode_dirty(old_inode);
+		if (err)
 			goto out_dir;
-		}
 		if (dir_de) {
 			inc_nlink(new_dir);
 			nilfs_mark_inode_dirty(new_dir);
@@ -431,7 +426,6 @@ static int nilfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	old_inode->i_ctime = CURRENT_TIME;
 
 	nilfs_delete_entry(old_de, old_page);
-	drop_nlink(old_inode);
 
 	if (dir_de) {
 		nilfs_set_link(old_inode, dir_de, dir_page, new_dir);
diff --git a/fs/nilfs2/page.c b/fs/nilfs2/page.c
index 0c432416cfe..a585b35fd6b 100644
--- a/fs/nilfs2/page.c
+++ b/fs/nilfs2/page.c
@@ -492,19 +492,6 @@ unsigned nilfs_page_count_clean_buffers(struct page *page,
 	return nc;
 }
 
-void nilfs_mapping_init_once(struct address_space *mapping)
-{
-	memset(mapping, 0, sizeof(*mapping));
-	INIT_RADIX_TREE(&mapping->page_tree, GFP_ATOMIC);
-	spin_lock_init(&mapping->tree_lock);
-	INIT_LIST_HEAD(&mapping->private_list);
-	spin_lock_init(&mapping->private_lock);
-
-	spin_lock_init(&mapping->i_mmap_lock);
-	INIT_RAW_PRIO_TREE_ROOT(&mapping->i_mmap);
-	INIT_LIST_HEAD(&mapping->i_mmap_nonlinear);
-}
-
 void nilfs_mapping_init(struct address_space *mapping,
 			struct backing_dev_info *bdi,
 			const struct address_space_operations *aops)
diff --git a/fs/nilfs2/page.h b/fs/nilfs2/page.h
index 622df27cd89..2a00953ebd5 100644
--- a/fs/nilfs2/page.h
+++ b/fs/nilfs2/page.h
@@ -61,7 +61,6 @@ void nilfs_free_private_page(struct page *);
 int nilfs_copy_dirty_pages(struct address_space *, struct address_space *);
 void nilfs_copy_back_pages(struct address_space *, struct address_space *);
 void nilfs_clear_dirty_pages(struct address_space *);
-void nilfs_mapping_init_once(struct address_space *mapping);
 void nilfs_mapping_init(struct address_space *mapping,
 			struct backing_dev_info *bdi,
 			const struct address_space_operations *aops);
diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c
index 55ebae5c7f3..2de9f636792 100644
--- a/fs/nilfs2/segment.c
+++ b/fs/nilfs2/segment.c
@@ -430,7 +430,8 @@ static void nilfs_segctor_begin_finfo(struct nilfs_sc_info *sci,
 	nilfs_segctor_map_segsum_entry(
 		sci, &sci->sc_binfo_ptr, sizeof(struct nilfs_finfo));
 
-	if (inode->i_sb && !test_bit(NILFS_SC_HAVE_DELTA, &sci->sc_flags))
+	if (NILFS_I(inode)->i_root &&
+	    !test_bit(NILFS_SC_HAVE_DELTA, &sci->sc_flags))
 		set_bit(NILFS_SC_HAVE_DELTA, &sci->sc_flags);
 	/* skip finfo */
 }
diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c
index 58fd707174e..1673b3d9984 100644
--- a/fs/nilfs2/super.c
+++ b/fs/nilfs2/super.c
@@ -1279,7 +1279,7 @@ static void nilfs_inode_init_once(void *obj)
 #ifdef CONFIG_NILFS_XATTR
 	init_rwsem(&ii->xattr_sem);
 #endif
-	nilfs_btnode_cache_init_once(&ii->i_btnode_cache);
+	address_space_init_once(&ii->i_btnode_cache);
 	ii->i_bmap = &ii->i_bmap_data;
 	inode_init_once(&ii->vfs_inode);
 }
diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c
index 704f6b1742f..69786e3eb43 100644
--- a/fs/ocfs2/acl.c
+++ b/fs/ocfs2/acl.c
@@ -24,7 +24,6 @@
 #include <linux/slab.h>
 #include <linux/string.h>
 
-#define MLOG_MASK_PREFIX ML_INODE
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index e4984e259cb..b27a0d86f8c 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -30,7 +30,6 @@
 #include <linux/swap.h>
 #include <linux/quotaops.h>
 
-#define MLOG_MASK_PREFIX ML_DISK_ALLOC
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -50,6 +49,7 @@
 #include "uptodate.h"
 #include "xattr.h"
 #include "refcounttree.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -886,8 +886,7 @@ static int ocfs2_validate_extent_block(struct super_block *sb,
 	struct ocfs2_extent_block *eb =
 		(struct ocfs2_extent_block *)bh->b_data;
 
-	mlog(0, "Validating extent block %llu\n",
-	     (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_validate_extent_block((unsigned long long)bh->b_blocknr);
 
 	BUG_ON(!buffer_uptodate(bh));
 
@@ -965,8 +964,6 @@ int ocfs2_num_free_extents(struct ocfs2_super *osb,
 	struct buffer_head *eb_bh = NULL;
 	u64 last_eb_blk = 0;
 
-	mlog_entry_void();
-
 	el = et->et_root_el;
 	last_eb_blk = ocfs2_et_get_last_eb_blk(et);
 
@@ -987,7 +984,7 @@ int ocfs2_num_free_extents(struct ocfs2_super *osb,
 bail:
 	brelse(eb_bh);
 
-	mlog_exit(retval);
+	trace_ocfs2_num_free_extents(retval);
 	return retval;
 }
 
@@ -1010,8 +1007,6 @@ static int ocfs2_create_new_meta_bhs(handle_t *handle,
 		OCFS2_SB(ocfs2_metadata_cache_get_super(et->et_ci));
 	struct ocfs2_extent_block *eb;
 
-	mlog_entry_void();
-
 	count = 0;
 	while (count < wanted) {
 		status = ocfs2_claim_metadata(handle,
@@ -1074,8 +1069,8 @@ bail:
 			brelse(bhs[i]);
 			bhs[i] = NULL;
 		}
+		mlog_errno(status);
 	}
-	mlog_exit(status);
 	return status;
 }
 
@@ -1173,8 +1168,6 @@ static int ocfs2_add_branch(handle_t *handle,
 	struct ocfs2_extent_list  *el;
 	u32 new_cpos, root_end;
 
-	mlog_entry_void();
-
 	BUG_ON(!last_eb_bh || !*last_eb_bh);
 
 	if (eb_bh) {
@@ -1200,8 +1193,11 @@ static int ocfs2_add_branch(handle_t *handle,
 	 * from new_cpos).
 	 */
 	if (root_end > new_cpos) {
-		mlog(0, "adjust the cluster end from %u to %u\n",
-		     root_end, new_cpos);
+		trace_ocfs2_adjust_rightmost_branch(
+			(unsigned long long)
+			ocfs2_metadata_cache_owner(et->et_ci),
+			root_end, new_cpos);
+
 		status = ocfs2_adjust_rightmost_branch(handle, et);
 		if (status) {
 			mlog_errno(status);
@@ -1332,7 +1328,6 @@ bail:
 		kfree(new_eb_bhs);
 	}
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1353,8 +1348,6 @@ static int ocfs2_shift_tree_depth(handle_t *handle,
 	struct ocfs2_extent_list  *root_el;
 	struct ocfs2_extent_list  *eb_el;
 
-	mlog_entry_void();
-
 	status = ocfs2_create_new_meta_bhs(handle, et, 1, meta_ac,
 					   &new_eb_bh);
 	if (status < 0) {
@@ -1415,7 +1408,6 @@ static int ocfs2_shift_tree_depth(handle_t *handle,
 bail:
 	brelse(new_eb_bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1446,8 +1438,6 @@ static int ocfs2_find_branch_target(struct ocfs2_extent_tree *et,
 	struct buffer_head *bh = NULL;
 	struct buffer_head *lowest_bh = NULL;
 
-	mlog_entry_void();
-
 	*target_bh = NULL;
 
 	el = et->et_root_el;
@@ -1503,7 +1493,6 @@ static int ocfs2_find_branch_target(struct ocfs2_extent_tree *et,
 bail:
 	brelse(bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1540,7 +1529,10 @@ static int ocfs2_grow_tree(handle_t *handle, struct ocfs2_extent_tree *et,
 	 * another tree level */
 	if (shift) {
 		BUG_ON(bh);
-		mlog(0, "need to shift tree depth (current = %d)\n", depth);
+		trace_ocfs2_grow_tree(
+			(unsigned long long)
+			ocfs2_metadata_cache_owner(et->et_ci),
+			depth);
 
 		/* ocfs2_shift_tree_depth will return us a buffer with
 		 * the new extent block (so we can pass that to
@@ -1570,7 +1562,6 @@ static int ocfs2_grow_tree(handle_t *handle, struct ocfs2_extent_tree *et,
 
 	/* call ocfs2_add_branch to add the final part of the tree with
 	 * the new data. */
-	mlog(0, "add branch. bh = %p\n", bh);
 	ret = ocfs2_add_branch(handle, et, bh, last_eb_bh,
 			       meta_ac);
 	if (ret < 0) {
@@ -1645,8 +1636,9 @@ static void ocfs2_rotate_leaf(struct ocfs2_extent_list *el,
 	}
 	insert_index = i;
 
-	mlog(0, "ins %u: index %d, has_empty %d, next_free %d, count %d\n",
-	     insert_cpos, insert_index, has_empty, next_free, le16_to_cpu(el->l_count));
+	trace_ocfs2_rotate_leaf(insert_cpos, insert_index,
+				has_empty, next_free,
+				le16_to_cpu(el->l_count));
 
 	BUG_ON(insert_index < 0);
 	BUG_ON(insert_index >= le16_to_cpu(el->l_count));
@@ -2059,7 +2051,7 @@ static void ocfs2_complete_edge_insert(handle_t *handle,
 	left_el = path_leaf_el(left_path);
 	right_el = path_leaf_el(right_path);
 	for(i = left_path->p_tree_depth - 1; i > subtree_index; i--) {
-		mlog(0, "Adjust records at index %u\n", i);
+		trace_ocfs2_complete_edge_insert(i);
 
 		/*
 		 * One nice property of knowing that all of these
@@ -2389,7 +2381,9 @@ static int ocfs2_rotate_tree_right(handle_t *handle,
 		goto out;
 	}
 
-	mlog(0, "Insert: %u, first left path cpos: %u\n", insert_cpos, cpos);
+	trace_ocfs2_rotate_tree_right(
+		(unsigned long long)ocfs2_metadata_cache_owner(et->et_ci),
+		insert_cpos, cpos);
 
 	/*
 	 * What we want to do here is:
@@ -2418,8 +2412,10 @@ static int ocfs2_rotate_tree_right(handle_t *handle,
 	 * rotating subtrees.
 	 */
 	while (cpos && insert_cpos <= cpos) {
-		mlog(0, "Rotating a tree: ins. cpos: %u, left path cpos: %u\n",
-		     insert_cpos, cpos);
+		trace_ocfs2_rotate_tree_right(
+			(unsigned long long)
+			ocfs2_metadata_cache_owner(et->et_ci),
+			insert_cpos, cpos);
 
 		ret = ocfs2_find_path(et->et_ci, left_path, cpos);
 		if (ret) {
@@ -2461,10 +2457,10 @@ static int ocfs2_rotate_tree_right(handle_t *handle,
 
 		start = ocfs2_find_subtree_root(et, left_path, right_path);
 
-		mlog(0, "Subtree root at index %d (blk %llu, depth %d)\n",
-		     start,
-		     (unsigned long long) right_path->p_node[start].bh->b_blocknr,
-		     right_path->p_tree_depth);
+		trace_ocfs2_rotate_subtree(start,
+			(unsigned long long)
+			right_path->p_node[start].bh->b_blocknr,
+			right_path->p_tree_depth);
 
 		ret = ocfs2_extend_rotate_transaction(handle, start,
 						      orig_credits, right_path);
@@ -2964,8 +2960,7 @@ static int __ocfs2_rotate_tree_left(handle_t *handle,
 		subtree_root = ocfs2_find_subtree_root(et, left_path,
 						       right_path);
 
-		mlog(0, "Subtree root at index %d (blk %llu, depth %d)\n",
-		     subtree_root,
+		trace_ocfs2_rotate_subtree(subtree_root,
 		     (unsigned long long)
 		     right_path->p_node[subtree_root].bh->b_blocknr,
 		     right_path->p_tree_depth);
@@ -3989,9 +3984,11 @@ static int ocfs2_append_rec_to_path(handle_t *handle,
 			goto out;
 		}
 
-		mlog(0, "Append may need a left path update. cpos: %u, "
-		     "left_cpos: %u\n", le32_to_cpu(insert_rec->e_cpos),
-		     left_cpos);
+		trace_ocfs2_append_rec_to_path(
+			(unsigned long long)
+			ocfs2_metadata_cache_owner(et->et_ci),
+			le32_to_cpu(insert_rec->e_cpos),
+			left_cpos);
 
 		/*
 		 * No need to worry if the append is already in the
@@ -4562,7 +4559,7 @@ static int ocfs2_figure_insert_type(struct ocfs2_extent_tree *et,
 					      ocfs2_et_get_last_eb_blk(et),
 					      &bh);
 		if (ret) {
-			mlog_exit(ret);
+			mlog_errno(ret);
 			goto out;
 		}
 		eb = (struct ocfs2_extent_block *) bh->b_data;
@@ -4678,9 +4675,9 @@ int ocfs2_insert_extent(handle_t *handle,
 	struct ocfs2_insert_type insert = {0, };
 	struct ocfs2_extent_rec rec;
 
-	mlog(0, "add %u clusters at position %u to owner %llu\n",
-	     new_clusters, cpos,
-	     (unsigned long long)ocfs2_metadata_cache_owner(et->et_ci));
+	trace_ocfs2_insert_extent_start(
+		(unsigned long long)ocfs2_metadata_cache_owner(et->et_ci),
+		cpos, new_clusters);
 
 	memset(&rec, 0, sizeof(rec));
 	rec.e_cpos = cpu_to_le32(cpos);
@@ -4700,11 +4697,9 @@ int ocfs2_insert_extent(handle_t *handle,
 		goto bail;
 	}
 
-	mlog(0, "Insert.appending: %u, Insert.Contig: %u, "
-	     "Insert.contig_index: %d, Insert.free_records: %d, "
-	     "Insert.tree_depth: %d\n",
-	     insert.ins_appending, insert.ins_contig, insert.ins_contig_index,
-	     free_records, insert.ins_tree_depth);
+	trace_ocfs2_insert_extent(insert.ins_appending, insert.ins_contig,
+				  insert.ins_contig_index, free_records,
+				  insert.ins_tree_depth);
 
 	if (insert.ins_contig == CONTIG_NONE && free_records == 0) {
 		status = ocfs2_grow_tree(handle, et,
@@ -4726,7 +4721,6 @@ int ocfs2_insert_extent(handle_t *handle,
 bail:
 	brelse(last_eb_bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -4746,7 +4740,7 @@ int ocfs2_add_clusters_in_btree(handle_t *handle,
 				struct ocfs2_alloc_context *meta_ac,
 				enum ocfs2_alloc_restarted *reason_ret)
 {
-	int status = 0;
+	int status = 0, err = 0;
 	int free_extents;
 	enum ocfs2_alloc_restarted reason = RESTART_NONE;
 	u32 bit_off, num_bits;
@@ -4773,14 +4767,14 @@ int ocfs2_add_clusters_in_btree(handle_t *handle,
 	 * 2) we are so fragmented, we've needed to add metadata too
 	 *    many times. */
 	if (!free_extents && !meta_ac) {
-		mlog(0, "we haven't reserved any metadata!\n");
+		err = -1;
 		status = -EAGAIN;
 		reason = RESTART_META;
 		goto leave;
 	} else if ((!free_extents)
 		   && (ocfs2_alloc_context_bits_left(meta_ac)
 		       < ocfs2_extend_meta_needed(et->et_root_el))) {
-		mlog(0, "filesystem is really fragmented...\n");
+		err = -2;
 		status = -EAGAIN;
 		reason = RESTART_META;
 		goto leave;
@@ -4805,9 +4799,9 @@ int ocfs2_add_clusters_in_btree(handle_t *handle,
 	}
 
 	block = ocfs2_clusters_to_blocks(osb->sb, bit_off);
-	mlog(0, "Allocating %u clusters at block %u for owner %llu\n",
-	     num_bits, bit_off,
-	     (unsigned long long)ocfs2_metadata_cache_owner(et->et_ci));
+	trace_ocfs2_add_clusters_in_btree(
+	     (unsigned long long)ocfs2_metadata_cache_owner(et->et_ci),
+	     bit_off, num_bits);
 	status = ocfs2_insert_extent(handle, et, *logical_offset, block,
 				     num_bits, flags, meta_ac);
 	if (status < 0) {
@@ -4821,16 +4815,15 @@ int ocfs2_add_clusters_in_btree(handle_t *handle,
 	*logical_offset += num_bits;
 
 	if (clusters_to_add) {
-		mlog(0, "need to alloc once more, wanted = %u\n",
-		     clusters_to_add);
+		err = clusters_to_add;
 		status = -EAGAIN;
 		reason = RESTART_TRANS;
 	}
 
 leave:
-	mlog_exit(status);
 	if (reason_ret)
 		*reason_ret = reason;
+	trace_ocfs2_add_clusters_in_btree_ret(status, reason, err);
 	return status;
 }
 
@@ -5039,7 +5032,7 @@ int ocfs2_split_extent(handle_t *handle,
 					      ocfs2_et_get_last_eb_blk(et),
 					      &last_eb_bh);
 		if (ret) {
-			mlog_exit(ret);
+			mlog_errno(ret);
 			goto out;
 		}
 
@@ -5056,9 +5049,9 @@ int ocfs2_split_extent(handle_t *handle,
 
 	ctxt.c_has_empty_extent = ocfs2_is_empty_extent(&el->l_recs[0]);
 
-	mlog(0, "index: %d, contig: %u, has_empty: %u, split_covers: %u\n",
-	     split_index, ctxt.c_contig_type, ctxt.c_has_empty_extent,
-	     ctxt.c_split_covers_rec);
+	trace_ocfs2_split_extent(split_index, ctxt.c_contig_type,
+				 ctxt.c_has_empty_extent,
+				 ctxt.c_split_covers_rec);
 
 	if (ctxt.c_contig_type == CONTIG_NONE) {
 		if (ctxt.c_split_covers_rec)
@@ -5192,8 +5185,9 @@ int ocfs2_mark_extent_written(struct inode *inode,
 {
 	int ret;
 
-	mlog(0, "Inode %lu cpos %u, len %u, phys clusters %u\n",
-	     inode->i_ino, cpos, len, phys);
+	trace_ocfs2_mark_extent_written(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno,
+		cpos, len, phys);
 
 	if (!ocfs2_writes_unwritten_extents(OCFS2_SB(inode->i_sb))) {
 		ocfs2_error(inode->i_sb, "Inode %llu has unwritten extents "
@@ -5512,11 +5506,10 @@ int ocfs2_remove_extent(handle_t *handle,
 
 	BUG_ON(cpos < le32_to_cpu(rec->e_cpos) || trunc_range > rec_range);
 
-	mlog(0, "Owner %llu, remove (cpos %u, len %u). Existing index %d "
-	     "(cpos %u, len %u)\n",
-	     (unsigned long long)ocfs2_metadata_cache_owner(et->et_ci),
-	     cpos, len, index,
-	     le32_to_cpu(rec->e_cpos), ocfs2_rec_clusters(el, rec));
+	trace_ocfs2_remove_extent(
+		(unsigned long long)ocfs2_metadata_cache_owner(et->et_ci),
+		cpos, len, index, le32_to_cpu(rec->e_cpos),
+		ocfs2_rec_clusters(el, rec));
 
 	if (le32_to_cpu(rec->e_cpos) == cpos || rec_range == trunc_range) {
 		ret = ocfs2_truncate_rec(handle, et, path, index, dealloc,
@@ -5795,9 +5788,6 @@ int ocfs2_truncate_log_append(struct ocfs2_super *osb,
 	struct ocfs2_dinode *di;
 	struct ocfs2_truncate_log *tl;
 
-	mlog_entry("start_blk = %llu, num_clusters = %u\n",
-		   (unsigned long long)start_blk, num_clusters);
-
 	BUG_ON(mutex_trylock(&tl_inode->i_mutex));
 
 	start_cluster = ocfs2_blocks_to_clusters(osb->sb, start_blk);
@@ -5834,10 +5824,9 @@ int ocfs2_truncate_log_append(struct ocfs2_super *osb,
 		goto bail;
 	}
 
-	mlog(0, "Log truncate of %u clusters starting at cluster %u to "
-	     "%llu (index = %d)\n", num_clusters, start_cluster,
-	     (unsigned long long)OCFS2_I(tl_inode)->ip_blkno, index);
-
+	trace_ocfs2_truncate_log_append(
+		(unsigned long long)OCFS2_I(tl_inode)->ip_blkno, index,
+		start_cluster, num_clusters);
 	if (ocfs2_truncate_log_can_coalesce(tl, start_cluster)) {
 		/*
 		 * Move index back to the record we are coalescing with.
@@ -5846,9 +5835,10 @@ int ocfs2_truncate_log_append(struct ocfs2_super *osb,
 		index--;
 
 		num_clusters += le32_to_cpu(tl->tl_recs[index].t_clusters);
-		mlog(0, "Coalesce with index %u (start = %u, clusters = %u)\n",
-		     index, le32_to_cpu(tl->tl_recs[index].t_start),
-		     num_clusters);
+		trace_ocfs2_truncate_log_append(
+			(unsigned long long)OCFS2_I(tl_inode)->ip_blkno,
+			index, le32_to_cpu(tl->tl_recs[index].t_start),
+			num_clusters);
 	} else {
 		tl->tl_recs[index].t_start = cpu_to_le32(start_cluster);
 		tl->tl_used = cpu_to_le16(index + 1);
@@ -5859,7 +5849,6 @@ int ocfs2_truncate_log_append(struct ocfs2_super *osb,
 
 	osb->truncated_clusters += num_clusters;
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -5878,8 +5867,6 @@ static int ocfs2_replay_truncate_records(struct ocfs2_super *osb,
 	struct inode *tl_inode = osb->osb_tl_inode;
 	struct buffer_head *tl_bh = osb->osb_tl_bh;
 
-	mlog_entry_void();
-
 	di = (struct ocfs2_dinode *) tl_bh->b_data;
 	tl = &di->id2.i_dealloc;
 	i = le16_to_cpu(tl->tl_used) - 1;
@@ -5915,8 +5902,9 @@ static int ocfs2_replay_truncate_records(struct ocfs2_super *osb,
 		/* if start_blk is not set, we ignore the record as
 		 * invalid. */
 		if (start_blk) {
-			mlog(0, "free record %d, start = %u, clusters = %u\n",
-			     i, le32_to_cpu(rec.t_start), num_clusters);
+			trace_ocfs2_replay_truncate_records(
+				(unsigned long long)OCFS2_I(tl_inode)->ip_blkno,
+				i, le32_to_cpu(rec.t_start), num_clusters);
 
 			status = ocfs2_free_clusters(handle, data_alloc_inode,
 						     data_alloc_bh, start_blk,
@@ -5932,7 +5920,6 @@ static int ocfs2_replay_truncate_records(struct ocfs2_super *osb,
 	osb->truncated_clusters = 0;
 
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -5949,8 +5936,6 @@ int __ocfs2_flush_truncate_log(struct ocfs2_super *osb)
 	struct ocfs2_dinode *di;
 	struct ocfs2_truncate_log *tl;
 
-	mlog_entry_void();
-
 	BUG_ON(mutex_trylock(&tl_inode->i_mutex));
 
 	di = (struct ocfs2_dinode *) tl_bh->b_data;
@@ -5962,8 +5947,9 @@ int __ocfs2_flush_truncate_log(struct ocfs2_super *osb)
 
 	tl = &di->id2.i_dealloc;
 	num_to_flush = le16_to_cpu(tl->tl_used);
-	mlog(0, "Flush %u records from truncate log #%llu\n",
-	     num_to_flush, (unsigned long long)OCFS2_I(tl_inode)->ip_blkno);
+	trace_ocfs2_flush_truncate_log(
+		(unsigned long long)OCFS2_I(tl_inode)->ip_blkno,
+		num_to_flush);
 	if (!num_to_flush) {
 		status = 0;
 		goto out;
@@ -6009,7 +5995,6 @@ out_mutex:
 	iput(data_alloc_inode);
 
 out:
-	mlog_exit(status);
 	return status;
 }
 
@@ -6032,15 +6017,11 @@ static void ocfs2_truncate_log_worker(struct work_struct *work)
 		container_of(work, struct ocfs2_super,
 			     osb_truncate_log_wq.work);
 
-	mlog_entry_void();
-
 	status = ocfs2_flush_truncate_log(osb);
 	if (status < 0)
 		mlog_errno(status);
 	else
 		ocfs2_init_steal_slots(osb);
-
-	mlog_exit(status);
 }
 
 #define OCFS2_TRUNCATE_LOG_FLUSH_INTERVAL (2 * HZ)
@@ -6086,7 +6067,6 @@ static int ocfs2_get_truncate_log_info(struct ocfs2_super *osb,
 	*tl_inode = inode;
 	*tl_bh    = bh;
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -6106,7 +6086,7 @@ int ocfs2_begin_truncate_log_recovery(struct ocfs2_super *osb,
 
 	*tl_copy = NULL;
 
-	mlog(0, "recover truncate log from slot %d\n", slot_num);
+	trace_ocfs2_begin_truncate_log_recovery(slot_num);
 
 	status = ocfs2_get_truncate_log_info(osb, slot_num, &tl_inode, &tl_bh);
 	if (status < 0) {
@@ -6123,8 +6103,7 @@ int ocfs2_begin_truncate_log_recovery(struct ocfs2_super *osb,
 
 	tl = &di->id2.i_dealloc;
 	if (le16_to_cpu(tl->tl_used)) {
-		mlog(0, "We'll have %u logs to recover\n",
-		     le16_to_cpu(tl->tl_used));
+		trace_ocfs2_truncate_log_recovery_num(le16_to_cpu(tl->tl_used));
 
 		*tl_copy = kmalloc(tl_bh->b_size, GFP_KERNEL);
 		if (!(*tl_copy)) {
@@ -6157,9 +6136,9 @@ bail:
 	if (status < 0 && (*tl_copy)) {
 		kfree(*tl_copy);
 		*tl_copy = NULL;
+		mlog_errno(status);
 	}
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -6174,8 +6153,6 @@ int ocfs2_complete_truncate_log_recovery(struct ocfs2_super *osb,
 	struct inode *tl_inode = osb->osb_tl_inode;
 	struct ocfs2_truncate_log *tl;
 
-	mlog_entry_void();
-
 	if (OCFS2_I(tl_inode)->ip_blkno == le64_to_cpu(tl_copy->i_blkno)) {
 		mlog(ML_ERROR, "Asked to recover my own truncate log!\n");
 		return -EINVAL;
@@ -6183,8 +6160,9 @@ int ocfs2_complete_truncate_log_recovery(struct ocfs2_super *osb,
 
 	tl = &tl_copy->id2.i_dealloc;
 	num_recs = le16_to_cpu(tl->tl_used);
-	mlog(0, "cleanup %u records from %llu\n", num_recs,
-	     (unsigned long long)le64_to_cpu(tl_copy->i_blkno));
+	trace_ocfs2_complete_truncate_log_recovery(
+		(unsigned long long)le64_to_cpu(tl_copy->i_blkno),
+		num_recs);
 
 	mutex_lock(&tl_inode->i_mutex);
 	for(i = 0; i < num_recs; i++) {
@@ -6219,7 +6197,6 @@ int ocfs2_complete_truncate_log_recovery(struct ocfs2_super *osb,
 bail_up:
 	mutex_unlock(&tl_inode->i_mutex);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -6228,8 +6205,6 @@ void ocfs2_truncate_log_shutdown(struct ocfs2_super *osb)
 	int status;
 	struct inode *tl_inode = osb->osb_tl_inode;
 
-	mlog_entry_void();
-
 	if (tl_inode) {
 		cancel_delayed_work(&osb->osb_truncate_log_wq);
 		flush_workqueue(ocfs2_wq);
@@ -6241,8 +6216,6 @@ void ocfs2_truncate_log_shutdown(struct ocfs2_super *osb)
 		brelse(osb->osb_tl_bh);
 		iput(osb->osb_tl_inode);
 	}
-
-	mlog_exit_void();
 }
 
 int ocfs2_truncate_log_init(struct ocfs2_super *osb)
@@ -6251,8 +6224,6 @@ int ocfs2_truncate_log_init(struct ocfs2_super *osb)
 	struct inode *tl_inode = NULL;
 	struct buffer_head *tl_bh = NULL;
 
-	mlog_entry_void();
-
 	status = ocfs2_get_truncate_log_info(osb,
 					     osb->slot_num,
 					     &tl_inode,
@@ -6268,7 +6239,6 @@ int ocfs2_truncate_log_init(struct ocfs2_super *osb)
 	osb->osb_tl_bh    = tl_bh;
 	osb->osb_tl_inode = tl_inode;
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -6350,8 +6320,8 @@ static int ocfs2_free_cached_blocks(struct ocfs2_super *osb,
 		else
 			bg_blkno = ocfs2_which_suballoc_group(head->free_blk,
 							      head->free_bit);
-		mlog(0, "Free bit: (bit %u, blkno %llu)\n",
-		     head->free_bit, (unsigned long long)head->free_blk);
+		trace_ocfs2_free_cached_blocks(
+		     (unsigned long long)head->free_blk, head->free_bit);
 
 		ret = ocfs2_free_suballoc_bits(handle, inode, di_bh,
 					       head->free_bit, bg_blkno, 1);
@@ -6404,8 +6374,7 @@ int ocfs2_cache_cluster_dealloc(struct ocfs2_cached_dealloc_ctxt *ctxt,
 		return ret;
 	}
 
-	mlog(0, "Insert clusters: (bit %u, blk %llu)\n",
-	     bit, (unsigned long long)blkno);
+	trace_ocfs2_cache_cluster_dealloc((unsigned long long)blkno, bit);
 
 	item->free_blk = blkno;
 	item->free_bit = bit;
@@ -6480,8 +6449,8 @@ int ocfs2_run_deallocs(struct ocfs2_super *osb,
 		fl = ctxt->c_first_suballocator;
 
 		if (fl->f_first) {
-			mlog(0, "Free items: (type %u, slot %d)\n",
-			     fl->f_inode_type, fl->f_slot);
+			trace_ocfs2_run_deallocs(fl->f_inode_type,
+						 fl->f_slot);
 			ret2 = ocfs2_free_cached_blocks(osb,
 							fl->f_inode_type,
 							fl->f_slot,
@@ -6558,8 +6527,9 @@ int ocfs2_cache_block_dealloc(struct ocfs2_cached_dealloc_ctxt *ctxt,
 		goto out;
 	}
 
-	mlog(0, "Insert: (type %d, slot %u, bit %u, blk %llu)\n",
-	     type, slot, bit, (unsigned long long)blkno);
+	trace_ocfs2_cache_block_dealloc(type, slot,
+					(unsigned long long)suballoc,
+					(unsigned long long)blkno, bit);
 
 	item->free_bg = suballoc;
 	item->free_blk = blkno;
@@ -7005,8 +6975,6 @@ int ocfs2_commit_truncate(struct ocfs2_super *osb,
 	struct ocfs2_extent_tree et;
 	struct ocfs2_cached_dealloc_ctxt dealloc;
 
-	mlog_entry_void();
-
 	ocfs2_init_dinode_extent_tree(&et, INODE_CACHE(inode), di_bh);
 	ocfs2_init_dealloc_ctxt(&dealloc);
 
@@ -7041,8 +7009,11 @@ start:
 		goto bail;
 	}
 
-	mlog(0, "inode->ip_clusters = %u, tree_depth = %u\n",
-	     OCFS2_I(inode)->ip_clusters, path->p_tree_depth);
+	trace_ocfs2_commit_truncate(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno,
+		new_highest_cpos,
+		OCFS2_I(inode)->ip_clusters,
+		path->p_tree_depth);
 
 	/*
 	 * By now, el will point to the extent list on the bottom most
@@ -7136,7 +7107,6 @@ bail:
 
 	ocfs2_free_path(path);
 
-	mlog_exit(status);
 	return status;
 }
 
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 1fbb0e20131..0d44b770184 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -29,7 +29,6 @@
 #include <linux/mpage.h>
 #include <linux/quotaops.h>
 
-#define MLOG_MASK_PREFIX ML_FILE_IO
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -45,6 +44,7 @@
 #include "super.h"
 #include "symlink.h"
 #include "refcounttree.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -59,8 +59,9 @@ static int ocfs2_symlink_get_block(struct inode *inode, sector_t iblock,
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	void *kaddr;
 
-	mlog_entry("(0x%p, %llu, 0x%p, %d)\n", inode,
-		   (unsigned long long)iblock, bh_result, create);
+	trace_ocfs2_symlink_get_block(
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			(unsigned long long)iblock, bh_result, create);
 
 	BUG_ON(ocfs2_inode_is_fast_symlink(inode));
 
@@ -123,7 +124,6 @@ static int ocfs2_symlink_get_block(struct inode *inode, sector_t iblock,
 bail:
 	brelse(bh);
 
-	mlog_exit(err);
 	return err;
 }
 
@@ -136,8 +136,8 @@ int ocfs2_get_block(struct inode *inode, sector_t iblock,
 	u64 p_blkno, count, past_eof;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
-	mlog_entry("(0x%p, %llu, 0x%p, %d)\n", inode,
-		   (unsigned long long)iblock, bh_result, create);
+	trace_ocfs2_get_block((unsigned long long)OCFS2_I(inode)->ip_blkno,
+			      (unsigned long long)iblock, bh_result, create);
 
 	if (OCFS2_I(inode)->ip_flags & OCFS2_INODE_SYSTEM_FILE)
 		mlog(ML_NOTICE, "get_block on system inode 0x%p (%lu)\n",
@@ -199,8 +199,9 @@ int ocfs2_get_block(struct inode *inode, sector_t iblock,
 	}
 
 	past_eof = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode));
-	mlog(0, "Inode %lu, past_eof = %llu\n", inode->i_ino,
-	     (unsigned long long)past_eof);
+
+	trace_ocfs2_get_block_end((unsigned long long)OCFS2_I(inode)->ip_blkno,
+				  (unsigned long long)past_eof);
 	if (create && (iblock >= past_eof))
 		set_buffer_new(bh_result);
 
@@ -208,7 +209,6 @@ bail:
 	if (err < 0)
 		err = -EIO;
 
-	mlog_exit(err);
 	return err;
 }
 
@@ -278,7 +278,8 @@ static int ocfs2_readpage(struct file *file, struct page *page)
 	loff_t start = (loff_t)page->index << PAGE_CACHE_SHIFT;
 	int ret, unlock = 1;
 
-	mlog_entry("(0x%p, %lu)\n", file, (page ? page->index : 0));
+	trace_ocfs2_readpage((unsigned long long)oi->ip_blkno,
+			     (page ? page->index : 0));
 
 	ret = ocfs2_inode_lock_with_page(inode, NULL, 0, page);
 	if (ret != 0) {
@@ -323,7 +324,6 @@ out_inode_unlock:
 out:
 	if (unlock)
 		unlock_page(page);
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -396,15 +396,11 @@ out_unlock:
  */
 static int ocfs2_writepage(struct page *page, struct writeback_control *wbc)
 {
-	int ret;
-
-	mlog_entry("(0x%p)\n", page);
-
-	ret = block_write_full_page(page, ocfs2_get_block, wbc);
+	trace_ocfs2_writepage(
+		(unsigned long long)OCFS2_I(page->mapping->host)->ip_blkno,
+		page->index);
 
-	mlog_exit(ret);
-
-	return ret;
+	return block_write_full_page(page, ocfs2_get_block, wbc);
 }
 
 /* Taken from ext3. We don't necessarily need the full blown
@@ -450,7 +446,8 @@ static sector_t ocfs2_bmap(struct address_space *mapping, sector_t block)
 	int err = 0;
 	struct inode *inode = mapping->host;
 
-	mlog_entry("(block = %llu)\n", (unsigned long long)block);
+	trace_ocfs2_bmap((unsigned long long)OCFS2_I(inode)->ip_blkno,
+			 (unsigned long long)block);
 
 	/* We don't need to lock journal system files, since they aren't
 	 * accessed concurrently from multiple nodes.
@@ -484,8 +481,6 @@ static sector_t ocfs2_bmap(struct address_space *mapping, sector_t block)
 bail:
 	status = err ? 0 : p_blkno;
 
-	mlog_exit((int)status);
-
 	return status;
 }
 
@@ -616,9 +611,6 @@ static ssize_t ocfs2_direct_IO(int rw,
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file->f_path.dentry->d_inode->i_mapping->host;
-	int ret;
-
-	mlog_entry_void();
 
 	/*
 	 * Fallback to buffered I/O if we see an inode without
@@ -631,13 +623,10 @@ static ssize_t ocfs2_direct_IO(int rw,
 	if (i_size_read(inode) <= offset)
 		return 0;
 
-	ret = __blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev,
-				   iov, offset, nr_segs,
-				   ocfs2_direct_IO_get_blocks,
-				   ocfs2_dio_end_io, NULL, 0);
-
-	mlog_exit(ret);
-	return ret;
+	return __blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev,
+				    iov, offset, nr_segs,
+				    ocfs2_direct_IO_get_blocks,
+				    ocfs2_dio_end_io, NULL, 0);
 }
 
 static void ocfs2_figure_cluster_boundaries(struct ocfs2_super *osb,
@@ -1534,9 +1523,9 @@ static int ocfs2_try_to_write_inline_data(struct address_space *mapping,
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 	struct ocfs2_dinode *di = NULL;
 
-	mlog(0, "Inode %llu, write of %u bytes at off %llu. features: 0x%x\n",
-	     (unsigned long long)oi->ip_blkno, len, (unsigned long long)pos,
-	     oi->ip_dyn_features);
+	trace_ocfs2_try_to_write_inline_data((unsigned long long)oi->ip_blkno,
+					     len, (unsigned long long)pos,
+					     oi->ip_dyn_features);
 
 	/*
 	 * Handle inodes which already have inline data 1st.
@@ -1739,6 +1728,13 @@ try_again:
 
 	di = (struct ocfs2_dinode *)wc->w_di_bh->b_data;
 
+	trace_ocfs2_write_begin_nolock(
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			(long long)i_size_read(inode),
+			le32_to_cpu(di->i_clusters),
+			pos, len, flags, mmap_page,
+			clusters_to_alloc, extents_to_split);
+
 	/*
 	 * We set w_target_from, w_target_to here so that
 	 * ocfs2_write_end() knows which range in the target page to
@@ -1751,12 +1747,6 @@ try_again:
 		 * ocfs2_lock_allocators(). It greatly over-estimates
 		 * the work to be done.
 		 */
-		mlog(0, "extend inode %llu, i_size = %lld, di->i_clusters = %u,"
-		     " clusters_to_add = %u, extents_to_split = %u\n",
-		     (unsigned long long)OCFS2_I(inode)->ip_blkno,
-		     (long long)i_size_read(inode), le32_to_cpu(di->i_clusters),
-		     clusters_to_alloc, extents_to_split);
-
 		ocfs2_init_dinode_extent_tree(&et, INODE_CACHE(inode),
 					      wc->w_di_bh);
 		ret = ocfs2_lock_allocators(inode, &et,
@@ -1938,8 +1928,8 @@ static void ocfs2_write_end_inline(struct inode *inode, loff_t pos,
 	memcpy(di->id2.i_data.id_data + pos, kaddr + pos, *copied);
 	kunmap_atomic(kaddr, KM_USER0);
 
-	mlog(0, "Data written to inode at offset %llu. "
-	     "id_count = %u, copied = %u, i_dyn_features = 0x%x\n",
+	trace_ocfs2_write_end_inline(
+	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
 	     (unsigned long long)pos, *copied,
 	     le16_to_cpu(di->id2.i_data.id_count),
 	     le16_to_cpu(di->i_dyn_features));
diff --git a/fs/ocfs2/buffer_head_io.c b/fs/ocfs2/buffer_head_io.c
index f9d5d3ffc75..5d18ad10c27 100644
--- a/fs/ocfs2/buffer_head_io.c
+++ b/fs/ocfs2/buffer_head_io.c
@@ -35,8 +35,8 @@
 #include "inode.h"
 #include "journal.h"
 #include "uptodate.h"
-
 #include "buffer_head_io.h"
+#include "ocfs2_trace.h"
 
 /*
  * Bits on bh->b_state used by ocfs2.
@@ -55,8 +55,7 @@ int ocfs2_write_block(struct ocfs2_super *osb, struct buffer_head *bh,
 {
 	int ret = 0;
 
-	mlog_entry("(bh->b_blocknr = %llu, ci=%p)\n",
-		   (unsigned long long)bh->b_blocknr, ci);
+	trace_ocfs2_write_block((unsigned long long)bh->b_blocknr, ci);
 
 	BUG_ON(bh->b_blocknr < OCFS2_SUPER_BLOCK_BLKNO);
 	BUG_ON(buffer_jbd(bh));
@@ -66,6 +65,7 @@ int ocfs2_write_block(struct ocfs2_super *osb, struct buffer_head *bh,
 	 * can get modified during recovery even if read-only. */
 	if (ocfs2_is_hard_readonly(osb)) {
 		ret = -EROFS;
+		mlog_errno(ret);
 		goto out;
 	}
 
@@ -91,11 +91,11 @@ int ocfs2_write_block(struct ocfs2_super *osb, struct buffer_head *bh,
 		 * uptodate. */
 		ret = -EIO;
 		put_bh(bh);
+		mlog_errno(ret);
 	}
 
 	ocfs2_metadata_cache_io_unlock(ci);
 out:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -106,10 +106,10 @@ int ocfs2_read_blocks_sync(struct ocfs2_super *osb, u64 block,
 	unsigned int i;
 	struct buffer_head *bh;
 
-	if (!nr) {
-		mlog(ML_BH_IO, "No buffers will be read!\n");
+	trace_ocfs2_read_blocks_sync((unsigned long long)block, nr);
+
+	if (!nr)
 		goto bail;
-	}
 
 	for (i = 0 ; i < nr ; i++) {
 		if (bhs[i] == NULL) {
@@ -123,10 +123,8 @@ int ocfs2_read_blocks_sync(struct ocfs2_super *osb, u64 block,
 		bh = bhs[i];
 
 		if (buffer_jbd(bh)) {
-			mlog(ML_BH_IO,
-			     "trying to sync read a jbd "
-			     "managed bh (blocknr = %llu), skipping\n",
-			     (unsigned long long)bh->b_blocknr);
+			trace_ocfs2_read_blocks_sync_jbd(
+					(unsigned long long)bh->b_blocknr);
 			continue;
 		}
 
@@ -186,8 +184,7 @@ int ocfs2_read_blocks(struct ocfs2_caching_info *ci, u64 block, int nr,
 	struct buffer_head *bh;
 	struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
 
-	mlog_entry("(ci=%p, block=(%llu), nr=(%d), flags=%d)\n",
-		   ci, (unsigned long long)block, nr, flags);
+	trace_ocfs2_read_blocks_begin(ci, (unsigned long long)block, nr, flags);
 
 	BUG_ON(!ci);
 	BUG_ON((flags & OCFS2_BH_READAHEAD) &&
@@ -207,7 +204,6 @@ int ocfs2_read_blocks(struct ocfs2_caching_info *ci, u64 block, int nr,
 	}
 
 	if (nr == 0) {
-		mlog(ML_BH_IO, "No buffers will be read!\n");
 		status = 0;
 		goto bail;
 	}
@@ -251,8 +247,7 @@ int ocfs2_read_blocks(struct ocfs2_caching_info *ci, u64 block, int nr,
 		 */
 
 		if (!ignore_cache && !ocfs2_buffer_uptodate(ci, bh)) {
-			mlog(ML_UPTODATE,
-			     "bh (%llu), owner %llu not uptodate\n",
+			trace_ocfs2_read_blocks_from_disk(
 			     (unsigned long long)bh->b_blocknr,
 			     (unsigned long long)ocfs2_metadata_cache_owner(ci));
 			/* We're using ignore_cache here to say
@@ -260,11 +255,10 @@ int ocfs2_read_blocks(struct ocfs2_caching_info *ci, u64 block, int nr,
 			ignore_cache = 1;
 		}
 
+		trace_ocfs2_read_blocks_bh((unsigned long long)bh->b_blocknr,
+			ignore_cache, buffer_jbd(bh), buffer_dirty(bh));
+
 		if (buffer_jbd(bh)) {
-			if (ignore_cache)
-				mlog(ML_BH_IO, "trying to sync read a jbd "
-					       "managed bh (blocknr = %llu)\n",
-				     (unsigned long long)bh->b_blocknr);
 			continue;
 		}
 
@@ -272,9 +266,6 @@ int ocfs2_read_blocks(struct ocfs2_caching_info *ci, u64 block, int nr,
 			if (buffer_dirty(bh)) {
 				/* This should probably be a BUG, or
 				 * at least return an error. */
-				mlog(ML_BH_IO, "asking me to sync read a dirty "
-					       "buffer! (blocknr = %llu)\n",
-				     (unsigned long long)bh->b_blocknr);
 				continue;
 			}
 
@@ -367,14 +358,11 @@ int ocfs2_read_blocks(struct ocfs2_caching_info *ci, u64 block, int nr,
 	}
 	ocfs2_metadata_cache_io_unlock(ci);
 
-	mlog(ML_BH_IO, "block=(%llu), nr=(%d), cached=%s, flags=0x%x\n",
-	     (unsigned long long)block, nr,
-	     ((flags & OCFS2_BH_IGNORE_CACHE) || ignore_cache) ? "no" : "yes",
-	     flags);
+	trace_ocfs2_read_blocks_end((unsigned long long)block, nr,
+				    flags, ignore_cache);
 
 bail:
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -408,13 +396,12 @@ int ocfs2_write_super_or_backup(struct ocfs2_super *osb,
 	int ret = 0;
 	struct ocfs2_dinode *di = (struct ocfs2_dinode *)bh->b_data;
 
-	mlog_entry_void();
-
 	BUG_ON(buffer_jbd(bh));
 	ocfs2_check_super_or_backup(osb->sb, bh->b_blocknr);
 
 	if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) {
 		ret = -EROFS;
+		mlog_errno(ret);
 		goto out;
 	}
 
@@ -434,9 +421,9 @@ int ocfs2_write_super_or_backup(struct ocfs2_super *osb,
 	if (!buffer_uptodate(bh)) {
 		ret = -EIO;
 		put_bh(bh);
+		mlog_errno(ret);
 	}
 
 out:
-	mlog_exit(ret);
 	return ret;
 }
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index b108e863d8f..fc9e96a03a6 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -1658,8 +1658,6 @@ static int o2hb_populate_slot_data(struct o2hb_region *reg)
 	struct o2hb_disk_slot *slot;
 	struct o2hb_disk_heartbeat_block *hb_block;
 
-	mlog_entry_void();
-
 	ret = o2hb_read_slots(reg, reg->hr_blocks);
 	if (ret) {
 		mlog_errno(ret);
@@ -1681,7 +1679,6 @@ static int o2hb_populate_slot_data(struct o2hb_region *reg)
 	}
 
 out:
-	mlog_exit(ret);
 	return ret;
 }
 
diff --git a/fs/ocfs2/cluster/masklog.c b/fs/ocfs2/cluster/masklog.c
index 6c61771469a..07ac24fd925 100644
--- a/fs/ocfs2/cluster/masklog.c
+++ b/fs/ocfs2/cluster/masklog.c
@@ -30,7 +30,7 @@
 
 struct mlog_bits mlog_and_bits = MLOG_BITS_RHS(MLOG_INITIAL_AND_MASK);
 EXPORT_SYMBOL_GPL(mlog_and_bits);
-struct mlog_bits mlog_not_bits = MLOG_BITS_RHS(MLOG_INITIAL_NOT_MASK);
+struct mlog_bits mlog_not_bits = MLOG_BITS_RHS(0);
 EXPORT_SYMBOL_GPL(mlog_not_bits);
 
 static ssize_t mlog_mask_show(u64 mask, char *buf)
@@ -80,8 +80,6 @@ struct mlog_attribute {
 }
 
 static struct mlog_attribute mlog_attrs[MLOG_MAX_BITS] = {
-	define_mask(ENTRY),
-	define_mask(EXIT),
 	define_mask(TCP),
 	define_mask(MSG),
 	define_mask(SOCKET),
@@ -93,27 +91,11 @@ static struct mlog_attribute mlog_attrs[MLOG_MAX_BITS] = {
 	define_mask(DLM_THREAD),
 	define_mask(DLM_MASTER),
 	define_mask(DLM_RECOVERY),
-	define_mask(AIO),
-	define_mask(JOURNAL),
-	define_mask(DISK_ALLOC),
-	define_mask(SUPER),
-	define_mask(FILE_IO),
-	define_mask(EXTENT_MAP),
 	define_mask(DLM_GLUE),
-	define_mask(BH_IO),
-	define_mask(UPTODATE),
-	define_mask(NAMEI),
-	define_mask(INODE),
 	define_mask(VOTE),
-	define_mask(DCACHE),
 	define_mask(CONN),
 	define_mask(QUORUM),
-	define_mask(EXPORT),
-	define_mask(XATTR),
-	define_mask(QUOTA),
-	define_mask(REFCOUNT),
 	define_mask(BASTS),
-	define_mask(RESERVATIONS),
 	define_mask(CLUSTER),
 	define_mask(ERROR),
 	define_mask(NOTICE),
diff --git a/fs/ocfs2/cluster/masklog.h b/fs/ocfs2/cluster/masklog.h
index 34d6544357d..baa2b9ef7ee 100644
--- a/fs/ocfs2/cluster/masklog.h
+++ b/fs/ocfs2/cluster/masklog.h
@@ -82,41 +82,23 @@
 
 /* bits that are frequently given and infrequently matched in the low word */
 /* NOTE: If you add a flag, you need to also update masklog.c! */
-#define ML_ENTRY	0x0000000000000001ULL /* func call entry */
-#define ML_EXIT		0x0000000000000002ULL /* func call exit */
-#define ML_TCP		0x0000000000000004ULL /* net cluster/tcp.c */
-#define ML_MSG		0x0000000000000008ULL /* net network messages */
-#define ML_SOCKET	0x0000000000000010ULL /* net socket lifetime */
-#define ML_HEARTBEAT	0x0000000000000020ULL /* hb all heartbeat tracking */
-#define ML_HB_BIO	0x0000000000000040ULL /* hb io tracing */
-#define ML_DLMFS	0x0000000000000080ULL /* dlm user dlmfs */
-#define ML_DLM		0x0000000000000100ULL /* dlm general debugging */
-#define ML_DLM_DOMAIN	0x0000000000000200ULL /* dlm domain debugging */
-#define ML_DLM_THREAD	0x0000000000000400ULL /* dlm domain thread */
-#define ML_DLM_MASTER	0x0000000000000800ULL /* dlm master functions */
-#define ML_DLM_RECOVERY	0x0000000000001000ULL /* dlm master functions */
-#define ML_AIO		0x0000000000002000ULL /* ocfs2 aio read and write */
-#define ML_JOURNAL	0x0000000000004000ULL /* ocfs2 journalling functions */
-#define ML_DISK_ALLOC	0x0000000000008000ULL /* ocfs2 disk allocation */
-#define ML_SUPER	0x0000000000010000ULL /* ocfs2 mount / umount */
-#define ML_FILE_IO	0x0000000000020000ULL /* ocfs2 file I/O */
-#define ML_EXTENT_MAP	0x0000000000040000ULL /* ocfs2 extent map caching */
-#define ML_DLM_GLUE	0x0000000000080000ULL /* ocfs2 dlm glue layer */
-#define ML_BH_IO	0x0000000000100000ULL /* ocfs2 buffer I/O */
-#define ML_UPTODATE	0x0000000000200000ULL /* ocfs2 caching sequence #'s */
-#define ML_NAMEI	0x0000000000400000ULL /* ocfs2 directory / namespace */
-#define ML_INODE	0x0000000000800000ULL /* ocfs2 inode manipulation */
-#define ML_VOTE		0x0000000001000000ULL /* ocfs2 node messaging  */
-#define ML_DCACHE	0x0000000002000000ULL /* ocfs2 dcache operations */
-#define ML_CONN		0x0000000004000000ULL /* net connection management */
-#define ML_QUORUM	0x0000000008000000ULL /* net connection quorum */
-#define ML_EXPORT	0x0000000010000000ULL /* ocfs2 export operations */
-#define ML_XATTR	0x0000000020000000ULL /* ocfs2 extended attributes */
-#define ML_QUOTA	0x0000000040000000ULL /* ocfs2 quota operations */
-#define ML_REFCOUNT	0x0000000080000000ULL /* refcount tree operations */
-#define ML_BASTS	0x0000000100000000ULL /* dlmglue asts and basts */
-#define ML_RESERVATIONS	0x0000000200000000ULL /* ocfs2 alloc reservations */
-#define ML_CLUSTER	0x0000000400000000ULL /* cluster stack */
+#define ML_TCP		0x0000000000000001ULL /* net cluster/tcp.c */
+#define ML_MSG		0x0000000000000002ULL /* net network messages */
+#define ML_SOCKET	0x0000000000000004ULL /* net socket lifetime */
+#define ML_HEARTBEAT	0x0000000000000008ULL /* hb all heartbeat tracking */
+#define ML_HB_BIO	0x0000000000000010ULL /* hb io tracing */
+#define ML_DLMFS	0x0000000000000020ULL /* dlm user dlmfs */
+#define ML_DLM		0x0000000000000040ULL /* dlm general debugging */
+#define ML_DLM_DOMAIN	0x0000000000000080ULL /* dlm domain debugging */
+#define ML_DLM_THREAD	0x0000000000000100ULL /* dlm domain thread */
+#define ML_DLM_MASTER	0x0000000000000200ULL /* dlm master functions */
+#define ML_DLM_RECOVERY	0x0000000000000400ULL /* dlm master functions */
+#define ML_DLM_GLUE	0x0000000000000800ULL /* ocfs2 dlm glue layer */
+#define ML_VOTE		0x0000000000001000ULL /* ocfs2 node messaging  */
+#define ML_CONN		0x0000000000002000ULL /* net connection management */
+#define ML_QUORUM	0x0000000000004000ULL /* net connection quorum */
+#define ML_BASTS	0x0000000000008000ULL /* dlmglue asts and basts */
+#define ML_CLUSTER	0x0000000000010000ULL /* cluster stack */
 
 /* bits that are infrequently given and frequently matched in the high word */
 #define ML_ERROR	0x1000000000000000ULL /* sent to KERN_ERR */
@@ -124,7 +106,6 @@
 #define ML_KTHREAD	0x4000000000000000ULL /* kernel thread activity */
 
 #define MLOG_INITIAL_AND_MASK (ML_ERROR|ML_NOTICE)
-#define MLOG_INITIAL_NOT_MASK (ML_ENTRY|ML_EXIT)
 #ifndef MLOG_MASK_PREFIX
 #define MLOG_MASK_PREFIX 0
 #endif
@@ -222,58 +203,6 @@ extern struct mlog_bits mlog_and_bits, mlog_not_bits;
 		mlog(ML_ERROR, "status = %lld\n", (long long)_st);	\
 } while (0)
 
-#if defined(CONFIG_OCFS2_DEBUG_MASKLOG)
-#define mlog_entry(fmt, args...) do {					\
-	mlog(ML_ENTRY, "ENTRY:" fmt , ##args);				\
-} while (0)
-
-#define mlog_entry_void() do {						\
-	mlog(ML_ENTRY, "ENTRY:\n");					\
-} while (0)
-
-/*
- * We disable this for sparse.
- */
-#if !defined(__CHECKER__)
-#define mlog_exit(st) do {						     \
-	if (__builtin_types_compatible_p(typeof(st), unsigned long))	     \
-		mlog(ML_EXIT, "EXIT: %lu\n", (unsigned long) (st));	     \
-	else if (__builtin_types_compatible_p(typeof(st), signed long))      \
-		mlog(ML_EXIT, "EXIT: %ld\n", (signed long) (st));	     \
-	else if (__builtin_types_compatible_p(typeof(st), unsigned int)	     \
-		 || __builtin_types_compatible_p(typeof(st), unsigned short) \
-		 || __builtin_types_compatible_p(typeof(st), unsigned char)) \
-		mlog(ML_EXIT, "EXIT: %u\n", (unsigned int) (st));	     \
-	else if (__builtin_types_compatible_p(typeof(st), signed int)	     \
-		 || __builtin_types_compatible_p(typeof(st), signed short)   \
-		 || __builtin_types_compatible_p(typeof(st), signed char))   \
-		mlog(ML_EXIT, "EXIT: %d\n", (signed int) (st));		     \
-	else if (__builtin_types_compatible_p(typeof(st), long long))	     \
-		mlog(ML_EXIT, "EXIT: %lld\n", (long long) (st));	     \
-	else								     \
-		mlog(ML_EXIT, "EXIT: %llu\n", (unsigned long long) (st));    \
-} while (0)
-#else
-#define mlog_exit(st) do {						     \
-	mlog(ML_EXIT, "EXIT: %lld\n", (long long) (st));		     \
-} while (0)
-#endif
-
-#define mlog_exit_ptr(ptr) do {						\
-	mlog(ML_EXIT, "EXIT: %p\n", ptr);				\
-} while (0)
-
-#define mlog_exit_void() do {						\
-	mlog(ML_EXIT, "EXIT\n");					\
-} while (0)
-#else
-#define mlog_entry(...)  do { } while (0)
-#define mlog_entry_void(...)  do { } while (0)
-#define mlog_exit(...)  do { } while (0)
-#define mlog_exit_ptr(...)  do { } while (0)
-#define mlog_exit_void(...)  do { } while (0)
-#endif  /* defined(CONFIG_OCFS2_DEBUG_MASKLOG) */
-
 #define mlog_bug_on_msg(cond, fmt, args...) do {			\
 	if (cond) {							\
 		mlog(ML_ERROR, "bug expression: " #cond "\n");		\
diff --git a/fs/ocfs2/dcache.c b/fs/ocfs2/dcache.c
index 6d80ecc7834..e5ba3481833 100644
--- a/fs/ocfs2/dcache.c
+++ b/fs/ocfs2/dcache.c
@@ -28,7 +28,6 @@
 #include <linux/slab.h>
 #include <linux/namei.h>
 
-#define MLOG_MASK_PREFIX ML_DCACHE
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -39,6 +38,7 @@
 #include "file.h"
 #include "inode.h"
 #include "super.h"
+#include "ocfs2_trace.h"
 
 void ocfs2_dentry_attach_gen(struct dentry *dentry)
 {
@@ -56,14 +56,14 @@ static int ocfs2_dentry_revalidate(struct dentry *dentry,
 	int ret = 0;    /* if all else fails, just return false */
 	struct ocfs2_super *osb;
 
-	if (nd->flags & LOOKUP_RCU)
+	if (nd && nd->flags & LOOKUP_RCU)
 		return -ECHILD;
 
 	inode = dentry->d_inode;
 	osb = OCFS2_SB(dentry->d_sb);
 
-	mlog_entry("(0x%p, '%.*s')\n", dentry,
-		   dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_dentry_revalidate(dentry, dentry->d_name.len,
+				      dentry->d_name.name);
 
 	/* For a negative dentry -
 	 * check the generation number of the parent and compare with the
@@ -73,9 +73,10 @@ static int ocfs2_dentry_revalidate(struct dentry *dentry,
 		unsigned long gen = (unsigned long) dentry->d_fsdata;
 		unsigned long pgen =
 			OCFS2_I(dentry->d_parent->d_inode)->ip_dir_lock_gen;
-		mlog(0, "negative dentry: %.*s parent gen: %lu "
-			"dentry gen: %lu\n",
-			dentry->d_name.len, dentry->d_name.name, pgen, gen);
+
+		trace_ocfs2_dentry_revalidate_negative(dentry->d_name.len,
+						       dentry->d_name.name,
+						       pgen, gen);
 		if (gen != pgen)
 			goto bail;
 		goto valid;
@@ -90,8 +91,8 @@ static int ocfs2_dentry_revalidate(struct dentry *dentry,
 	/* did we or someone else delete this inode? */
 	if (OCFS2_I(inode)->ip_flags & OCFS2_INODE_DELETED) {
 		spin_unlock(&OCFS2_I(inode)->ip_lock);
-		mlog(0, "inode (%llu) deleted, returning false\n",
-		     (unsigned long long)OCFS2_I(inode)->ip_blkno);
+		trace_ocfs2_dentry_revalidate_delete(
+				(unsigned long long)OCFS2_I(inode)->ip_blkno);
 		goto bail;
 	}
 	spin_unlock(&OCFS2_I(inode)->ip_lock);
@@ -101,10 +102,9 @@ static int ocfs2_dentry_revalidate(struct dentry *dentry,
 	 * inode nlink hits zero, it never goes back.
 	 */
 	if (inode->i_nlink == 0) {
-		mlog(0, "Inode %llu orphaned, returning false "
-		     "dir = %d\n",
-		     (unsigned long long)OCFS2_I(inode)->ip_blkno,
-		     S_ISDIR(inode->i_mode));
+		trace_ocfs2_dentry_revalidate_orphaned(
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			S_ISDIR(inode->i_mode));
 		goto bail;
 	}
 
@@ -113,9 +113,8 @@ static int ocfs2_dentry_revalidate(struct dentry *dentry,
 	 * redo it.
 	 */
 	if (!dentry->d_fsdata) {
-		mlog(0, "Inode %llu doesn't have dentry lock, "
-		     "returning false\n",
-		     (unsigned long long)OCFS2_I(inode)->ip_blkno);
+		trace_ocfs2_dentry_revalidate_nofsdata(
+				(unsigned long long)OCFS2_I(inode)->ip_blkno);
 		goto bail;
 	}
 
@@ -123,8 +122,7 @@ valid:
 	ret = 1;
 
 bail:
-	mlog_exit(ret);
-
+	trace_ocfs2_dentry_revalidate_ret(ret);
 	return ret;
 }
 
@@ -181,8 +179,8 @@ struct dentry *ocfs2_find_local_alias(struct inode *inode,
 
 		spin_lock(&dentry->d_lock);
 		if (ocfs2_match_dentry(dentry, parent_blkno, skip_unhashed)) {
-			mlog(0, "dentry found: %.*s\n",
-			     dentry->d_name.len, dentry->d_name.name);
+			trace_ocfs2_find_local_alias(dentry->d_name.len,
+						     dentry->d_name.name);
 
 			dget_dlock(dentry);
 			spin_unlock(&dentry->d_lock);
@@ -240,9 +238,8 @@ int ocfs2_dentry_attach_lock(struct dentry *dentry,
 	struct dentry *alias;
 	struct ocfs2_dentry_lock *dl = dentry->d_fsdata;
 
-	mlog(0, "Attach \"%.*s\", parent %llu, fsdata: %p\n",
-	     dentry->d_name.len, dentry->d_name.name,
-	     (unsigned long long)parent_blkno, dl);
+	trace_ocfs2_dentry_attach_lock(dentry->d_name.len, dentry->d_name.name,
+				       (unsigned long long)parent_blkno, dl);
 
 	/*
 	 * Negative dentry. We ignore these for now.
@@ -292,7 +289,9 @@ int ocfs2_dentry_attach_lock(struct dentry *dentry,
 				(unsigned long long)parent_blkno,
 				(unsigned long long)dl->dl_parent_blkno);
 
-		mlog(0, "Found: %s\n", dl->dl_lockres.l_name);
+		trace_ocfs2_dentry_attach_lock_found(dl->dl_lockres.l_name,
+				(unsigned long long)parent_blkno,
+				(unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 		goto out_attach;
 	}
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index 41b356fefaa..b0c17490bbb 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -43,7 +43,6 @@
 #include <linux/quotaops.h>
 #include <linux/sort.h>
 
-#define MLOG_MASK_PREFIX ML_NAMEI
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -61,6 +60,7 @@
 #include "super.h"
 #include "sysfile.h"
 #include "uptodate.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -369,8 +369,6 @@ static int inline ocfs2_search_dirblock(struct buffer_head *bh,
 	int de_len;
 	int ret = 0;
 
-	mlog_entry_void();
-
 	de_buf = first_de;
 	dlimit = de_buf + bytes;
 
@@ -404,7 +402,7 @@ static int inline ocfs2_search_dirblock(struct buffer_head *bh,
 	}
 
 bail:
-	mlog_exit(ret);
+	trace_ocfs2_search_dirblock(ret);
 	return ret;
 }
 
@@ -449,8 +447,7 @@ static int ocfs2_validate_dir_block(struct super_block *sb,
 	 * We don't validate dirents here, that's handled
 	 * in-place when the code walks them.
 	 */
-	mlog(0, "Validating dirblock %llu\n",
-	     (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_validate_dir_block((unsigned long long)bh->b_blocknr);
 
 	BUG_ON(!buffer_uptodate(bh));
 
@@ -708,8 +705,6 @@ static struct buffer_head *ocfs2_find_entry_el(const char *name, int namelen,
 	int num = 0;
 	int nblocks, i, err;
 
-	mlog_entry_void();
-
 	sb = dir->i_sb;
 
 	nblocks = i_size_read(dir) >> sb->s_blocksize_bits;
@@ -790,7 +785,7 @@ cleanup_and_exit:
 	for (; ra_ptr < ra_max; ra_ptr++)
 		brelse(bh_use[ra_ptr]);
 
-	mlog_exit_ptr(ret);
+	trace_ocfs2_find_entry_el(ret);
 	return ret;
 }
 
@@ -952,11 +947,9 @@ static int ocfs2_dx_dir_search(const char *name, int namelen,
 		goto out;
 	}
 
-	mlog(0, "Dir %llu: name: \"%.*s\", lookup of hash: %u.0x%x "
-	     "returns: %llu\n",
-	     (unsigned long long)OCFS2_I(dir)->ip_blkno,
-	     namelen, name, hinfo->major_hash, hinfo->minor_hash,
-	     (unsigned long long)phys);
+	trace_ocfs2_dx_dir_search((unsigned long long)OCFS2_I(dir)->ip_blkno,
+				  namelen, name, hinfo->major_hash,
+				  hinfo->minor_hash, (unsigned long long)phys);
 
 	ret = ocfs2_read_dx_leaf(dir, phys, &dx_leaf_bh);
 	if (ret) {
@@ -966,9 +959,9 @@ static int ocfs2_dx_dir_search(const char *name, int namelen,
 
 	dx_leaf = (struct ocfs2_dx_leaf *) dx_leaf_bh->b_data;
 
-	mlog(0, "leaf info: num_used: %d, count: %d\n",
-	     le16_to_cpu(dx_leaf->dl_list.de_num_used),
-	     le16_to_cpu(dx_leaf->dl_list.de_count));
+	trace_ocfs2_dx_dir_search_leaf_info(
+			le16_to_cpu(dx_leaf->dl_list.de_num_used),
+			le16_to_cpu(dx_leaf->dl_list.de_count));
 
 	entry_list = &dx_leaf->dl_list;
 
@@ -1168,8 +1161,6 @@ static int __ocfs2_delete_entry(handle_t *handle, struct inode *dir,
 	int i, status = -ENOENT;
 	ocfs2_journal_access_func access = ocfs2_journal_access_db;
 
-	mlog_entry("(0x%p, 0x%p, 0x%p, 0x%p)\n", handle, dir, de_del, bh);
-
 	if (OCFS2_I(dir)->ip_dyn_features & OCFS2_INLINE_DATA_FL)
 		access = ocfs2_journal_access_di;
 
@@ -1204,7 +1195,6 @@ static int __ocfs2_delete_entry(handle_t *handle, struct inode *dir,
 		de = (struct ocfs2_dir_entry *)((char *)de + le16_to_cpu(de->rec_len));
 	}
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -1350,8 +1340,8 @@ static int ocfs2_delete_entry_dx(handle_t *handle, struct inode *dir,
 		}
 	}
 
-	mlog(0, "Dir %llu: delete entry at index: %d\n",
-	     (unsigned long long)OCFS2_I(dir)->ip_blkno, index);
+	trace_ocfs2_delete_entry_dx((unsigned long long)OCFS2_I(dir)->ip_blkno,
+				    index);
 
 	ret = __ocfs2_delete_entry(handle, dir, lookup->dl_entry,
 				   leaf_bh, leaf_bh->b_data, leaf_bh->b_size);
@@ -1634,8 +1624,6 @@ int __ocfs2_add_entry(handle_t *handle,
 	struct buffer_head *insert_bh = lookup->dl_leaf_bh;
 	char *data_start = insert_bh->b_data;
 
-	mlog_entry_void();
-
 	if (!namelen)
 		return -EINVAL;
 
@@ -1767,8 +1755,9 @@ int __ocfs2_add_entry(handle_t *handle,
 	 * from ever getting here. */
 	retval = -ENOSPC;
 bail:
+	if (retval)
+		mlog_errno(retval);
 
-	mlog_exit(retval);
 	return retval;
 }
 
@@ -2030,8 +2019,7 @@ int ocfs2_readdir(struct file * filp, void * dirent, filldir_t filldir)
 	struct inode *inode = filp->f_path.dentry->d_inode;
 	int lock_level = 0;
 
-	mlog_entry("dirino=%llu\n",
-		   (unsigned long long)OCFS2_I(inode)->ip_blkno);
+	trace_ocfs2_readdir((unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 	error = ocfs2_inode_lock_atime(inode, filp->f_vfsmnt, &lock_level);
 	if (lock_level && error >= 0) {
@@ -2053,9 +2041,10 @@ int ocfs2_readdir(struct file * filp, void * dirent, filldir_t filldir)
 				      dirent, filldir, NULL);
 
 	ocfs2_inode_unlock(inode, lock_level);
+	if (error)
+		mlog_errno(error);
 
 bail_nolock:
-	mlog_exit(error);
 
 	return error;
 }
@@ -2071,8 +2060,8 @@ int ocfs2_find_files_on_disk(const char *name,
 {
 	int status = -ENOENT;
 
-	mlog(0, "name=%.*s, blkno=%p, inode=%llu\n", namelen, name, blkno,
-	     (unsigned long long)OCFS2_I(inode)->ip_blkno);
+	trace_ocfs2_find_files_on_disk(namelen, name, blkno,
+				(unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 	status = ocfs2_find_entry(name, namelen, inode, lookup);
 	if (status)
@@ -2116,8 +2105,8 @@ int ocfs2_check_dir_for_entry(struct inode *dir,
 	int ret;
 	struct ocfs2_dir_lookup_result lookup = { NULL, };
 
-	mlog_entry("dir %llu, name '%.*s'\n",
-		   (unsigned long long)OCFS2_I(dir)->ip_blkno, namelen, name);
+	trace_ocfs2_check_dir_for_entry(
+		(unsigned long long)OCFS2_I(dir)->ip_blkno, namelen, name);
 
 	ret = -EEXIST;
 	if (ocfs2_find_entry(name, namelen, dir, &lookup) == 0)
@@ -2127,7 +2116,8 @@ int ocfs2_check_dir_for_entry(struct inode *dir,
 bail:
 	ocfs2_free_dir_lookup_result(&lookup);
 
-	mlog_exit(ret);
+	if (ret)
+		mlog_errno(ret);
 	return ret;
 }
 
@@ -2326,8 +2316,6 @@ static int ocfs2_fill_new_dir_el(struct ocfs2_super *osb,
 	struct buffer_head *new_bh = NULL;
 	struct ocfs2_dir_entry *de;
 
-	mlog_entry_void();
-
 	if (ocfs2_new_dir_wants_trailer(inode))
 		size = ocfs2_dir_trailer_blk_off(parent->i_sb);
 
@@ -2382,7 +2370,6 @@ static int ocfs2_fill_new_dir_el(struct ocfs2_super *osb,
 bail:
 	brelse(new_bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -2411,9 +2398,9 @@ static int ocfs2_dx_dir_attach_index(struct ocfs2_super *osb,
 		goto out;
 	}
 
-	mlog(0, "Dir %llu, attach new index block: %llu\n",
-	     (unsigned long long)OCFS2_I(dir)->ip_blkno,
-	     (unsigned long long)dr_blkno);
+	trace_ocfs2_dx_dir_attach_index(
+				(unsigned long long)OCFS2_I(dir)->ip_blkno,
+				(unsigned long long)dr_blkno);
 
 	dx_root_bh = sb_getblk(osb->sb, dr_blkno);
 	if (dx_root_bh == NULL) {
@@ -2513,11 +2500,10 @@ static int ocfs2_dx_dir_format_cluster(struct ocfs2_super *osb,
 		dx_leaf->dl_list.de_count =
 			cpu_to_le16(ocfs2_dx_entries_per_leaf(osb->sb));
 
-		mlog(0,
-		     "Dir %llu, format dx_leaf: %llu, entry count: %u\n",
-		     (unsigned long long)OCFS2_I(dir)->ip_blkno,
-		     (unsigned long long)bh->b_blocknr,
-		     le16_to_cpu(dx_leaf->dl_list.de_count));
+		trace_ocfs2_dx_dir_format_cluster(
+				(unsigned long long)OCFS2_I(dir)->ip_blkno,
+				(unsigned long long)bh->b_blocknr,
+				le16_to_cpu(dx_leaf->dl_list.de_count));
 
 		ocfs2_journal_dirty(handle, bh);
 	}
@@ -2761,12 +2747,11 @@ static void ocfs2_dx_dir_index_root_block(struct inode *dir,
 
 		ocfs2_dx_dir_name_hash(dir, de->name, de->name_len, &hinfo);
 
-		mlog(0,
-		     "dir: %llu, major: 0x%x minor: 0x%x, index: %u, name: %.*s\n",
-		     (unsigned long long)dir->i_ino, hinfo.major_hash,
-		     hinfo.minor_hash,
-		     le16_to_cpu(dx_root->dr_entries.de_num_used),
-		     de->name_len, de->name);
+		trace_ocfs2_dx_dir_index_root_block(
+				(unsigned long long)dir->i_ino,
+				hinfo.major_hash, hinfo.minor_hash,
+				de->name_len, de->name,
+				le16_to_cpu(dx_root->dr_entries.de_num_used));
 
 		ocfs2_dx_entry_list_insert(&dx_root->dr_entries, &hinfo,
 					   dirent_blk);
@@ -3237,7 +3222,6 @@ static int ocfs2_do_extend_dir(struct super_block *sb,
 bail:
 	if (did_quota && status < 0)
 		dquot_free_space_nodirty(dir, ocfs2_clusters_to_bytes(sb, 1));
-	mlog_exit(status);
 	return status;
 }
 
@@ -3272,8 +3256,6 @@ static int ocfs2_extend_dir(struct ocfs2_super *osb,
 	struct ocfs2_extent_tree et;
 	struct buffer_head *dx_root_bh = lookup->dl_dx_root_bh;
 
-	mlog_entry_void();
-
 	if (OCFS2_I(dir)->ip_dyn_features & OCFS2_INLINE_DATA_FL) {
 		/*
 		 * This would be a code error as an inline directory should
@@ -3322,8 +3304,8 @@ static int ocfs2_extend_dir(struct ocfs2_super *osb,
 	down_write(&OCFS2_I(dir)->ip_alloc_sem);
 	drop_alloc_sem = 1;
 	dir_i_size = i_size_read(dir);
-	mlog(0, "extending dir %llu (i_size = %lld)\n",
-	     (unsigned long long)OCFS2_I(dir)->ip_blkno, dir_i_size);
+	trace_ocfs2_extend_dir((unsigned long long)OCFS2_I(dir)->ip_blkno,
+			       dir_i_size);
 
 	/* dir->i_size is always block aligned. */
 	spin_lock(&OCFS2_I(dir)->ip_lock);
@@ -3438,7 +3420,6 @@ bail:
 
 	brelse(new_bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -3585,8 +3566,9 @@ next:
 	status = 0;
 bail:
 	brelse(bh);
+	if (status)
+		mlog_errno(status);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -3817,9 +3799,9 @@ static int ocfs2_dx_dir_rebalance(struct ocfs2_super *osb, struct inode *dir,
 	struct ocfs2_dx_root_block *dx_root;
 	struct ocfs2_dx_leaf *tmp_dx_leaf = NULL;
 
-	mlog(0, "DX Dir: %llu, rebalance leaf leaf_blkno: %llu insert: %u\n",
-	     (unsigned long long)OCFS2_I(dir)->ip_blkno,
-	     (unsigned long long)leaf_blkno, insert_hash);
+	trace_ocfs2_dx_dir_rebalance((unsigned long long)OCFS2_I(dir)->ip_blkno,
+				     (unsigned long long)leaf_blkno,
+				     insert_hash);
 
 	ocfs2_init_dx_root_extent_tree(&et, INODE_CACHE(dir), dx_root_bh);
 
@@ -3899,8 +3881,7 @@ static int ocfs2_dx_dir_rebalance(struct ocfs2_super *osb, struct inode *dir,
 		goto  out_commit;
 	}
 
-	mlog(0, "Split leaf (%u) at %u, insert major hash is %u\n",
-	     leaf_cpos, split_hash, insert_hash);
+	trace_ocfs2_dx_dir_rebalance_split(leaf_cpos, split_hash, insert_hash);
 
 	/*
 	 * We have to carefully order operations here. There are items
@@ -4357,8 +4338,8 @@ int ocfs2_prepare_dir_for_insert(struct ocfs2_super *osb,
 	unsigned int blocks_wanted = 1;
 	struct buffer_head *bh = NULL;
 
-	mlog(0, "getting ready to insert namelen %d into dir %llu\n",
-	     namelen, (unsigned long long)OCFS2_I(dir)->ip_blkno);
+	trace_ocfs2_prepare_dir_for_insert(
+		(unsigned long long)OCFS2_I(dir)->ip_blkno, namelen);
 
 	if (!namelen) {
 		ret = -EINVAL;
diff --git a/fs/ocfs2/dlm/dlmconvert.c b/fs/ocfs2/dlm/dlmconvert.c
index 9f30491e5e8..29a886d1e82 100644
--- a/fs/ocfs2/dlm/dlmconvert.c
+++ b/fs/ocfs2/dlm/dlmconvert.c
@@ -128,8 +128,8 @@ static enum dlm_status __dlmconvert_master(struct dlm_ctxt *dlm,
 
 	assert_spin_locked(&res->spinlock);
 
-	mlog_entry("type=%d, convert_type=%d, new convert_type=%d\n",
-		   lock->ml.type, lock->ml.convert_type, type);
+	mlog(0, "type=%d, convert_type=%d, new convert_type=%d\n",
+	     lock->ml.type, lock->ml.convert_type, type);
 
 	spin_lock(&lock->spinlock);
 
@@ -353,7 +353,7 @@ static enum dlm_status dlm_send_remote_convert_request(struct dlm_ctxt *dlm,
 	struct kvec vec[2];
 	size_t veclen = 1;
 
-	mlog_entry("%.*s\n", res->lockname.len, res->lockname.name);
+	mlog(0, "%.*s\n", res->lockname.len, res->lockname.name);
 
 	memset(&convert, 0, sizeof(struct dlm_convert_lock));
 	convert.node_idx = dlm->node_num;
diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index 99805d578ab..7540a492eab 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -188,7 +188,7 @@ struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm,
 	struct hlist_head *bucket;
 	struct hlist_node *list;
 
-	mlog_entry("%.*s\n", len, name);
+	mlog(0, "%.*s\n", len, name);
 
 	assert_spin_locked(&dlm->spinlock);
 
@@ -222,7 +222,7 @@ struct dlm_lock_resource * __dlm_lookup_lockres(struct dlm_ctxt *dlm,
 {
 	struct dlm_lock_resource *res = NULL;
 
-	mlog_entry("%.*s\n", len, name);
+	mlog(0, "%.*s\n", len, name);
 
 	assert_spin_locked(&dlm->spinlock);
 
@@ -531,7 +531,7 @@ static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data,
 	unsigned int node;
 	struct dlm_exit_domain *exit_msg = (struct dlm_exit_domain *) msg->buf;
 
-	mlog_entry("%p %u %p", msg, len, data);
+	mlog(0, "%p %u %p", msg, len, data);
 
 	if (!dlm_grab(dlm))
 		return 0;
@@ -1557,7 +1557,7 @@ static int dlm_try_to_join_domain(struct dlm_ctxt *dlm)
 	struct domain_join_ctxt *ctxt;
 	enum dlm_query_join_response_code response = JOIN_DISALLOW;
 
-	mlog_entry("%p", dlm);
+	mlog(0, "%p", dlm);
 
 	ctxt = kzalloc(sizeof(*ctxt), GFP_KERNEL);
 	if (!ctxt) {
diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c
index 7009292aac5..8d39e0fd66f 100644
--- a/fs/ocfs2/dlm/dlmlock.c
+++ b/fs/ocfs2/dlm/dlmlock.c
@@ -128,7 +128,7 @@ static enum dlm_status dlmlock_master(struct dlm_ctxt *dlm,
 	int call_ast = 0, kick_thread = 0;
 	enum dlm_status status = DLM_NORMAL;
 
-	mlog_entry("type=%d\n", lock->ml.type);
+	mlog(0, "type=%d\n", lock->ml.type);
 
 	spin_lock(&res->spinlock);
 	/* if called from dlm_create_lock_handler, need to
@@ -227,8 +227,8 @@ static enum dlm_status dlmlock_remote(struct dlm_ctxt *dlm,
 	enum dlm_status status = DLM_DENIED;
 	int lockres_changed = 1;
 
-	mlog_entry("type=%d\n", lock->ml.type);
-	mlog(0, "lockres %.*s, flags = 0x%x\n", res->lockname.len,
+	mlog(0, "type=%d, lockres %.*s, flags = 0x%x\n",
+	     lock->ml.type, res->lockname.len,
 	     res->lockname.name, flags);
 
 	spin_lock(&res->spinlock);
@@ -308,8 +308,6 @@ static enum dlm_status dlm_send_remote_lock_request(struct dlm_ctxt *dlm,
 	int tmpret, status = 0;
 	enum dlm_status ret;
 
-	mlog_entry_void();
-
 	memset(&create, 0, sizeof(create));
 	create.node_idx = dlm->node_num;
 	create.requested_type = lock->ml.type;
@@ -477,8 +475,6 @@ int dlm_create_lock_handler(struct o2net_msg *msg, u32 len, void *data,
 
 	BUG_ON(!dlm);
 
-	mlog_entry_void();
-
 	if (!dlm_grab(dlm))
 		return DLM_REJECTED;
 
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index 59f0f6bdfc6..9d67610dfc7 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -426,8 +426,6 @@ static void dlm_mle_release(struct kref *kref)
 	struct dlm_master_list_entry *mle;
 	struct dlm_ctxt *dlm;
 
-	mlog_entry_void();
-
 	mle = container_of(kref, struct dlm_master_list_entry, mle_refs);
 	dlm = mle->dlm;
 
@@ -3120,8 +3118,6 @@ static int dlm_add_migration_mle(struct dlm_ctxt *dlm,
 
 	*oldmle = NULL;
 
-	mlog_entry_void();
-
 	assert_spin_locked(&dlm->spinlock);
 	assert_spin_locked(&dlm->master_lock);
 
@@ -3261,7 +3257,7 @@ void dlm_clean_master_list(struct dlm_ctxt *dlm, u8 dead_node)
 	struct hlist_node *list;
 	unsigned int i;
 
-	mlog_entry("dlm=%s, dead node=%u\n", dlm->name, dead_node);
+	mlog(0, "dlm=%s, dead node=%u\n", dlm->name, dead_node);
 top:
 	assert_spin_locked(&dlm->spinlock);
 
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index aaaffbcbe91..f1beb6fc254 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -727,7 +727,6 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node)
 	if (destroy)
 		dlm_destroy_recovery_area(dlm, dead_node);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1496,9 +1495,9 @@ leave:
 			kfree(buf);
 		if (item)
 			kfree(item);
+		mlog_errno(ret);
 	}
 
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -1567,7 +1566,6 @@ leave:
 		dlm_lockres_put(res);
 	}
 	kfree(data);
-	mlog_exit(ret);
 }
 
 
@@ -1986,7 +1984,6 @@ leave:
 			dlm_lock_put(newlock);
 	}
 
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -2083,8 +2080,6 @@ static void dlm_finish_local_lockres_recovery(struct dlm_ctxt *dlm,
 	struct hlist_head *bucket;
 	struct dlm_lock_resource *res, *next;
 
-	mlog_entry_void();
-
 	assert_spin_locked(&dlm->spinlock);
 
 	list_for_each_entry_safe(res, next, &dlm->reco.resources, recovering) {
@@ -2607,8 +2602,6 @@ static int dlm_send_begin_reco_message(struct dlm_ctxt *dlm, u8 dead_node)
 	int nodenum;
 	int status;
 
-	mlog_entry("%u\n", dead_node);
-
 	mlog(0, "%s: dead node is %u\n", dlm->name, dead_node);
 
 	spin_lock(&dlm->spinlock);
diff --git a/fs/ocfs2/dlm/dlmunlock.c b/fs/ocfs2/dlm/dlmunlock.c
index 817287c6a6d..850aa7e8753 100644
--- a/fs/ocfs2/dlm/dlmunlock.c
+++ b/fs/ocfs2/dlm/dlmunlock.c
@@ -317,7 +317,7 @@ static enum dlm_status dlm_send_remote_unlock_request(struct dlm_ctxt *dlm,
 	struct kvec vec[2];
 	size_t veclen = 1;
 
-	mlog_entry("%.*s\n", res->lockname.len, res->lockname.name);
+	mlog(0, "%.*s\n", res->lockname.len, res->lockname.name);
 
 	if (owner == dlm->node_num) {
 		/* ended up trying to contact ourself.  this means
@@ -588,8 +588,6 @@ enum dlm_status dlmunlock(struct dlm_ctxt *dlm, struct dlm_lockstatus *lksb,
 	struct dlm_lock *lock = NULL;
 	int call_ast, is_master;
 
-	mlog_entry_void();
-
 	if (!lksb) {
 		dlm_error(DLM_BADARGS);
 		return DLM_BADARGS;
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 7484bb9a343..7642d7ca73e 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -397,8 +397,6 @@ static void ocfs2_build_lock_name(enum ocfs2_lock_type type,
 {
 	int len;
 
-	mlog_entry_void();
-
 	BUG_ON(type >= OCFS2_NUM_LOCK_TYPES);
 
 	len = snprintf(name, OCFS2_LOCK_ID_MAX_LEN, "%c%s%016llx%08x",
@@ -408,8 +406,6 @@ static void ocfs2_build_lock_name(enum ocfs2_lock_type type,
 	BUG_ON(len != (OCFS2_LOCK_ID_MAX_LEN - 1));
 
 	mlog(0, "built lock resource with name: %s\n", name);
-
-	mlog_exit_void();
 }
 
 static DEFINE_SPINLOCK(ocfs2_dlm_tracking_lock);
@@ -725,8 +721,6 @@ void ocfs2_refcount_lock_res_init(struct ocfs2_lock_res *lockres,
 
 void ocfs2_lock_res_free(struct ocfs2_lock_res *res)
 {
-	mlog_entry_void();
-
 	if (!(res->l_flags & OCFS2_LOCK_INITIALIZED))
 		return;
 
@@ -752,14 +746,11 @@ void ocfs2_lock_res_free(struct ocfs2_lock_res *res)
 	memset(&res->l_lksb, 0, sizeof(res->l_lksb));
 
 	res->l_flags = 0UL;
-	mlog_exit_void();
 }
 
 static inline void ocfs2_inc_holders(struct ocfs2_lock_res *lockres,
 				     int level)
 {
-	mlog_entry_void();
-
 	BUG_ON(!lockres);
 
 	switch(level) {
@@ -772,15 +763,11 @@ static inline void ocfs2_inc_holders(struct ocfs2_lock_res *lockres,
 	default:
 		BUG();
 	}
-
-	mlog_exit_void();
 }
 
 static inline void ocfs2_dec_holders(struct ocfs2_lock_res *lockres,
 				     int level)
 {
-	mlog_entry_void();
-
 	BUG_ON(!lockres);
 
 	switch(level) {
@@ -795,7 +782,6 @@ static inline void ocfs2_dec_holders(struct ocfs2_lock_res *lockres,
 	default:
 		BUG();
 	}
-	mlog_exit_void();
 }
 
 /* WARNING: This function lives in a world where the only three lock
@@ -842,8 +828,6 @@ static void lockres_clear_flags(struct ocfs2_lock_res *lockres,
 
 static inline void ocfs2_generic_handle_downconvert_action(struct ocfs2_lock_res *lockres)
 {
-	mlog_entry_void();
-
 	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BUSY));
 	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_ATTACHED));
 	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BLOCKED));
@@ -856,14 +840,10 @@ static inline void ocfs2_generic_handle_downconvert_action(struct ocfs2_lock_res
 		lockres_clear_flags(lockres, OCFS2_LOCK_BLOCKED);
 	}
 	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
-
-	mlog_exit_void();
 }
 
 static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lockres)
 {
-	mlog_entry_void();
-
 	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_BUSY));
 	BUG_ON(!(lockres->l_flags & OCFS2_LOCK_ATTACHED));
 
@@ -885,14 +865,10 @@ static inline void ocfs2_generic_handle_convert_action(struct ocfs2_lock_res *lo
 	lockres_or_flags(lockres, OCFS2_LOCK_UPCONVERT_FINISHING);
 
 	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
-
-	mlog_exit_void();
 }
 
 static inline void ocfs2_generic_handle_attach_action(struct ocfs2_lock_res *lockres)
 {
-	mlog_entry_void();
-
 	BUG_ON((!(lockres->l_flags & OCFS2_LOCK_BUSY)));
 	BUG_ON(lockres->l_flags & OCFS2_LOCK_ATTACHED);
 
@@ -904,15 +880,12 @@ static inline void ocfs2_generic_handle_attach_action(struct ocfs2_lock_res *loc
 	lockres->l_level = lockres->l_requested;
 	lockres_or_flags(lockres, OCFS2_LOCK_ATTACHED);
 	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
-
-	mlog_exit_void();
 }
 
 static int ocfs2_generic_handle_bast(struct ocfs2_lock_res *lockres,
 				     int level)
 {
 	int needs_downconvert = 0;
-	mlog_entry_void();
 
 	assert_spin_locked(&lockres->l_lock);
 
@@ -934,8 +907,7 @@ static int ocfs2_generic_handle_bast(struct ocfs2_lock_res *lockres,
 
 	if (needs_downconvert)
 		lockres_or_flags(lockres, OCFS2_LOCK_BLOCKED);
-
-	mlog_exit(needs_downconvert);
+	mlog(0, "needs_downconvert = %d\n", needs_downconvert);
 	return needs_downconvert;
 }
 
@@ -1147,8 +1119,6 @@ static void ocfs2_unlock_ast(struct ocfs2_dlm_lksb *lksb, int error)
 	struct ocfs2_lock_res *lockres = ocfs2_lksb_to_lock_res(lksb);
 	unsigned long flags;
 
-	mlog_entry_void();
-
 	mlog(ML_BASTS, "UNLOCK AST fired for lockres %s, action = %d\n",
 	     lockres->l_name, lockres->l_unlock_action);
 
@@ -1158,7 +1128,6 @@ static void ocfs2_unlock_ast(struct ocfs2_dlm_lksb *lksb, int error)
 		     "unlock_action %d\n", error, lockres->l_name,
 		     lockres->l_unlock_action);
 		spin_unlock_irqrestore(&lockres->l_lock, flags);
-		mlog_exit_void();
 		return;
 	}
 
@@ -1182,8 +1151,6 @@ static void ocfs2_unlock_ast(struct ocfs2_dlm_lksb *lksb, int error)
 	lockres->l_unlock_action = OCFS2_UNLOCK_INVALID;
 	wake_up(&lockres->l_event);
 	spin_unlock_irqrestore(&lockres->l_lock, flags);
-
-	mlog_exit_void();
 }
 
 /*
@@ -1229,7 +1196,6 @@ static inline void ocfs2_recover_from_dlm_error(struct ocfs2_lock_res *lockres,
 {
 	unsigned long flags;
 
-	mlog_entry_void();
 	spin_lock_irqsave(&lockres->l_lock, flags);
 	lockres_clear_flags(lockres, OCFS2_LOCK_BUSY);
 	lockres_clear_flags(lockres, OCFS2_LOCK_UPCONVERT_FINISHING);
@@ -1240,7 +1206,6 @@ static inline void ocfs2_recover_from_dlm_error(struct ocfs2_lock_res *lockres,
 	spin_unlock_irqrestore(&lockres->l_lock, flags);
 
 	wake_up(&lockres->l_event);
-	mlog_exit_void();
 }
 
 /* Note: If we detect another process working on the lock (i.e.,
@@ -1256,8 +1221,6 @@ static int ocfs2_lock_create(struct ocfs2_super *osb,
 	unsigned long flags;
 	unsigned int gen;
 
-	mlog_entry_void();
-
 	mlog(0, "lock %s, level = %d, flags = %u\n", lockres->l_name, level,
 	     dlm_flags);
 
@@ -1289,7 +1252,6 @@ static int ocfs2_lock_create(struct ocfs2_super *osb,
 	mlog(0, "lock %s, return from ocfs2_dlm_lock\n", lockres->l_name);
 
 bail:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -1412,8 +1374,6 @@ static int __ocfs2_cluster_lock(struct ocfs2_super *osb,
 	unsigned int gen;
 	int noqueue_attempted = 0;
 
-	mlog_entry_void();
-
 	ocfs2_init_mask_waiter(&mw);
 
 	if (lockres->l_ops->flags & LOCK_TYPE_USES_LVB)
@@ -1579,7 +1539,6 @@ out:
 				caller_ip);
 	}
 #endif
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -1601,7 +1560,6 @@ static void __ocfs2_cluster_unlock(struct ocfs2_super *osb,
 {
 	unsigned long flags;
 
-	mlog_entry_void();
 	spin_lock_irqsave(&lockres->l_lock, flags);
 	ocfs2_dec_holders(lockres, level);
 	ocfs2_downconvert_on_unlock(osb, lockres);
@@ -1610,7 +1568,6 @@ static void __ocfs2_cluster_unlock(struct ocfs2_super *osb,
 	if (lockres->l_lockdep_map.key != NULL)
 		rwsem_release(&lockres->l_lockdep_map, 1, caller_ip);
 #endif
-	mlog_exit_void();
 }
 
 static int ocfs2_create_new_lock(struct ocfs2_super *osb,
@@ -1644,8 +1601,6 @@ int ocfs2_create_new_inode_locks(struct inode *inode)
 	BUG_ON(!inode);
 	BUG_ON(!ocfs2_inode_is_new(inode));
 
-	mlog_entry_void();
-
 	mlog(0, "Inode %llu\n", (unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 	/* NOTE: That we don't increment any of the holder counts, nor
@@ -1679,7 +1634,6 @@ int ocfs2_create_new_inode_locks(struct inode *inode)
 	}
 
 bail:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -1691,16 +1645,12 @@ int ocfs2_rw_lock(struct inode *inode, int write)
 
 	BUG_ON(!inode);
 
-	mlog_entry_void();
-
 	mlog(0, "inode %llu take %s RW lock\n",
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
 	     write ? "EXMODE" : "PRMODE");
 
-	if (ocfs2_mount_local(osb)) {
-		mlog_exit(0);
+	if (ocfs2_mount_local(osb))
 		return 0;
-	}
 
 	lockres = &OCFS2_I(inode)->ip_rw_lockres;
 
@@ -1711,7 +1661,6 @@ int ocfs2_rw_lock(struct inode *inode, int write)
 	if (status < 0)
 		mlog_errno(status);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1721,16 +1670,12 @@ void ocfs2_rw_unlock(struct inode *inode, int write)
 	struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_rw_lockres;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
-	mlog_entry_void();
-
 	mlog(0, "inode %llu drop %s RW lock\n",
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
 	     write ? "EXMODE" : "PRMODE");
 
 	if (!ocfs2_mount_local(osb))
 		ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level);
-
-	mlog_exit_void();
 }
 
 /*
@@ -1744,8 +1689,6 @@ int ocfs2_open_lock(struct inode *inode)
 
 	BUG_ON(!inode);
 
-	mlog_entry_void();
-
 	mlog(0, "inode %llu take PRMODE open lock\n",
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno);
 
@@ -1760,7 +1703,6 @@ int ocfs2_open_lock(struct inode *inode)
 		mlog_errno(status);
 
 out:
-	mlog_exit(status);
 	return status;
 }
 
@@ -1772,8 +1714,6 @@ int ocfs2_try_open_lock(struct inode *inode, int write)
 
 	BUG_ON(!inode);
 
-	mlog_entry_void();
-
 	mlog(0, "inode %llu try to take %s open lock\n",
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
 	     write ? "EXMODE" : "PRMODE");
@@ -1795,7 +1735,6 @@ int ocfs2_try_open_lock(struct inode *inode, int write)
 				    level, DLM_LKF_NOQUEUE, 0);
 
 out:
-	mlog_exit(status);
 	return status;
 }
 
@@ -1807,8 +1746,6 @@ void ocfs2_open_unlock(struct inode *inode)
 	struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_open_lockres;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
-	mlog_entry_void();
-
 	mlog(0, "inode %llu drop open lock\n",
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno);
 
@@ -1823,7 +1760,7 @@ void ocfs2_open_unlock(struct inode *inode)
 				     DLM_LOCK_EX);
 
 out:
-	mlog_exit_void();
+	return;
 }
 
 static int ocfs2_flock_handle_signal(struct ocfs2_lock_res *lockres,
@@ -2039,8 +1976,6 @@ static void ocfs2_downconvert_on_unlock(struct ocfs2_super *osb,
 {
 	int kick = 0;
 
-	mlog_entry_void();
-
 	/* If we know that another node is waiting on our lock, kick
 	 * the downconvert thread * pre-emptively when we reach a release
 	 * condition. */
@@ -2061,8 +1996,6 @@ static void ocfs2_downconvert_on_unlock(struct ocfs2_super *osb,
 
 	if (kick)
 		ocfs2_wake_downconvert_thread(osb);
-
-	mlog_exit_void();
 }
 
 #define OCFS2_SEC_BITS   34
@@ -2091,8 +2024,6 @@ static void __ocfs2_stuff_meta_lvb(struct inode *inode)
 	struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres;
 	struct ocfs2_meta_lvb *lvb;
 
-	mlog_entry_void();
-
 	lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
 
 	/*
@@ -2124,8 +2055,6 @@ static void __ocfs2_stuff_meta_lvb(struct inode *inode)
 
 out:
 	mlog_meta_lvb(0, lockres);
-
-	mlog_exit_void();
 }
 
 static void ocfs2_unpack_timespec(struct timespec *spec,
@@ -2141,8 +2070,6 @@ static void ocfs2_refresh_inode_from_lvb(struct inode *inode)
 	struct ocfs2_lock_res *lockres = &oi->ip_inode_lockres;
 	struct ocfs2_meta_lvb *lvb;
 
-	mlog_entry_void();
-
 	mlog_meta_lvb(0, lockres);
 
 	lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
@@ -2173,8 +2100,6 @@ static void ocfs2_refresh_inode_from_lvb(struct inode *inode)
 	ocfs2_unpack_timespec(&inode->i_ctime,
 			      be64_to_cpu(lvb->lvb_ictime_packed));
 	spin_unlock(&oi->ip_lock);
-
-	mlog_exit_void();
 }
 
 static inline int ocfs2_meta_lvb_is_trustable(struct inode *inode,
@@ -2201,8 +2126,6 @@ static int ocfs2_should_refresh_lock_res(struct ocfs2_lock_res *lockres)
 	unsigned long flags;
 	int status = 0;
 
-	mlog_entry_void();
-
 refresh_check:
 	spin_lock_irqsave(&lockres->l_lock, flags);
 	if (!(lockres->l_flags & OCFS2_LOCK_NEEDS_REFRESH)) {
@@ -2223,7 +2146,7 @@ refresh_check:
 
 	status = 1;
 bail:
-	mlog_exit(status);
+	mlog(0, "status %d\n", status);
 	return status;
 }
 
@@ -2233,7 +2156,6 @@ static inline void ocfs2_complete_lock_res_refresh(struct ocfs2_lock_res *lockre
 						   int status)
 {
 	unsigned long flags;
-	mlog_entry_void();
 
 	spin_lock_irqsave(&lockres->l_lock, flags);
 	lockres_clear_flags(lockres, OCFS2_LOCK_REFRESHING);
@@ -2242,8 +2164,6 @@ static inline void ocfs2_complete_lock_res_refresh(struct ocfs2_lock_res *lockre
 	spin_unlock_irqrestore(&lockres->l_lock, flags);
 
 	wake_up(&lockres->l_event);
-
-	mlog_exit_void();
 }
 
 /* may or may not return a bh if it went to disk. */
@@ -2256,8 +2176,6 @@ static int ocfs2_inode_lock_update(struct inode *inode,
 	struct ocfs2_dinode *fe;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
-	mlog_entry_void();
-
 	if (ocfs2_mount_local(osb))
 		goto bail;
 
@@ -2326,7 +2244,6 @@ static int ocfs2_inode_lock_update(struct inode *inode,
 bail_refresh:
 	ocfs2_complete_lock_res_refresh(lockres, status);
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -2370,8 +2287,6 @@ int ocfs2_inode_lock_full_nested(struct inode *inode,
 
 	BUG_ON(!inode);
 
-	mlog_entry_void();
-
 	mlog(0, "inode %llu, take %s META lock\n",
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
 	     ex ? "EXMODE" : "PRMODE");
@@ -2463,7 +2378,6 @@ bail:
 	if (local_bh)
 		brelse(local_bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -2513,7 +2427,6 @@ int ocfs2_inode_lock_atime(struct inode *inode,
 {
 	int ret;
 
-	mlog_entry_void();
 	ret = ocfs2_inode_lock(inode, NULL, 0);
 	if (ret < 0) {
 		mlog_errno(ret);
@@ -2541,7 +2454,6 @@ int ocfs2_inode_lock_atime(struct inode *inode,
 	} else
 		*level = 0;
 
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -2552,8 +2464,6 @@ void ocfs2_inode_unlock(struct inode *inode,
 	struct ocfs2_lock_res *lockres = &OCFS2_I(inode)->ip_inode_lockres;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
-	mlog_entry_void();
-
 	mlog(0, "inode %llu drop %s META lock\n",
 	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
 	     ex ? "EXMODE" : "PRMODE");
@@ -2561,8 +2471,6 @@ void ocfs2_inode_unlock(struct inode *inode,
 	if (!ocfs2_is_hard_readonly(OCFS2_SB(inode->i_sb)) &&
 	    !ocfs2_mount_local(osb))
 		ocfs2_cluster_unlock(OCFS2_SB(inode->i_sb), lockres, level);
-
-	mlog_exit_void();
 }
 
 int ocfs2_orphan_scan_lock(struct ocfs2_super *osb, u32 *seqno)
@@ -2613,8 +2521,6 @@ int ocfs2_super_lock(struct ocfs2_super *osb,
 	int level = ex ? DLM_LOCK_EX : DLM_LOCK_PR;
 	struct ocfs2_lock_res *lockres = &osb->osb_super_lockres;
 
-	mlog_entry_void();
-
 	if (ocfs2_is_hard_readonly(osb))
 		return -EROFS;
 
@@ -2646,7 +2552,6 @@ int ocfs2_super_lock(struct ocfs2_super *osb,
 		ocfs2_track_lock_refresh(lockres);
 	}
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -3057,8 +2962,6 @@ int ocfs2_dlm_init(struct ocfs2_super *osb)
 	int status = 0;
 	struct ocfs2_cluster_connection *conn = NULL;
 
-	mlog_entry_void();
-
 	if (ocfs2_mount_local(osb)) {
 		osb->node_num = 0;
 		goto local;
@@ -3115,15 +3018,12 @@ bail:
 			kthread_stop(osb->dc_task);
 	}
 
-	mlog_exit(status);
 	return status;
 }
 
 void ocfs2_dlm_shutdown(struct ocfs2_super *osb,
 			int hangup_pending)
 {
-	mlog_entry_void();
-
 	ocfs2_drop_osb_locks(osb);
 
 	/*
@@ -3146,8 +3046,6 @@ void ocfs2_dlm_shutdown(struct ocfs2_super *osb,
 	osb->cconn = NULL;
 
 	ocfs2_dlm_shutdown_debug(osb);
-
-	mlog_exit_void();
 }
 
 static int ocfs2_drop_lock(struct ocfs2_super *osb,
@@ -3229,7 +3127,6 @@ static int ocfs2_drop_lock(struct ocfs2_super *osb,
 
 	ocfs2_wait_on_busy_lock(lockres);
 out:
-	mlog_exit(0);
 	return 0;
 }
 
@@ -3287,8 +3184,6 @@ int ocfs2_drop_inode_locks(struct inode *inode)
 {
 	int status, err;
 
-	mlog_entry_void();
-
 	/* No need to call ocfs2_mark_lockres_freeing here -
 	 * ocfs2_clear_inode has done it for us. */
 
@@ -3313,7 +3208,6 @@ int ocfs2_drop_inode_locks(struct inode *inode)
 	if (err < 0 && !status)
 		status = err;
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -3355,8 +3249,6 @@ static int ocfs2_downconvert_lock(struct ocfs2_super *osb,
 	int ret;
 	u32 dlm_flags = DLM_LKF_CONVERT;
 
-	mlog_entry_void();
-
 	mlog(ML_BASTS, "lockres %s, level %d => %d\n", lockres->l_name,
 	     lockres->l_level, new_level);
 
@@ -3378,7 +3270,6 @@ static int ocfs2_downconvert_lock(struct ocfs2_super *osb,
 
 	ret = 0;
 bail:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -3388,8 +3279,6 @@ static int ocfs2_prepare_cancel_convert(struct ocfs2_super *osb,
 {
 	assert_spin_locked(&lockres->l_lock);
 
-	mlog_entry_void();
-
 	if (lockres->l_unlock_action == OCFS2_UNLOCK_CANCEL_CONVERT) {
 		/* If we're already trying to cancel a lock conversion
 		 * then just drop the spinlock and allow the caller to
@@ -3419,8 +3308,6 @@ static int ocfs2_cancel_convert(struct ocfs2_super *osb,
 {
 	int ret;
 
-	mlog_entry_void();
-
 	ret = ocfs2_dlm_unlock(osb->cconn, &lockres->l_lksb,
 			       DLM_LKF_CANCEL);
 	if (ret) {
@@ -3430,7 +3317,6 @@ static int ocfs2_cancel_convert(struct ocfs2_super *osb,
 
 	mlog(ML_BASTS, "lockres %s\n", lockres->l_name);
 
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -3446,8 +3332,6 @@ static int ocfs2_unblock_lock(struct ocfs2_super *osb,
 	int set_lvb = 0;
 	unsigned int gen;
 
-	mlog_entry_void();
-
 	spin_lock_irqsave(&lockres->l_lock, flags);
 
 recheck:
@@ -3622,14 +3506,14 @@ downconvert:
 				     gen);
 
 leave:
-	mlog_exit(ret);
+	if (ret)
+		mlog_errno(ret);
 	return ret;
 
 leave_requeue:
 	spin_unlock_irqrestore(&lockres->l_lock, flags);
 	ctl->requeue = 1;
 
-	mlog_exit(0);
 	return 0;
 }
 
@@ -3862,8 +3746,6 @@ static void ocfs2_set_qinfo_lvb(struct ocfs2_lock_res *lockres)
 	struct mem_dqinfo *info = sb_dqinfo(oinfo->dqi_gi.dqi_sb,
 					    oinfo->dqi_gi.dqi_type);
 
-	mlog_entry_void();
-
 	lvb = ocfs2_dlm_lvb(&lockres->l_lksb);
 	lvb->lvb_version = OCFS2_QINFO_LVB_VERSION;
 	lvb->lvb_bgrace = cpu_to_be32(info->dqi_bgrace);
@@ -3872,8 +3754,6 @@ static void ocfs2_set_qinfo_lvb(struct ocfs2_lock_res *lockres)
 	lvb->lvb_blocks = cpu_to_be32(oinfo->dqi_gi.dqi_blocks);
 	lvb->lvb_free_blk = cpu_to_be32(oinfo->dqi_gi.dqi_free_blk);
 	lvb->lvb_free_entry = cpu_to_be32(oinfo->dqi_gi.dqi_free_entry);
-
-	mlog_exit_void();
 }
 
 void ocfs2_qinfo_unlock(struct ocfs2_mem_dqinfo *oinfo, int ex)
@@ -3882,10 +3762,8 @@ void ocfs2_qinfo_unlock(struct ocfs2_mem_dqinfo *oinfo, int ex)
 	struct ocfs2_super *osb = OCFS2_SB(oinfo->dqi_gi.dqi_sb);
 	int level = ex ? DLM_LOCK_EX : DLM_LOCK_PR;
 
-	mlog_entry_void();
 	if (!ocfs2_is_hard_readonly(osb) && !ocfs2_mount_local(osb))
 		ocfs2_cluster_unlock(osb, lockres, level);
-	mlog_exit_void();
 }
 
 static int ocfs2_refresh_qinfo(struct ocfs2_mem_dqinfo *oinfo)
@@ -3940,8 +3818,6 @@ int ocfs2_qinfo_lock(struct ocfs2_mem_dqinfo *oinfo, int ex)
 	int level = ex ? DLM_LOCK_EX : DLM_LOCK_PR;
 	int status = 0;
 
-	mlog_entry_void();
-
 	/* On RO devices, locking really isn't needed... */
 	if (ocfs2_is_hard_readonly(osb)) {
 		if (ex)
@@ -3964,7 +3840,6 @@ int ocfs2_qinfo_lock(struct ocfs2_mem_dqinfo *oinfo, int ex)
 		ocfs2_qinfo_unlock(oinfo, ex);
 	ocfs2_complete_lock_res_refresh(lockres, status);
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -4010,8 +3885,6 @@ static void ocfs2_process_blocked_lock(struct ocfs2_super *osb,
 	 * considered valid until we remove the OCFS2_LOCK_QUEUED
 	 * flag. */
 
-	mlog_entry_void();
-
 	BUG_ON(!lockres);
 	BUG_ON(!lockres->l_ops);
 
@@ -4045,15 +3918,11 @@ unqueue:
 	if (ctl.unblock_action != UNBLOCK_CONTINUE
 	    && lockres->l_ops->post_unlock)
 		lockres->l_ops->post_unlock(osb, lockres);
-
-	mlog_exit_void();
 }
 
 static void ocfs2_schedule_blocked_lock(struct ocfs2_super *osb,
 					struct ocfs2_lock_res *lockres)
 {
-	mlog_entry_void();
-
 	assert_spin_locked(&lockres->l_lock);
 
 	if (lockres->l_flags & OCFS2_LOCK_FREEING) {
@@ -4074,8 +3943,6 @@ static void ocfs2_schedule_blocked_lock(struct ocfs2_super *osb,
 		osb->blocked_lock_count++;
 	}
 	spin_unlock(&osb->dc_task_lock);
-
-	mlog_exit_void();
 }
 
 static void ocfs2_downconvert_thread_do_work(struct ocfs2_super *osb)
@@ -4083,8 +3950,6 @@ static void ocfs2_downconvert_thread_do_work(struct ocfs2_super *osb)
 	unsigned long processed;
 	struct ocfs2_lock_res *lockres;
 
-	mlog_entry_void();
-
 	spin_lock(&osb->dc_task_lock);
 	/* grab this early so we know to try again if a state change and
 	 * wake happens part-way through our work  */
@@ -4108,8 +3973,6 @@ static void ocfs2_downconvert_thread_do_work(struct ocfs2_super *osb)
 		spin_lock(&osb->dc_task_lock);
 	}
 	spin_unlock(&osb->dc_task_lock);
-
-	mlog_exit_void();
 }
 
 static int ocfs2_downconvert_thread_lists_empty(struct ocfs2_super *osb)
diff --git a/fs/ocfs2/export.c b/fs/ocfs2/export.c
index 5dbc3062b4f..745db42528d 100644
--- a/fs/ocfs2/export.c
+++ b/fs/ocfs2/export.c
@@ -26,7 +26,6 @@
 #include <linux/fs.h>
 #include <linux/types.h>
 
-#define MLOG_MASK_PREFIX ML_EXPORT
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -40,6 +39,7 @@
 
 #include "buffer_head_io.h"
 #include "suballoc.h"
+#include "ocfs2_trace.h"
 
 struct ocfs2_inode_handle
 {
@@ -56,10 +56,9 @@ static struct dentry *ocfs2_get_dentry(struct super_block *sb,
 	int status, set;
 	struct dentry *result;
 
-	mlog_entry("(0x%p, 0x%p)\n", sb, handle);
+	trace_ocfs2_get_dentry_begin(sb, handle, (unsigned long long)blkno);
 
 	if (blkno == 0) {
-		mlog(0, "nfs wants inode with blkno: 0\n");
 		result = ERR_PTR(-ESTALE);
 		goto bail;
 	}
@@ -83,6 +82,7 @@ static struct dentry *ocfs2_get_dentry(struct super_block *sb,
 	}
 
 	status = ocfs2_test_inode_bit(osb, blkno, &set);
+	trace_ocfs2_get_dentry_test_bit(status, set);
 	if (status < 0) {
 		if (status == -EINVAL) {
 			/*
@@ -90,18 +90,14 @@ static struct dentry *ocfs2_get_dentry(struct super_block *sb,
 			 * as an inode, we return -ESTALE to be
 			 * nice
 			 */
-			mlog(0, "test inode bit failed %d\n", status);
 			status = -ESTALE;
-		} else {
+		} else
 			mlog(ML_ERROR, "test inode bit failed %d\n", status);
-		}
 		goto unlock_nfs_sync;
 	}
 
 	/* If the inode allocator bit is clear, this inode must be stale */
 	if (!set) {
-		mlog(0, "inode %llu suballoc bit is clear\n",
-		     (unsigned long long)blkno);
 		status = -ESTALE;
 		goto unlock_nfs_sync;
 	}
@@ -114,8 +110,8 @@ unlock_nfs_sync:
 check_err:
 	if (status < 0) {
 		if (status == -ESTALE) {
-			mlog(0, "stale inode ino: %llu generation: %u\n",
-			     (unsigned long long)blkno, handle->ih_generation);
+			trace_ocfs2_get_dentry_stale((unsigned long long)blkno,
+						     handle->ih_generation);
 		}
 		result = ERR_PTR(status);
 		goto bail;
@@ -130,8 +126,9 @@ check_err:
 check_gen:
 	if (handle->ih_generation != inode->i_generation) {
 		iput(inode);
-		mlog(0, "stale inode ino: %llu generation: %u\n",
-		     (unsigned long long)blkno, handle->ih_generation);
+		trace_ocfs2_get_dentry_generation((unsigned long long)blkno,
+						  handle->ih_generation,
+						  inode->i_generation);
 		result = ERR_PTR(-ESTALE);
 		goto bail;
 	}
@@ -141,7 +138,7 @@ check_gen:
 		mlog_errno(PTR_ERR(result));
 
 bail:
-	mlog_exit_ptr(result);
+	trace_ocfs2_get_dentry_end(result);
 	return result;
 }
 
@@ -152,11 +149,8 @@ static struct dentry *ocfs2_get_parent(struct dentry *child)
 	struct dentry *parent;
 	struct inode *dir = child->d_inode;
 
-	mlog_entry("(0x%p, '%.*s')\n", child,
-		   child->d_name.len, child->d_name.name);
-
-	mlog(0, "find parent of directory %llu\n",
-	     (unsigned long long)OCFS2_I(dir)->ip_blkno);
+	trace_ocfs2_get_parent(child, child->d_name.len, child->d_name.name,
+			       (unsigned long long)OCFS2_I(dir)->ip_blkno);
 
 	status = ocfs2_inode_lock(dir, NULL, 0);
 	if (status < 0) {
@@ -178,7 +172,7 @@ bail_unlock:
 	ocfs2_inode_unlock(dir, 0);
 
 bail:
-	mlog_exit_ptr(parent);
+	trace_ocfs2_get_parent_end(parent);
 
 	return parent;
 }
@@ -193,12 +187,16 @@ static int ocfs2_encode_fh(struct dentry *dentry, u32 *fh_in, int *max_len,
 	u32 generation;
 	__le32 *fh = (__force __le32 *) fh_in;
 
-	mlog_entry("(0x%p, '%.*s', 0x%p, %d, %d)\n", dentry,
-		   dentry->d_name.len, dentry->d_name.name,
-		   fh, len, connectable);
+	trace_ocfs2_encode_fh_begin(dentry, dentry->d_name.len,
+				    dentry->d_name.name,
+				    fh, len, connectable);
 
-	if (len < 3 || (connectable && len < 6)) {
-		mlog(ML_ERROR, "fh buffer is too small for encoding\n");
+	if (connectable && (len < 6)) {
+		*max_len = 6;
+		type = 255;
+		goto bail;
+	} else if (len < 3) {
+		*max_len = 3;
 		type = 255;
 		goto bail;
 	}
@@ -206,8 +204,7 @@ static int ocfs2_encode_fh(struct dentry *dentry, u32 *fh_in, int *max_len,
 	blkno = OCFS2_I(inode)->ip_blkno;
 	generation = inode->i_generation;
 
-	mlog(0, "Encoding fh: blkno: %llu, generation: %u\n",
-	     (unsigned long long)blkno, generation);
+	trace_ocfs2_encode_fh_self((unsigned long long)blkno, generation);
 
 	len = 3;
 	fh[0] = cpu_to_le32((u32)(blkno >> 32));
@@ -232,14 +229,14 @@ static int ocfs2_encode_fh(struct dentry *dentry, u32 *fh_in, int *max_len,
 		len = 6;
 		type = 2;
 
-		mlog(0, "Encoding parent: blkno: %llu, generation: %u\n",
-		     (unsigned long long)blkno, generation);
+		trace_ocfs2_encode_fh_parent((unsigned long long)blkno,
+					     generation);
 	}
 
 	*max_len = len;
 
 bail:
-	mlog_exit(type);
+	trace_ocfs2_encode_fh_type(type);
 	return type;
 }
 
diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c
index 09e3fdfa6d3..23457b491e8 100644
--- a/fs/ocfs2/extent_map.c
+++ b/fs/ocfs2/extent_map.c
@@ -28,7 +28,6 @@
 #include <linux/types.h>
 #include <linux/fiemap.h>
 
-#define MLOG_MASK_PREFIX ML_EXTENT_MAP
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -39,6 +38,7 @@
 #include "inode.h"
 #include "super.h"
 #include "symlink.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -841,10 +841,9 @@ int ocfs2_read_virt_blocks(struct inode *inode, u64 v_block, int nr,
 	u64 p_block, p_count;
 	int i, count, done = 0;
 
-	mlog_entry("(inode = %p, v_block = %llu, nr = %d, bhs = %p, "
-		   "flags = %x, validate = %p)\n",
-		   inode, (unsigned long long)v_block, nr, bhs, flags,
-		   validate);
+	trace_ocfs2_read_virt_blocks(
+	     inode, (unsigned long long)v_block, nr, bhs, flags,
+	     validate);
 
 	if (((v_block + nr - 1) << inode->i_sb->s_blocksize_bits) >=
 	    i_size_read(inode)) {
@@ -897,7 +896,6 @@ int ocfs2_read_virt_blocks(struct inode *inode, u64 v_block, int nr,
 	}
 
 out:
-	mlog_exit(rc);
 	return rc;
 }
 
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index a6651956482..41565ae5285 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -38,7 +38,6 @@
 #include <linux/quotaops.h>
 #include <linux/blkdev.h>
 
-#define MLOG_MASK_PREFIX ML_INODE
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -61,6 +60,7 @@
 #include "acl.h"
 #include "quota.h"
 #include "refcounttree.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -99,8 +99,10 @@ static int ocfs2_file_open(struct inode *inode, struct file *file)
 	int mode = file->f_flags;
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 
-	mlog_entry("(0x%p, 0x%p, '%.*s')\n", inode, file,
-		   file->f_path.dentry->d_name.len, file->f_path.dentry->d_name.name);
+	trace_ocfs2_file_open(inode, file, file->f_path.dentry,
+			      (unsigned long long)OCFS2_I(inode)->ip_blkno,
+			      file->f_path.dentry->d_name.len,
+			      file->f_path.dentry->d_name.name, mode);
 
 	if (file->f_mode & FMODE_WRITE)
 		dquot_initialize(inode);
@@ -135,7 +137,6 @@ static int ocfs2_file_open(struct inode *inode, struct file *file)
 	}
 
 leave:
-	mlog_exit(status);
 	return status;
 }
 
@@ -143,19 +144,19 @@ static int ocfs2_file_release(struct inode *inode, struct file *file)
 {
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 
-	mlog_entry("(0x%p, 0x%p, '%.*s')\n", inode, file,
-		       file->f_path.dentry->d_name.len,
-		       file->f_path.dentry->d_name.name);
-
 	spin_lock(&oi->ip_lock);
 	if (!--oi->ip_open_count)
 		oi->ip_flags &= ~OCFS2_INODE_OPEN_DIRECT;
+
+	trace_ocfs2_file_release(inode, file, file->f_path.dentry,
+				 oi->ip_blkno,
+				 file->f_path.dentry->d_name.len,
+				 file->f_path.dentry->d_name.name,
+				 oi->ip_open_count);
 	spin_unlock(&oi->ip_lock);
 
 	ocfs2_free_file_private(inode, file);
 
-	mlog_exit(0);
-
 	return 0;
 }
 
@@ -177,9 +178,11 @@ static int ocfs2_sync_file(struct file *file, int datasync)
 	struct inode *inode = file->f_mapping->host;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
-	mlog_entry("(0x%p, %d, 0x%p, '%.*s')\n", file, datasync,
-		   file->f_path.dentry, file->f_path.dentry->d_name.len,
-		   file->f_path.dentry->d_name.name);
+	trace_ocfs2_sync_file(inode, file, file->f_path.dentry,
+			      OCFS2_I(inode)->ip_blkno,
+			      file->f_path.dentry->d_name.len,
+			      file->f_path.dentry->d_name.name,
+			      (unsigned long long)datasync);
 
 	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) {
 		/*
@@ -195,7 +198,8 @@ static int ocfs2_sync_file(struct file *file, int datasync)
 	err = jbd2_journal_force_commit(journal);
 
 bail:
-	mlog_exit(err);
+	if (err)
+		mlog_errno(err);
 
 	return (err < 0) ? -EIO : 0;
 }
@@ -251,8 +255,6 @@ int ocfs2_update_inode_atime(struct inode *inode,
 	handle_t *handle;
 	struct ocfs2_dinode *di = (struct ocfs2_dinode *) bh->b_data;
 
-	mlog_entry_void();
-
 	handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
 	if (IS_ERR(handle)) {
 		ret = PTR_ERR(handle);
@@ -280,7 +282,6 @@ int ocfs2_update_inode_atime(struct inode *inode,
 out_commit:
 	ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
 out:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -291,7 +292,6 @@ static int ocfs2_set_inode_size(handle_t *handle,
 {
 	int status;
 
-	mlog_entry_void();
 	i_size_write(inode, new_i_size);
 	inode->i_blocks = ocfs2_inode_sector_count(inode);
 	inode->i_ctime = inode->i_mtime = CURRENT_TIME;
@@ -303,7 +303,6 @@ static int ocfs2_set_inode_size(handle_t *handle,
 	}
 
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -375,8 +374,6 @@ static int ocfs2_orphan_for_truncate(struct ocfs2_super *osb,
 	struct ocfs2_dinode *di;
 	u64 cluster_bytes;
 
-	mlog_entry_void();
-
 	/*
 	 * We need to CoW the cluster contains the offset if it is reflinked
 	 * since we will call ocfs2_zero_range_for_truncate later which will
@@ -429,8 +426,6 @@ static int ocfs2_orphan_for_truncate(struct ocfs2_super *osb,
 out_commit:
 	ocfs2_commit_trans(osb, handle);
 out:
-
-	mlog_exit(status);
 	return status;
 }
 
@@ -442,14 +437,14 @@ static int ocfs2_truncate_file(struct inode *inode,
 	struct ocfs2_dinode *fe = NULL;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
-	mlog_entry("(inode = %llu, new_i_size = %llu\n",
-		   (unsigned long long)OCFS2_I(inode)->ip_blkno,
-		   (unsigned long long)new_i_size);
-
 	/* We trust di_bh because it comes from ocfs2_inode_lock(), which
 	 * already validated it */
 	fe = (struct ocfs2_dinode *) di_bh->b_data;
 
+	trace_ocfs2_truncate_file((unsigned long long)OCFS2_I(inode)->ip_blkno,
+				  (unsigned long long)le64_to_cpu(fe->i_size),
+				  (unsigned long long)new_i_size);
+
 	mlog_bug_on_msg(le64_to_cpu(fe->i_size) != i_size_read(inode),
 			"Inode %llu, inode i_size = %lld != di "
 			"i_size = %llu, i_flags = 0x%x\n",
@@ -459,19 +454,14 @@ static int ocfs2_truncate_file(struct inode *inode,
 			le32_to_cpu(fe->i_flags));
 
 	if (new_i_size > le64_to_cpu(fe->i_size)) {
-		mlog(0, "asked to truncate file with size (%llu) to size (%llu)!\n",
-		     (unsigned long long)le64_to_cpu(fe->i_size),
-		     (unsigned long long)new_i_size);
+		trace_ocfs2_truncate_file_error(
+			(unsigned long long)le64_to_cpu(fe->i_size),
+			(unsigned long long)new_i_size);
 		status = -EINVAL;
 		mlog_errno(status);
 		goto bail;
 	}
 
-	mlog(0, "inode %llu, i_size = %llu, new_i_size = %llu\n",
-	     (unsigned long long)le64_to_cpu(fe->i_blkno),
-	     (unsigned long long)le64_to_cpu(fe->i_size),
-	     (unsigned long long)new_i_size);
-
 	/* lets handle the simple truncate cases before doing any more
 	 * cluster locking. */
 	if (new_i_size == le64_to_cpu(fe->i_size))
@@ -525,7 +515,6 @@ bail:
 	if (!status && OCFS2_I(inode)->ip_clusters == 0)
 		status = ocfs2_try_remove_refcount_tree(inode, di_bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -578,8 +567,6 @@ static int __ocfs2_extend_allocation(struct inode *inode, u32 logical_start,
 	struct ocfs2_extent_tree et;
 	int did_quota = 0;
 
-	mlog_entry("(clusters_to_add = %u)\n", clusters_to_add);
-
 	/*
 	 * This function only exists for file systems which don't
 	 * support holes.
@@ -596,11 +583,6 @@ static int __ocfs2_extend_allocation(struct inode *inode, u32 logical_start,
 restart_all:
 	BUG_ON(le32_to_cpu(fe->i_clusters) != OCFS2_I(inode)->ip_clusters);
 
-	mlog(0, "extend inode %llu, i_size = %lld, di->i_clusters = %u, "
-	     "clusters_to_add = %u\n",
-	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
-	     (long long)i_size_read(inode), le32_to_cpu(fe->i_clusters),
-	     clusters_to_add);
 	ocfs2_init_dinode_extent_tree(&et, INODE_CACHE(inode), bh);
 	status = ocfs2_lock_allocators(inode, &et, clusters_to_add, 0,
 				       &data_ac, &meta_ac);
@@ -620,6 +602,12 @@ restart_all:
 	}
 
 restarted_transaction:
+	trace_ocfs2_extend_allocation(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno,
+		(unsigned long long)i_size_read(inode),
+		le32_to_cpu(fe->i_clusters), clusters_to_add,
+		why, restart_func);
+
 	status = dquot_alloc_space_nodirty(inode,
 			ocfs2_clusters_to_bytes(osb->sb, clusters_to_add));
 	if (status)
@@ -666,13 +654,11 @@ restarted_transaction:
 
 	if (why != RESTART_NONE && clusters_to_add) {
 		if (why == RESTART_META) {
-			mlog(0, "restarting function.\n");
 			restart_func = 1;
 			status = 0;
 		} else {
 			BUG_ON(why != RESTART_TRANS);
 
-			mlog(0, "restarting transaction.\n");
 			/* TODO: This can be more intelligent. */
 			credits = ocfs2_calc_extend_credits(osb->sb,
 							    &fe->id2.i_list,
@@ -689,11 +675,11 @@ restarted_transaction:
 		}
 	}
 
-	mlog(0, "fe: i_clusters = %u, i_size=%llu\n",
+	trace_ocfs2_extend_allocation_end(OCFS2_I(inode)->ip_blkno,
 	     le32_to_cpu(fe->i_clusters),
-	     (unsigned long long)le64_to_cpu(fe->i_size));
-	mlog(0, "inode: ip_clusters=%u, i_size=%lld\n",
-	     OCFS2_I(inode)->ip_clusters, (long long)i_size_read(inode));
+	     (unsigned long long)le64_to_cpu(fe->i_size),
+	     OCFS2_I(inode)->ip_clusters,
+	     (unsigned long long)i_size_read(inode));
 
 leave:
 	if (status < 0 && did_quota)
@@ -718,7 +704,6 @@ leave:
 	brelse(bh);
 	bh = NULL;
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -785,10 +770,11 @@ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from,
 	if (!zero_to)
 		zero_to = PAGE_CACHE_SIZE;
 
-	mlog(0,
-	     "abs_from = %llu, abs_to = %llu, index = %lu, zero_from = %u, zero_to = %u\n",
-	     (unsigned long long)abs_from, (unsigned long long)abs_to,
-	     index, zero_from, zero_to);
+	trace_ocfs2_write_zero_page(
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			(unsigned long long)abs_from,
+			(unsigned long long)abs_to,
+			index, zero_from, zero_to);
 
 	/* We know that zero_from is block aligned */
 	for (block_start = zero_from; block_start < zero_to;
@@ -928,9 +914,10 @@ static int ocfs2_zero_extend_range(struct inode *inode, u64 range_start,
 	u64 next_pos;
 	u64 zero_pos = range_start;
 
-	mlog(0, "range_start = %llu, range_end = %llu\n",
-	     (unsigned long long)range_start,
-	     (unsigned long long)range_end);
+	trace_ocfs2_zero_extend_range(
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			(unsigned long long)range_start,
+			(unsigned long long)range_end);
 	BUG_ON(range_start >= range_end);
 
 	while (zero_pos < range_end) {
@@ -962,9 +949,9 @@ int ocfs2_zero_extend(struct inode *inode, struct buffer_head *di_bh,
 	struct super_block *sb = inode->i_sb;
 
 	zero_start = ocfs2_align_bytes_to_blocks(sb, i_size_read(inode));
-	mlog(0, "zero_start %llu for i_size %llu\n",
-	     (unsigned long long)zero_start,
-	     (unsigned long long)i_size_read(inode));
+	trace_ocfs2_zero_extend((unsigned long long)OCFS2_I(inode)->ip_blkno,
+				(unsigned long long)zero_start,
+				(unsigned long long)i_size_read(inode));
 	while (zero_start < zero_to_size) {
 		ret = ocfs2_zero_extend_get_range(inode, di_bh, zero_start,
 						  zero_to_size,
@@ -1113,30 +1100,20 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr)
 	struct dquot *transfer_to[MAXQUOTAS] = { };
 	int qtype;
 
-	mlog_entry("(0x%p, '%.*s')\n", dentry,
-	           dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_setattr(inode, dentry,
+			    (unsigned long long)OCFS2_I(inode)->ip_blkno,
+			    dentry->d_name.len, dentry->d_name.name,
+			    attr->ia_valid, attr->ia_mode,
+			    attr->ia_uid, attr->ia_gid);
 
 	/* ensuring we don't even attempt to truncate a symlink */
 	if (S_ISLNK(inode->i_mode))
 		attr->ia_valid &= ~ATTR_SIZE;
 
-	if (attr->ia_valid & ATTR_MODE)
-		mlog(0, "mode change: %d\n", attr->ia_mode);
-	if (attr->ia_valid & ATTR_UID)
-		mlog(0, "uid change: %d\n", attr->ia_uid);
-	if (attr->ia_valid & ATTR_GID)
-		mlog(0, "gid change: %d\n", attr->ia_gid);
-	if (attr->ia_valid & ATTR_SIZE)
-		mlog(0, "size change...\n");
-	if (attr->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_CTIME))
-		mlog(0, "time change...\n");
-
 #define OCFS2_VALID_ATTRS (ATTR_ATIME | ATTR_MTIME | ATTR_CTIME | ATTR_SIZE \
 			   | ATTR_GID | ATTR_UID | ATTR_MODE)
-	if (!(attr->ia_valid & OCFS2_VALID_ATTRS)) {
-		mlog(0, "can't handle attrs: 0x%x\n", attr->ia_valid);
+	if (!(attr->ia_valid & OCFS2_VALID_ATTRS))
 		return 0;
-	}
 
 	status = inode_change_ok(inode, attr);
 	if (status)
@@ -1274,7 +1251,6 @@ bail:
 			mlog_errno(status);
 	}
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1287,8 +1263,6 @@ int ocfs2_getattr(struct vfsmount *mnt,
 	struct ocfs2_super *osb = sb->s_fs_info;
 	int err;
 
-	mlog_entry_void();
-
 	err = ocfs2_inode_revalidate(dentry);
 	if (err) {
 		if (err != -ENOENT)
@@ -1302,8 +1276,6 @@ int ocfs2_getattr(struct vfsmount *mnt,
 	stat->blksize = osb->s_clustersize;
 
 bail:
-	mlog_exit(err);
-
 	return err;
 }
 
@@ -1314,8 +1286,6 @@ int ocfs2_permission(struct inode *inode, int mask, unsigned int flags)
 	if (flags & IPERM_FLAG_RCU)
 		return -ECHILD;
 
-	mlog_entry_void();
-
 	ret = ocfs2_inode_lock(inode, NULL, 0);
 	if (ret) {
 		if (ret != -ENOENT)
@@ -1327,7 +1297,6 @@ int ocfs2_permission(struct inode *inode, int mask, unsigned int flags)
 
 	ocfs2_inode_unlock(inode, 0);
 out:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -1339,8 +1308,9 @@ static int __ocfs2_write_remove_suid(struct inode *inode,
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	struct ocfs2_dinode *di;
 
-	mlog_entry("(Inode %llu, mode 0%o)\n",
-		   (unsigned long long)OCFS2_I(inode)->ip_blkno, inode->i_mode);
+	trace_ocfs2_write_remove_suid(
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			inode->i_mode);
 
 	handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
 	if (IS_ERR(handle)) {
@@ -1368,7 +1338,6 @@ static int __ocfs2_write_remove_suid(struct inode *inode,
 out_trans:
 	ocfs2_commit_trans(osb, handle);
 out:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -1547,8 +1516,9 @@ static int ocfs2_zero_partial_clusters(struct inode *inode,
 	 * partial clusters here. There's no need to worry about
 	 * physical allocation - the zeroing code knows to skip holes.
 	 */
-	mlog(0, "byte start: %llu, end: %llu\n",
-	     (unsigned long long)start, (unsigned long long)end);
+	trace_ocfs2_zero_partial_clusters(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno,
+		(unsigned long long)start, (unsigned long long)end);
 
 	/*
 	 * If both edges are on a cluster boundary then there's no
@@ -1572,8 +1542,8 @@ static int ocfs2_zero_partial_clusters(struct inode *inode,
 	if (tmpend > end)
 		tmpend = end;
 
-	mlog(0, "1st range: start: %llu, tmpend: %llu\n",
-	     (unsigned long long)start, (unsigned long long)tmpend);
+	trace_ocfs2_zero_partial_clusters_range1((unsigned long long)start,
+						 (unsigned long long)tmpend);
 
 	ret = ocfs2_zero_range_for_truncate(inode, handle, start, tmpend);
 	if (ret)
@@ -1587,8 +1557,8 @@ static int ocfs2_zero_partial_clusters(struct inode *inode,
 		 */
 		start = end & ~(osb->s_clustersize - 1);
 
-		mlog(0, "2nd range: start: %llu, end: %llu\n",
-		     (unsigned long long)start, (unsigned long long)end);
+		trace_ocfs2_zero_partial_clusters_range2(
+			(unsigned long long)start, (unsigned long long)end);
 
 		ret = ocfs2_zero_range_for_truncate(inode, handle, start, end);
 		if (ret)
@@ -1688,6 +1658,11 @@ static int ocfs2_remove_inode_range(struct inode *inode,
 	ocfs2_init_dinode_extent_tree(&et, INODE_CACHE(inode), di_bh);
 	ocfs2_init_dealloc_ctxt(&dealloc);
 
+	trace_ocfs2_remove_inode_range(
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			(unsigned long long)byte_start,
+			(unsigned long long)byte_len);
+
 	if (byte_len == 0)
 		return 0;
 
@@ -1734,11 +1709,6 @@ static int ocfs2_remove_inode_range(struct inode *inode,
 	trunc_end = (byte_start + byte_len) >> osb->s_clustersize_bits;
 	cluster_in_el = trunc_end;
 
-	mlog(0, "Inode: %llu, start: %llu, len: %llu, cstart: %u, cend: %u\n",
-	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
-	     (unsigned long long)byte_start,
-	     (unsigned long long)byte_len, trunc_start, trunc_end);
-
 	ret = ocfs2_zero_partial_clusters(inode, byte_start, byte_len);
 	if (ret) {
 		mlog_errno(ret);
@@ -2093,7 +2063,7 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 	int ret = 0, meta_level = 0;
 	struct dentry *dentry = file->f_path.dentry;
 	struct inode *inode = dentry->d_inode;
-	loff_t saved_pos, end;
+	loff_t saved_pos = 0, end;
 
 	/*
 	 * We start with a read level meta lock and only jump to an ex
@@ -2132,12 +2102,10 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 
 		/* work on a copy of ppos until we're sure that we won't have
 		 * to recalculate it due to relocking. */
-		if (appending) {
+		if (appending)
 			saved_pos = i_size_read(inode);
-			mlog(0, "O_APPEND: inode->i_size=%llu\n", saved_pos);
-		} else {
+		else
 			saved_pos = *ppos;
-		}
 
 		end = saved_pos + count;
 
@@ -2208,6 +2176,10 @@ static int ocfs2_prepare_inode_for_write(struct file *file,
 		*ppos = saved_pos;
 
 out_unlock:
+	trace_ocfs2_prepare_inode_for_write(OCFS2_I(inode)->ip_blkno,
+					    saved_pos, appending, count,
+					    direct_io, has_refcount);
+
 	if (meta_level >= 0)
 		ocfs2_inode_unlock(inode, meta_level);
 
@@ -2233,10 +2205,11 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
 	int full_coherency = !(osb->s_mount_opt &
 			       OCFS2_MOUNT_COHERENCY_BUFFERED);
 
-	mlog_entry("(0x%p, %u, '%.*s')\n", file,
-		   (unsigned int)nr_segs,
-		   file->f_path.dentry->d_name.len,
-		   file->f_path.dentry->d_name.name);
+	trace_ocfs2_file_aio_write(inode, file, file->f_path.dentry,
+		(unsigned long long)OCFS2_I(inode)->ip_blkno,
+		file->f_path.dentry->d_name.len,
+		file->f_path.dentry->d_name.name,
+		(unsigned int)nr_segs);
 
 	if (iocb->ki_left == 0)
 		return 0;
@@ -2402,7 +2375,6 @@ out_sems:
 
 	if (written)
 		ret = written;
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -2438,10 +2410,11 @@ static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe,
 		.u.file = out,
 	};
 
-	mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", out, pipe,
-		   (unsigned int)len,
-		   out->f_path.dentry->d_name.len,
-		   out->f_path.dentry->d_name.name);
+
+	trace_ocfs2_file_splice_write(inode, out, out->f_path.dentry,
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			out->f_path.dentry->d_name.len,
+			out->f_path.dentry->d_name.name, len);
 
 	if (pipe->inode)
 		mutex_lock_nested(&pipe->inode->i_mutex, I_MUTEX_PARENT);
@@ -2485,7 +2458,6 @@ static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe,
 		balance_dirty_pages_ratelimited_nr(mapping, nr_pages);
 	}
 
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -2498,10 +2470,10 @@ static ssize_t ocfs2_file_splice_read(struct file *in,
 	int ret = 0, lock_level = 0;
 	struct inode *inode = in->f_path.dentry->d_inode;
 
-	mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", in, pipe,
-		   (unsigned int)len,
-		   in->f_path.dentry->d_name.len,
-		   in->f_path.dentry->d_name.name);
+	trace_ocfs2_file_splice_read(inode, in, in->f_path.dentry,
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			in->f_path.dentry->d_name.len,
+			in->f_path.dentry->d_name.name, len);
 
 	/*
 	 * See the comment in ocfs2_file_aio_read()
@@ -2516,7 +2488,6 @@ static ssize_t ocfs2_file_splice_read(struct file *in,
 	ret = generic_file_splice_read(in, ppos, pipe, len, flags);
 
 bail:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -2529,10 +2500,11 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
 	struct file *filp = iocb->ki_filp;
 	struct inode *inode = filp->f_path.dentry->d_inode;
 
-	mlog_entry("(0x%p, %u, '%.*s')\n", filp,
-		   (unsigned int)nr_segs,
-		   filp->f_path.dentry->d_name.len,
-		   filp->f_path.dentry->d_name.name);
+	trace_ocfs2_file_aio_read(inode, filp, filp->f_path.dentry,
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			filp->f_path.dentry->d_name.len,
+			filp->f_path.dentry->d_name.name, nr_segs);
+
 
 	if (!inode) {
 		ret = -EINVAL;
@@ -2578,8 +2550,7 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
 	ocfs2_inode_unlock(inode, lock_level);
 
 	ret = generic_file_aio_read(iocb, iov, nr_segs, iocb->ki_pos);
-	if (ret == -EINVAL)
-		mlog(0, "generic_file_aio_read returned -EINVAL\n");
+	trace_generic_file_aio_read_ret(ret);
 
 	/* buffered aio wouldn't have proper lock coverage today */
 	BUG_ON(ret == -EIOCBQUEUED && !(filp->f_flags & O_DIRECT));
@@ -2597,7 +2568,6 @@ bail:
 	}
 	if (rw_level != -1)
 		ocfs2_rw_unlock(inode, rw_level);
-	mlog_exit(ret);
 
 	return ret;
 }
diff --git a/fs/ocfs2/heartbeat.c b/fs/ocfs2/heartbeat.c
index 1aa863dd901..d8208b20dc5 100644
--- a/fs/ocfs2/heartbeat.c
+++ b/fs/ocfs2/heartbeat.c
@@ -28,7 +28,6 @@
 #include <linux/types.h>
 #include <linux/highmem.h>
 
-#define MLOG_MASK_PREFIX ML_SUPER
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -37,6 +36,7 @@
 #include "heartbeat.h"
 #include "inode.h"
 #include "journal.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -66,7 +66,7 @@ void ocfs2_do_node_down(int node_num, void *data)
 
 	BUG_ON(osb->node_num == node_num);
 
-	mlog(0, "ocfs2: node down event for %d\n", node_num);
+	trace_ocfs2_do_node_down(node_num);
 
 	if (!osb->cconn) {
 		/*
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 4068c6c4c6f..177d3a6c2a5 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -31,7 +31,6 @@
 
 #include <asm/byteorder.h>
 
-#define MLOG_MASK_PREFIX ML_INODE
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -53,6 +52,7 @@
 #include "uptodate.h"
 #include "xattr.h"
 #include "refcounttree.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -131,7 +131,8 @@ struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, unsigned flags,
 	struct super_block *sb = osb->sb;
 	struct ocfs2_find_inode_args args;
 
-	mlog_entry("(blkno = %llu)\n", (unsigned long long)blkno);
+	trace_ocfs2_iget_begin((unsigned long long)blkno, flags,
+			       sysfile_type);
 
 	/* Ok. By now we've either got the offsets passed to us by the
 	 * caller, or we just pulled them off the bh. Lets do some
@@ -152,16 +153,16 @@ struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, unsigned flags,
 	/* inode was *not* in the inode cache. 2.6.x requires
 	 * us to do our own read_inode call and unlock it
 	 * afterwards. */
-	if (inode && inode->i_state & I_NEW) {
-		mlog(0, "Inode was not in inode cache, reading it.\n");
-		ocfs2_read_locked_inode(inode, &args);
-		unlock_new_inode(inode);
-	}
 	if (inode == NULL) {
 		inode = ERR_PTR(-ENOMEM);
 		mlog_errno(PTR_ERR(inode));
 		goto bail;
 	}
+	trace_ocfs2_iget5_locked(inode->i_state);
+	if (inode->i_state & I_NEW) {
+		ocfs2_read_locked_inode(inode, &args);
+		unlock_new_inode(inode);
+	}
 	if (is_bad_inode(inode)) {
 		iput(inode);
 		inode = ERR_PTR(-ESTALE);
@@ -170,9 +171,8 @@ struct inode *ocfs2_iget(struct ocfs2_super *osb, u64 blkno, unsigned flags,
 
 bail:
 	if (!IS_ERR(inode)) {
-		mlog(0, "returning inode with number %llu\n",
-		     (unsigned long long)OCFS2_I(inode)->ip_blkno);
-		mlog_exit_ptr(inode);
+		trace_ocfs2_iget_end(inode, 
+			(unsigned long long)OCFS2_I(inode)->ip_blkno);
 	}
 
 	return inode;
@@ -192,18 +192,17 @@ static int ocfs2_find_actor(struct inode *inode, void *opaque)
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 	int ret = 0;
 
-	mlog_entry("(0x%p, %lu, 0x%p)\n", inode, inode->i_ino, opaque);
-
 	args = opaque;
 
 	mlog_bug_on_msg(!inode, "No inode in find actor!\n");
 
+	trace_ocfs2_find_actor(inode, inode->i_ino, opaque, args->fi_blkno);
+
 	if (oi->ip_blkno != args->fi_blkno)
 		goto bail;
 
 	ret = 1;
 bail:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -218,8 +217,6 @@ static int ocfs2_init_locked_inode(struct inode *inode, void *opaque)
 	static struct lock_class_key ocfs2_quota_ip_alloc_sem_key,
 				     ocfs2_file_ip_alloc_sem_key;
 
-	mlog_entry("inode = %p, opaque = %p\n", inode, opaque);
-
 	inode->i_ino = args->fi_ino;
 	OCFS2_I(inode)->ip_blkno = args->fi_blkno;
 	if (args->fi_sysfile_type != 0)
@@ -235,7 +232,6 @@ static int ocfs2_init_locked_inode(struct inode *inode, void *opaque)
 		lockdep_set_class(&OCFS2_I(inode)->ip_alloc_sem,
 				  &ocfs2_file_ip_alloc_sem_key);
 
-	mlog_exit(0);
 	return 0;
 }
 
@@ -246,9 +242,6 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
 	struct ocfs2_super *osb;
 	int use_plocks = 1;
 
-	mlog_entry("(0x%p, size:%llu)\n", inode,
-		   (unsigned long long)le64_to_cpu(fe->i_size));
-
 	sb = inode->i_sb;
 	osb = OCFS2_SB(sb);
 
@@ -300,20 +293,20 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
 
 	inode->i_nlink = ocfs2_read_links_count(fe);
 
+	trace_ocfs2_populate_inode(OCFS2_I(inode)->ip_blkno,
+				   le32_to_cpu(fe->i_flags));
 	if (fe->i_flags & cpu_to_le32(OCFS2_SYSTEM_FL)) {
 		OCFS2_I(inode)->ip_flags |= OCFS2_INODE_SYSTEM_FILE;
 		inode->i_flags |= S_NOQUOTA;
 	}
-
+  
 	if (fe->i_flags & cpu_to_le32(OCFS2_LOCAL_ALLOC_FL)) {
 		OCFS2_I(inode)->ip_flags |= OCFS2_INODE_BITMAP;
-		mlog(0, "local alloc inode: i_ino=%lu\n", inode->i_ino);
 	} else if (fe->i_flags & cpu_to_le32(OCFS2_BITMAP_FL)) {
 		OCFS2_I(inode)->ip_flags |= OCFS2_INODE_BITMAP;
 	} else if (fe->i_flags & cpu_to_le32(OCFS2_QUOTA_FL)) {
 		inode->i_flags |= S_NOQUOTA;
 	} else if (fe->i_flags & cpu_to_le32(OCFS2_SUPER_BLOCK_FL)) {
-		mlog(0, "superblock inode: i_ino=%lu\n", inode->i_ino);
 		/* we can't actually hit this as read_inode can't
 		 * handle superblocks today ;-) */
 		BUG();
@@ -381,7 +374,6 @@ void ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
 	if (S_ISDIR(inode->i_mode))
 		ocfs2_resv_set_type(&OCFS2_I(inode)->ip_la_data_resv,
 				    OCFS2_RESV_FLAG_DIR);
-	mlog_exit_void();
 }
 
 static int ocfs2_read_locked_inode(struct inode *inode,
@@ -394,8 +386,6 @@ static int ocfs2_read_locked_inode(struct inode *inode,
 	int status, can_lock;
 	u32 generation = 0;
 
-	mlog_entry("(0x%p, 0x%p)\n", inode, args);
-
 	status = -EINVAL;
 	if (inode == NULL || inode->i_sb == NULL) {
 		mlog(ML_ERROR, "bad inode\n");
@@ -443,6 +433,9 @@ static int ocfs2_read_locked_inode(struct inode *inode,
 		&& !(args->fi_flags & OCFS2_FI_FLAG_ORPHAN_RECOVERY)
 		&& !ocfs2_mount_local(osb);
 
+	trace_ocfs2_read_locked_inode(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno, can_lock);
+
 	/*
 	 * To maintain backwards compatibility with older versions of
 	 * ocfs2-tools, we still store the generation value for system
@@ -534,7 +527,6 @@ bail:
 	if (args && bh)
 		brelse(bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -551,8 +543,6 @@ static int ocfs2_truncate_for_delete(struct ocfs2_super *osb,
 	struct ocfs2_dinode *fe;
 	handle_t *handle = NULL;
 
-	mlog_entry_void();
-
 	fe = (struct ocfs2_dinode *) fe_bh->b_data;
 
 	/*
@@ -600,7 +590,6 @@ static int ocfs2_truncate_for_delete(struct ocfs2_super *osb,
 out:
 	if (handle)
 		ocfs2_commit_trans(osb, handle);
-	mlog_exit(status);
 	return status;
 }
 
@@ -696,8 +685,6 @@ static int ocfs2_check_orphan_recovery_state(struct ocfs2_super *osb,
 
 	spin_lock(&osb->osb_lock);
 	if (ocfs2_node_map_test_bit(osb, &osb->osb_recovering_orphan_dirs, slot)) {
-		mlog(0, "Recovery is happening on orphan dir %d, will skip "
-		     "this inode\n", slot);
 		ret = -EDEADLK;
 		goto out;
 	}
@@ -706,6 +693,7 @@ static int ocfs2_check_orphan_recovery_state(struct ocfs2_super *osb,
 	osb->osb_orphan_wipes[slot]++;
 out:
 	spin_unlock(&osb->osb_lock);
+	trace_ocfs2_check_orphan_recovery_state(slot, ret);
 	return ret;
 }
 
@@ -816,6 +804,10 @@ static int ocfs2_inode_is_valid_to_delete(struct inode *inode)
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
+	trace_ocfs2_inode_is_valid_to_delete(current, osb->dc_task,
+					     (unsigned long long)oi->ip_blkno,
+					     oi->ip_flags);
+
 	/* We shouldn't be getting here for the root directory
 	 * inode.. */
 	if (inode == osb->root_inode) {
@@ -828,11 +820,8 @@ static int ocfs2_inode_is_valid_to_delete(struct inode *inode)
 	 * have to skip deleting this guy. That's OK though because
 	 * the node who's doing the actual deleting should handle it
 	 * anyway. */
-	if (current == osb->dc_task) {
-		mlog(0, "Skipping delete of %lu because we're currently "
-		     "in downconvert\n", inode->i_ino);
+	if (current == osb->dc_task)
 		goto bail;
-	}
 
 	spin_lock(&oi->ip_lock);
 	/* OCFS2 *never* deletes system files. This should technically
@@ -847,11 +836,8 @@ static int ocfs2_inode_is_valid_to_delete(struct inode *inode)
 	/* If we have allowd wipe of this inode for another node, it
 	 * will be marked here so we can safely skip it. Recovery will
 	 * cleanup any inodes we might inadvertantly skip here. */
-	if (oi->ip_flags & OCFS2_INODE_SKIP_DELETE) {
-		mlog(0, "Skipping delete of %lu because another node "
-		     "has done this for us.\n", inode->i_ino);
+	if (oi->ip_flags & OCFS2_INODE_SKIP_DELETE)
 		goto bail_unlock;
-	}
 
 	ret = 1;
 bail_unlock:
@@ -868,28 +854,27 @@ static int ocfs2_query_inode_wipe(struct inode *inode,
 				  struct buffer_head *di_bh,
 				  int *wipe)
 {
-	int status = 0;
+	int status = 0, reason = 0;
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 	struct ocfs2_dinode *di;
 
 	*wipe = 0;
 
+	trace_ocfs2_query_inode_wipe_begin((unsigned long long)oi->ip_blkno,
+					   inode->i_nlink);
+
 	/* While we were waiting for the cluster lock in
 	 * ocfs2_delete_inode, another node might have asked to delete
 	 * the inode. Recheck our flags to catch this. */
 	if (!ocfs2_inode_is_valid_to_delete(inode)) {
-		mlog(0, "Skipping delete of %llu because flags changed\n",
-		     (unsigned long long)oi->ip_blkno);
+		reason = 1;
 		goto bail;
 	}
 
 	/* Now that we have an up to date inode, we can double check
 	 * the link count. */
-	if (inode->i_nlink) {
-		mlog(0, "Skipping delete of %llu because nlink = %u\n",
-		     (unsigned long long)oi->ip_blkno, inode->i_nlink);
+	if (inode->i_nlink)
 		goto bail;
-	}
 
 	/* Do some basic inode verification... */
 	di = (struct ocfs2_dinode *) di_bh->b_data;
@@ -904,9 +889,7 @@ static int ocfs2_query_inode_wipe(struct inode *inode,
 		 * ORPHANED_FL not.
 		 */
 		if (di->i_dyn_features & cpu_to_le16(OCFS2_HAS_REFCOUNT_FL)) {
-			mlog(0, "Reflinked inode %llu is no longer orphaned.  "
-			     "it shouldn't be deleted\n",
-			     (unsigned long long)oi->ip_blkno);
+			reason = 2;
 			goto bail;
 		}
 
@@ -943,8 +926,7 @@ static int ocfs2_query_inode_wipe(struct inode *inode,
 	status = ocfs2_try_open_lock(inode, 1);
 	if (status == -EAGAIN) {
 		status = 0;
-		mlog(0, "Skipping delete of %llu because it is in use on "
-		     "other nodes\n", (unsigned long long)oi->ip_blkno);
+		reason = 3;
 		goto bail;
 	}
 	if (status < 0) {
@@ -953,11 +935,10 @@ static int ocfs2_query_inode_wipe(struct inode *inode,
 	}
 
 	*wipe = 1;
-	mlog(0, "Inode %llu is ok to wipe from orphan dir %u\n",
-	     (unsigned long long)oi->ip_blkno,
-	     le16_to_cpu(di->i_orphaned_slot));
+	trace_ocfs2_query_inode_wipe_succ(le16_to_cpu(di->i_orphaned_slot));
 
 bail:
+	trace_ocfs2_query_inode_wipe_end(status, reason);
 	return status;
 }
 
@@ -967,8 +948,8 @@ bail:
 static void ocfs2_cleanup_delete_inode(struct inode *inode,
 				       int sync_data)
 {
-	mlog(0, "Cleanup inode %llu, sync = %d\n",
-	     (unsigned long long)OCFS2_I(inode)->ip_blkno, sync_data);
+	trace_ocfs2_cleanup_delete_inode(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno, sync_data);
 	if (sync_data)
 		write_inode_now(inode, 1);
 	truncate_inode_pages(&inode->i_data, 0);
@@ -980,15 +961,15 @@ static void ocfs2_delete_inode(struct inode *inode)
 	sigset_t oldset;
 	struct buffer_head *di_bh = NULL;
 
-	mlog_entry("(inode->i_ino = %lu)\n", inode->i_ino);
+	trace_ocfs2_delete_inode(inode->i_ino,
+				 (unsigned long long)OCFS2_I(inode)->ip_blkno,
+				 is_bad_inode(inode));
 
 	/* When we fail in read_inode() we mark inode as bad. The second test
 	 * catches the case when inode allocation fails before allocating
 	 * a block for inode. */
-	if (is_bad_inode(inode) || !OCFS2_I(inode)->ip_blkno) {
-		mlog(0, "Skipping delete of bad inode\n");
+	if (is_bad_inode(inode) || !OCFS2_I(inode)->ip_blkno)
 		goto bail;
-	}
 
 	dquot_initialize(inode);
 
@@ -1080,7 +1061,7 @@ bail_unlock_nfs_sync:
 bail_unblock:
 	ocfs2_unblock_signals(&oldset);
 bail:
-	mlog_exit_void();
+	return;
 }
 
 static void ocfs2_clear_inode(struct inode *inode)
@@ -1088,11 +1069,9 @@ static void ocfs2_clear_inode(struct inode *inode)
 	int status;
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 
-	mlog_entry_void();
-
 	end_writeback(inode);
-	mlog(0, "Clearing inode: %llu, nlink = %u\n",
-	     (unsigned long long)OCFS2_I(inode)->ip_blkno, inode->i_nlink);
+	trace_ocfs2_clear_inode((unsigned long long)oi->ip_blkno,
+				inode->i_nlink);
 
 	mlog_bug_on_msg(OCFS2_SB(inode->i_sb) == NULL,
 			"Inode=%lu\n", inode->i_ino);
@@ -1181,8 +1160,6 @@ static void ocfs2_clear_inode(struct inode *inode)
 	 */
 	jbd2_journal_release_jbd_inode(OCFS2_SB(inode->i_sb)->journal->j_journal,
 				       &oi->ip_jinode);
-
-	mlog_exit_void();
 }
 
 void ocfs2_evict_inode(struct inode *inode)
@@ -1204,17 +1181,14 @@ int ocfs2_drop_inode(struct inode *inode)
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 	int res;
 
-	mlog_entry_void();
-
-	mlog(0, "Drop inode %llu, nlink = %u, ip_flags = 0x%x\n",
-	     (unsigned long long)oi->ip_blkno, inode->i_nlink, oi->ip_flags);
+	trace_ocfs2_drop_inode((unsigned long long)oi->ip_blkno,
+				inode->i_nlink, oi->ip_flags);
 
 	if (oi->ip_flags & OCFS2_INODE_MAYBE_ORPHANED)
 		res = 1;
 	else
 		res = generic_drop_inode(inode);
 
-	mlog_exit_void();
 	return res;
 }
 
@@ -1226,11 +1200,11 @@ int ocfs2_inode_revalidate(struct dentry *dentry)
 	struct inode *inode = dentry->d_inode;
 	int status = 0;
 
-	mlog_entry("(inode = 0x%p, ino = %llu)\n", inode,
-		   inode ? (unsigned long long)OCFS2_I(inode)->ip_blkno : 0ULL);
+	trace_ocfs2_inode_revalidate(inode,
+		inode ? (unsigned long long)OCFS2_I(inode)->ip_blkno : 0ULL,
+		inode ? (unsigned long long)OCFS2_I(inode)->ip_flags : 0);
 
 	if (!inode) {
-		mlog(0, "eep, no inode!\n");
 		status = -ENOENT;
 		goto bail;
 	}
@@ -1238,7 +1212,6 @@ int ocfs2_inode_revalidate(struct dentry *dentry)
 	spin_lock(&OCFS2_I(inode)->ip_lock);
 	if (OCFS2_I(inode)->ip_flags & OCFS2_INODE_DELETED) {
 		spin_unlock(&OCFS2_I(inode)->ip_lock);
-		mlog(0, "inode deleted!\n");
 		status = -ENOENT;
 		goto bail;
 	}
@@ -1254,8 +1227,6 @@ int ocfs2_inode_revalidate(struct dentry *dentry)
 	}
 	ocfs2_inode_unlock(inode, 0);
 bail:
-	mlog_exit(status);
-
 	return status;
 }
 
@@ -1271,8 +1242,7 @@ int ocfs2_mark_inode_dirty(handle_t *handle,
 	int status;
 	struct ocfs2_dinode *fe = (struct ocfs2_dinode *) bh->b_data;
 
-	mlog_entry("(inode %llu)\n",
-		   (unsigned long long)OCFS2_I(inode)->ip_blkno);
+	trace_ocfs2_mark_inode_dirty((unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 	status = ocfs2_journal_access_di(handle, INODE_CACHE(inode), bh,
 					 OCFS2_JOURNAL_ACCESS_WRITE);
@@ -1302,7 +1272,6 @@ int ocfs2_mark_inode_dirty(handle_t *handle,
 
 	ocfs2_journal_dirty(handle, bh);
 leave:
-	mlog_exit(status);
 	return status;
 }
 
@@ -1345,8 +1314,7 @@ int ocfs2_validate_inode_block(struct super_block *sb,
 	int rc;
 	struct ocfs2_dinode *di = (struct ocfs2_dinode *)bh->b_data;
 
-	mlog(0, "Validating dinode %llu\n",
-	     (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_validate_inode_block((unsigned long long)bh->b_blocknr);
 
 	BUG_ON(!buffer_uptodate(bh));
 
diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c
index 731cf4670b9..f97a213223e 100644
--- a/fs/ocfs2/ioctl.c
+++ b/fs/ocfs2/ioctl.c
@@ -9,7 +9,6 @@
 #include <linux/mount.h>
 #include <linux/compat.h>
 
-#define MLOG_MASK_PREFIX ML_INODE
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -75,7 +74,6 @@ static int ocfs2_get_inode_attr(struct inode *inode, unsigned *flags)
 	*flags = OCFS2_I(inode)->ip_attr;
 	ocfs2_inode_unlock(inode, 0);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -141,7 +139,6 @@ bail:
 
 	brelse(bh);
 
-	mlog_exit(status);
 	return status;
 }
 
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index faa2303dbf0..dcc2d932715 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -31,7 +31,6 @@
 #include <linux/time.h>
 #include <linux/random.h>
 
-#define MLOG_MASK_PREFIX ML_JOURNAL
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -52,6 +51,7 @@
 #include "quota.h"
 
 #include "buffer_head_io.h"
+#include "ocfs2_trace.h"
 
 DEFINE_SPINLOCK(trans_inc_lock);
 
@@ -303,16 +303,15 @@ static int ocfs2_commit_cache(struct ocfs2_super *osb)
 	unsigned int flushed;
 	struct ocfs2_journal *journal = NULL;
 
-	mlog_entry_void();
-
 	journal = osb->journal;
 
 	/* Flush all pending commits and checkpoint the journal. */
 	down_write(&journal->j_trans_barrier);
 
-	if (atomic_read(&journal->j_num_trans) == 0) {
+	flushed = atomic_read(&journal->j_num_trans);
+	trace_ocfs2_commit_cache_begin(flushed);
+	if (flushed == 0) {
 		up_write(&journal->j_trans_barrier);
-		mlog(0, "No transactions for me to flush!\n");
 		goto finally;
 	}
 
@@ -331,13 +330,11 @@ static int ocfs2_commit_cache(struct ocfs2_super *osb)
 	atomic_set(&journal->j_num_trans, 0);
 	up_write(&journal->j_trans_barrier);
 
-	mlog(0, "commit_thread: flushed transaction %lu (%u handles)\n",
-	     journal->j_trans_id, flushed);
+	trace_ocfs2_commit_cache_end(journal->j_trans_id, flushed);
 
 	ocfs2_wake_downconvert_thread(osb);
 	wake_up(&journal->j_checkpointed);
 finally:
-	mlog_exit(status);
 	return status;
 }
 
@@ -425,9 +422,8 @@ int ocfs2_extend_trans(handle_t *handle, int nblocks)
 		return 0;
 
 	old_nblocks = handle->h_buffer_credits;
-	mlog_entry_void();
 
-	mlog(0, "Trying to extend transaction by %d blocks\n", nblocks);
+	trace_ocfs2_extend_trans(old_nblocks, nblocks);
 
 #ifdef CONFIG_OCFS2_DEBUG_FS
 	status = 1;
@@ -440,9 +436,7 @@ int ocfs2_extend_trans(handle_t *handle, int nblocks)
 #endif
 
 	if (status > 0) {
-		mlog(0,
-		     "jbd2_journal_extend failed, trying "
-		     "jbd2_journal_restart\n");
+		trace_ocfs2_extend_trans_restart(old_nblocks + nblocks);
 		status = jbd2_journal_restart(handle,
 					      old_nblocks + nblocks);
 		if (status < 0) {
@@ -453,8 +447,6 @@ int ocfs2_extend_trans(handle_t *handle, int nblocks)
 
 	status = 0;
 bail:
-
-	mlog_exit(status);
 	return status;
 }
 
@@ -622,12 +614,9 @@ static int __ocfs2_journal_access(handle_t *handle,
 	BUG_ON(!handle);
 	BUG_ON(!bh);
 
-	mlog_entry("bh->b_blocknr=%llu, type=%d (\"%s\"), bh->b_size = %zu\n",
-		   (unsigned long long)bh->b_blocknr, type,
-		   (type == OCFS2_JOURNAL_ACCESS_CREATE) ?
-		   "OCFS2_JOURNAL_ACCESS_CREATE" :
-		   "OCFS2_JOURNAL_ACCESS_WRITE",
-		   bh->b_size);
+	trace_ocfs2_journal_access(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)bh->b_blocknr, type, bh->b_size);
 
 	/* we can safely remove this assertion after testing. */
 	if (!buffer_uptodate(bh)) {
@@ -668,7 +657,6 @@ static int __ocfs2_journal_access(handle_t *handle,
 		mlog(ML_ERROR, "Error %d getting %d access to buffer!\n",
 		     status, type);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -737,13 +725,10 @@ void ocfs2_journal_dirty(handle_t *handle, struct buffer_head *bh)
 {
 	int status;
 
-	mlog_entry("(bh->b_blocknr=%llu)\n",
-		   (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_journal_dirty((unsigned long long)bh->b_blocknr);
 
 	status = jbd2_journal_dirty_metadata(handle, bh);
 	BUG_ON(status);
-
-	mlog_exit_void();
 }
 
 #define OCFS2_DEFAULT_COMMIT_INTERVAL	(HZ * JBD2_DEFAULT_MAX_COMMIT_AGE)
@@ -775,8 +760,6 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty)
 	struct ocfs2_super *osb;
 	int inode_lock = 0;
 
-	mlog_entry_void();
-
 	BUG_ON(!journal);
 
 	osb = journal->j_osb;
@@ -820,10 +803,9 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty)
 		goto done;
 	}
 
-	mlog(0, "inode->i_size = %lld\n", inode->i_size);
-	mlog(0, "inode->i_blocks = %llu\n",
-			(unsigned long long)inode->i_blocks);
-	mlog(0, "inode->ip_clusters = %u\n", OCFS2_I(inode)->ip_clusters);
+	trace_ocfs2_journal_init(inode->i_size,
+				 (unsigned long long)inode->i_blocks,
+				 OCFS2_I(inode)->ip_clusters);
 
 	/* call the kernels journal init function now */
 	j_journal = jbd2_journal_init_inode(inode);
@@ -833,8 +815,7 @@ int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty)
 		goto done;
 	}
 
-	mlog(0, "Returned from jbd2_journal_init_inode\n");
-	mlog(0, "j_journal->j_maxlen = %u\n", j_journal->j_maxlen);
+	trace_ocfs2_journal_init_maxlen(j_journal->j_maxlen);
 
 	*dirty = (le32_to_cpu(di->id1.journal1.ij_flags) &
 		  OCFS2_JOURNAL_DIRTY_FL);
@@ -859,7 +840,6 @@ done:
 		}
 	}
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -882,8 +862,6 @@ static int ocfs2_journal_toggle_dirty(struct ocfs2_super *osb,
 	struct buffer_head *bh = journal->j_bh;
 	struct ocfs2_dinode *fe;
 
-	mlog_entry_void();
-
 	fe = (struct ocfs2_dinode *)bh->b_data;
 
 	/* The journal bh on the osb always comes from ocfs2_journal_init()
@@ -906,7 +884,6 @@ static int ocfs2_journal_toggle_dirty(struct ocfs2_super *osb,
 	if (status < 0)
 		mlog_errno(status);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -921,8 +898,6 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 	struct inode *inode = NULL;
 	int num_running_trans = 0;
 
-	mlog_entry_void();
-
 	BUG_ON(!osb);
 
 	journal = osb->journal;
@@ -939,10 +914,7 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 		BUG();
 
 	num_running_trans = atomic_read(&(osb->journal->j_num_trans));
-	if (num_running_trans > 0)
-		mlog(0, "Shutting down journal: must wait on %d "
-		     "running transactions!\n",
-		     num_running_trans);
+	trace_ocfs2_journal_shutdown(num_running_trans);
 
 	/* Do a commit_cache here. It will flush our journal, *and*
 	 * release any locks that are still held.
@@ -955,7 +927,7 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 	 * completely destroy the journal. */
 	if (osb->commit_task) {
 		/* Wait for the commit thread */
-		mlog(0, "Waiting for ocfs2commit to exit....\n");
+		trace_ocfs2_journal_shutdown_wait(osb->commit_task);
 		kthread_stop(osb->commit_task);
 		osb->commit_task = NULL;
 	}
@@ -998,7 +970,6 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 done:
 	if (inode)
 		iput(inode);
-	mlog_exit_void();
 }
 
 static void ocfs2_clear_journal_error(struct super_block *sb,
@@ -1024,8 +995,6 @@ int ocfs2_journal_load(struct ocfs2_journal *journal, int local, int replayed)
 	int status = 0;
 	struct ocfs2_super *osb;
 
-	mlog_entry_void();
-
 	BUG_ON(!journal);
 
 	osb = journal->j_osb;
@@ -1059,7 +1028,6 @@ int ocfs2_journal_load(struct ocfs2_journal *journal, int local, int replayed)
 		osb->commit_task = NULL;
 
 done:
-	mlog_exit(status);
 	return status;
 }
 
@@ -1070,8 +1038,6 @@ int ocfs2_journal_wipe(struct ocfs2_journal *journal, int full)
 {
 	int status;
 
-	mlog_entry_void();
-
 	BUG_ON(!journal);
 
 	status = jbd2_journal_wipe(journal->j_journal, full);
@@ -1085,7 +1051,6 @@ int ocfs2_journal_wipe(struct ocfs2_journal *journal, int full)
 		mlog_errno(status);
 
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -1124,8 +1089,6 @@ static int ocfs2_force_read_journal(struct inode *inode)
 #define CONCURRENT_JOURNAL_FILL 32ULL
 	struct buffer_head *bhs[CONCURRENT_JOURNAL_FILL];
 
-	mlog_entry_void();
-
 	memset(bhs, 0, sizeof(struct buffer_head *) * CONCURRENT_JOURNAL_FILL);
 
 	num_blocks = ocfs2_blocks_for_bytes(inode->i_sb, inode->i_size);
@@ -1161,7 +1124,6 @@ static int ocfs2_force_read_journal(struct inode *inode)
 bail:
 	for(i = 0; i < CONCURRENT_JOURNAL_FILL; i++)
 		brelse(bhs[i]);
-	mlog_exit(status);
 	return status;
 }
 
@@ -1185,7 +1147,7 @@ struct ocfs2_la_recovery_item {
  */
 void ocfs2_complete_recovery(struct work_struct *work)
 {
-	int ret;
+	int ret = 0;
 	struct ocfs2_journal *journal =
 		container_of(work, struct ocfs2_journal, j_recovery_work);
 	struct ocfs2_super *osb = journal->j_osb;
@@ -1194,9 +1156,8 @@ void ocfs2_complete_recovery(struct work_struct *work)
 	struct ocfs2_quota_recovery *qrec;
 	LIST_HEAD(tmp_la_list);
 
-	mlog_entry_void();
-
-	mlog(0, "completing recovery from keventd\n");
+	trace_ocfs2_complete_recovery(
+		(unsigned long long)OCFS2_I(journal->j_inode)->ip_blkno);
 
 	spin_lock(&journal->j_lock);
 	list_splice_init(&journal->j_la_cleanups, &tmp_la_list);
@@ -1205,15 +1166,18 @@ void ocfs2_complete_recovery(struct work_struct *work)
 	list_for_each_entry_safe(item, n, &tmp_la_list, lri_list) {
 		list_del_init(&item->lri_list);
 
-		mlog(0, "Complete recovery for slot %d\n", item->lri_slot);
-
 		ocfs2_wait_on_quotas(osb);
 
 		la_dinode = item->lri_la_dinode;
-		if (la_dinode) {
-			mlog(0, "Clean up local alloc %llu\n",
-			     (unsigned long long)le64_to_cpu(la_dinode->i_blkno));
+		tl_dinode = item->lri_tl_dinode;
+		qrec = item->lri_qrec;
+
+		trace_ocfs2_complete_recovery_slot(item->lri_slot,
+			la_dinode ? le64_to_cpu(la_dinode->i_blkno) : 0,
+			tl_dinode ? le64_to_cpu(tl_dinode->i_blkno) : 0,
+			qrec);
 
+		if (la_dinode) {
 			ret = ocfs2_complete_local_alloc_recovery(osb,
 								  la_dinode);
 			if (ret < 0)
@@ -1222,11 +1186,7 @@ void ocfs2_complete_recovery(struct work_struct *work)
 			kfree(la_dinode);
 		}
 
-		tl_dinode = item->lri_tl_dinode;
 		if (tl_dinode) {
-			mlog(0, "Clean up truncate log %llu\n",
-			     (unsigned long long)le64_to_cpu(tl_dinode->i_blkno));
-
 			ret = ocfs2_complete_truncate_log_recovery(osb,
 								   tl_dinode);
 			if (ret < 0)
@@ -1239,9 +1199,7 @@ void ocfs2_complete_recovery(struct work_struct *work)
 		if (ret < 0)
 			mlog_errno(ret);
 
-		qrec = item->lri_qrec;
 		if (qrec) {
-			mlog(0, "Recovering quota files");
 			ret = ocfs2_finish_quota_recovery(osb, qrec,
 							  item->lri_slot);
 			if (ret < 0)
@@ -1252,8 +1210,7 @@ void ocfs2_complete_recovery(struct work_struct *work)
 		kfree(item);
 	}
 
-	mlog(0, "Recovery completion\n");
-	mlog_exit_void();
+	trace_ocfs2_complete_recovery_end(ret);
 }
 
 /* NOTE: This function always eats your references to la_dinode and
@@ -1339,8 +1296,6 @@ static int __ocfs2_recovery_thread(void *arg)
 	int rm_quota_used = 0, i;
 	struct ocfs2_quota_recovery *qrec;
 
-	mlog_entry_void();
-
 	status = ocfs2_wait_on_mount(osb);
 	if (status < 0) {
 		goto bail;
@@ -1372,15 +1327,12 @@ restart:
 		 * clear it until ocfs2_recover_node() has succeeded. */
 		node_num = rm->rm_entries[0];
 		spin_unlock(&osb->osb_lock);
-		mlog(0, "checking node %d\n", node_num);
 		slot_num = ocfs2_node_num_to_slot(osb, node_num);
+		trace_ocfs2_recovery_thread_node(node_num, slot_num);
 		if (slot_num == -ENOENT) {
 			status = 0;
-			mlog(0, "no slot for this node, so no recovery"
-			     "required.\n");
 			goto skip_recovery;
 		}
-		mlog(0, "node %d was using slot %d\n", node_num, slot_num);
 
 		/* It is a bit subtle with quota recovery. We cannot do it
 		 * immediately because we have to obtain cluster locks from
@@ -1407,7 +1359,7 @@ skip_recovery:
 		spin_lock(&osb->osb_lock);
 	}
 	spin_unlock(&osb->osb_lock);
-	mlog(0, "All nodes recovered\n");
+	trace_ocfs2_recovery_thread_end(status);
 
 	/* Refresh all journal recovery generations from disk */
 	status = ocfs2_check_journals_nolocks(osb);
@@ -1451,7 +1403,6 @@ bail:
 	if (rm_quota)
 		kfree(rm_quota);
 
-	mlog_exit(status);
 	/* no one is callint kthread_stop() for us so the kthread() api
 	 * requires that we call do_exit().  And it isn't exported, but
 	 * complete_and_exit() seems to be a minimal wrapper around it. */
@@ -1461,19 +1412,15 @@ bail:
 
 void ocfs2_recovery_thread(struct ocfs2_super *osb, int node_num)
 {
-	mlog_entry("(node_num=%d, osb->node_num = %d)\n",
-		   node_num, osb->node_num);
-
 	mutex_lock(&osb->recovery_lock);
-	if (osb->disable_recovery)
-		goto out;
 
-	/* People waiting on recovery will wait on
-	 * the recovery map to empty. */
-	if (ocfs2_recovery_map_set(osb, node_num))
-		mlog(0, "node %d already in recovery map.\n", node_num);
+	trace_ocfs2_recovery_thread(node_num, osb->node_num,
+		osb->disable_recovery, osb->recovery_thread_task,
+		osb->disable_recovery ?
+		-1 : ocfs2_recovery_map_set(osb, node_num));
 
-	mlog(0, "starting recovery thread...\n");
+	if (osb->disable_recovery)
+		goto out;
 
 	if (osb->recovery_thread_task)
 		goto out;
@@ -1488,8 +1435,6 @@ void ocfs2_recovery_thread(struct ocfs2_super *osb, int node_num)
 out:
 	mutex_unlock(&osb->recovery_lock);
 	wake_up(&osb->recovery_event);
-
-	mlog_exit_void();
 }
 
 static int ocfs2_read_journal_inode(struct ocfs2_super *osb,
@@ -1563,7 +1508,7 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb,
 	 * If not, it needs recovery.
 	 */
 	if (osb->slot_recovery_generations[slot_num] != slot_reco_gen) {
-		mlog(0, "Slot %u already recovered (old/new=%u/%u)\n", slot_num,
+		trace_ocfs2_replay_journal_recovered(slot_num,
 		     osb->slot_recovery_generations[slot_num], slot_reco_gen);
 		osb->slot_recovery_generations[slot_num] = slot_reco_gen;
 		status = -EBUSY;
@@ -1574,7 +1519,7 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb,
 
 	status = ocfs2_inode_lock_full(inode, &bh, 1, OCFS2_META_LOCK_RECOVERY);
 	if (status < 0) {
-		mlog(0, "status returned from ocfs2_inode_lock=%d\n", status);
+		trace_ocfs2_replay_journal_lock_err(status);
 		if (status != -ERESTARTSYS)
 			mlog(ML_ERROR, "Could not lock journal!\n");
 		goto done;
@@ -1587,7 +1532,7 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb,
 	slot_reco_gen = ocfs2_get_recovery_generation(fe);
 
 	if (!(flags & OCFS2_JOURNAL_DIRTY_FL)) {
-		mlog(0, "No recovery required for node %d\n", node_num);
+		trace_ocfs2_replay_journal_skip(node_num);
 		/* Refresh recovery generation for the slot */
 		osb->slot_recovery_generations[slot_num] = slot_reco_gen;
 		goto done;
@@ -1608,7 +1553,6 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb,
 		goto done;
 	}
 
-	mlog(0, "calling journal_init_inode\n");
 	journal = jbd2_journal_init_inode(inode);
 	if (journal == NULL) {
 		mlog(ML_ERROR, "Linux journal layer error\n");
@@ -1628,7 +1572,6 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb,
 	ocfs2_clear_journal_error(osb->sb, journal, slot_num);
 
 	/* wipe the journal */
-	mlog(0, "flushing the journal.\n");
 	jbd2_journal_lock_updates(journal);
 	status = jbd2_journal_flush(journal);
 	jbd2_journal_unlock_updates(journal);
@@ -1665,7 +1608,6 @@ done:
 
 	brelse(bh);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1688,8 +1630,7 @@ static int ocfs2_recover_node(struct ocfs2_super *osb,
 	struct ocfs2_dinode *la_copy = NULL;
 	struct ocfs2_dinode *tl_copy = NULL;
 
-	mlog_entry("(node_num=%d, slot_num=%d, osb->node_num = %d)\n",
-		   node_num, slot_num, osb->node_num);
+	trace_ocfs2_recover_node(node_num, slot_num, osb->node_num);
 
 	/* Should not ever be called to recover ourselves -- in that
 	 * case we should've called ocfs2_journal_load instead. */
@@ -1698,9 +1639,7 @@ static int ocfs2_recover_node(struct ocfs2_super *osb,
 	status = ocfs2_replay_journal(osb, node_num, slot_num);
 	if (status < 0) {
 		if (status == -EBUSY) {
-			mlog(0, "Skipping recovery for slot %u (node %u) "
-			     "as another node has recovered it\n", slot_num,
-			     node_num);
+			trace_ocfs2_recover_node_skip(slot_num, node_num);
 			status = 0;
 			goto done;
 		}
@@ -1735,7 +1674,6 @@ static int ocfs2_recover_node(struct ocfs2_super *osb,
 	status = 0;
 done:
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1808,8 +1746,8 @@ int ocfs2_mark_dead_nodes(struct ocfs2_super *osb)
 		spin_lock(&osb->osb_lock);
 		osb->slot_recovery_generations[i] = gen;
 
-		mlog(0, "Slot %u recovery generation is %u\n", i,
-		     osb->slot_recovery_generations[i]);
+		trace_ocfs2_mark_dead_nodes(i,
+					    osb->slot_recovery_generations[i]);
 
 		if (i == osb->slot_num) {
 			spin_unlock(&osb->osb_lock);
@@ -1845,7 +1783,6 @@ int ocfs2_mark_dead_nodes(struct ocfs2_super *osb)
 
 	status = 0;
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -1884,11 +1821,12 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 
 	os = &osb->osb_orphan_scan;
 
-	mlog(0, "Begin orphan scan\n");
-
 	if (atomic_read(&os->os_state) == ORPHAN_SCAN_INACTIVE)
 		goto out;
 
+	trace_ocfs2_queue_orphan_scan_begin(os->os_count, os->os_seqno,
+					    atomic_read(&os->os_state));
+
 	status = ocfs2_orphan_scan_lock(osb, &seqno);
 	if (status < 0) {
 		if (status != -EAGAIN)
@@ -1918,7 +1856,8 @@ void ocfs2_queue_orphan_scan(struct ocfs2_super *osb)
 unlock:
 	ocfs2_orphan_scan_unlock(osb, seqno);
 out:
-	mlog(0, "Orphan scan completed\n");
+	trace_ocfs2_queue_orphan_scan_end(os->os_count, os->os_seqno,
+					  atomic_read(&os->os_state));
 	return;
 }
 
@@ -2002,8 +1941,7 @@ static int ocfs2_orphan_filldir(void *priv, const char *name, int name_len,
 	if (IS_ERR(iter))
 		return 0;
 
-	mlog(0, "queue orphan %llu\n",
-	     (unsigned long long)OCFS2_I(iter)->ip_blkno);
+	trace_ocfs2_orphan_filldir((unsigned long long)OCFS2_I(iter)->ip_blkno);
 	/* No locking is required for the next_orphan queue as there
 	 * is only ever a single process doing orphan recovery. */
 	OCFS2_I(iter)->ip_next_orphan = p->head;
@@ -2119,7 +2057,7 @@ static int ocfs2_recover_orphans(struct ocfs2_super *osb,
 	struct inode *iter;
 	struct ocfs2_inode_info *oi;
 
-	mlog(0, "Recover inodes from orphan dir in slot %d\n", slot);
+	trace_ocfs2_recover_orphans(slot);
 
 	ocfs2_mark_recovering_orphan_dir(osb, slot);
 	ret = ocfs2_queue_orphans(osb, slot, &inode);
@@ -2132,7 +2070,8 @@ static int ocfs2_recover_orphans(struct ocfs2_super *osb,
 
 	while (inode) {
 		oi = OCFS2_I(inode);
-		mlog(0, "iput orphan %llu\n", (unsigned long long)oi->ip_blkno);
+		trace_ocfs2_recover_orphans_iput(
+					(unsigned long long)oi->ip_blkno);
 
 		iter = oi->ip_next_orphan;
 
@@ -2170,6 +2109,7 @@ static int __ocfs2_wait_on_mount(struct ocfs2_super *osb, int quota)
 	 * MOUNTED flag, but this is set right before
 	 * dismount_volume() so we can trust it. */
 	if (atomic_read(&osb->vol_state) == VOLUME_DISABLED) {
+		trace_ocfs2_wait_on_mount(VOLUME_DISABLED);
 		mlog(0, "mount error, exiting!\n");
 		return -EBUSY;
 	}
diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h
index 43e56b97f9c..6180da1e37e 100644
--- a/fs/ocfs2/journal.h
+++ b/fs/ocfs2/journal.h
@@ -405,9 +405,9 @@ static inline int ocfs2_remove_extent_credits(struct super_block *sb)
 	       ocfs2_quota_trans_credits(sb);
 }
 
-/* data block for new dir/symlink, 2 for bitmap updates (bitmap fe +
- * bitmap block for the new bit) dx_root update for free list */
-#define OCFS2_DIR_LINK_ADDITIONAL_CREDITS (1 + 2 + 1)
+/* data block for new dir/symlink, allocation of directory block, dx_root
+ * update for free list */
+#define OCFS2_DIR_LINK_ADDITIONAL_CREDITS (1 + OCFS2_SUBALLOC_ALLOC + 1)
 
 static inline int ocfs2_add_dir_index_credits(struct super_block *sb)
 {
diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c
index ec6adbf8f55..210c3523754 100644
--- a/fs/ocfs2/localalloc.c
+++ b/fs/ocfs2/localalloc.c
@@ -29,7 +29,6 @@
 #include <linux/highmem.h>
 #include <linux/bitops.h>
 
-#define MLOG_MASK_PREFIX ML_DISK_ALLOC
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -43,6 +42,7 @@
 #include "suballoc.h"
 #include "super.h"
 #include "sysfile.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -201,8 +201,7 @@ void ocfs2_la_set_sizes(struct ocfs2_super *osb, int requested_mb)
 	la_max_mb = ocfs2_clusters_to_megabytes(sb,
 						ocfs2_local_alloc_size(sb) * 8);
 
-	mlog(0, "requested: %dM, max: %uM, default: %uM\n",
-	     requested_mb, la_max_mb, la_default_mb);
+	trace_ocfs2_la_set_sizes(requested_mb, la_max_mb, la_default_mb);
 
 	if (requested_mb == -1) {
 		/* No user request - use defaults */
@@ -276,8 +275,8 @@ int ocfs2_alloc_should_use_local(struct ocfs2_super *osb, u64 bits)
 
 	ret = 1;
 bail:
-	mlog(0, "state=%d, bits=%llu, la_bits=%d, ret=%d\n",
-	     osb->local_alloc_state, (unsigned long long)bits, la_bits, ret);
+	trace_ocfs2_alloc_should_use_local(
+	     (unsigned long long)bits, osb->local_alloc_state, la_bits, ret);
 	spin_unlock(&osb->osb_lock);
 	return ret;
 }
@@ -291,8 +290,6 @@ int ocfs2_load_local_alloc(struct ocfs2_super *osb)
 	struct inode *inode = NULL;
 	struct ocfs2_local_alloc *la;
 
-	mlog_entry_void();
-
 	if (osb->local_alloc_bits == 0)
 		goto bail;
 
@@ -364,9 +361,10 @@ bail:
 	if (inode)
 		iput(inode);
 
-	mlog(0, "Local alloc window bits = %d\n", osb->local_alloc_bits);
+	trace_ocfs2_load_local_alloc(osb->local_alloc_bits);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -388,8 +386,6 @@ void ocfs2_shutdown_local_alloc(struct ocfs2_super *osb)
 	struct ocfs2_dinode *alloc_copy = NULL;
 	struct ocfs2_dinode *alloc = NULL;
 
-	mlog_entry_void();
-
 	cancel_delayed_work(&osb->la_enable_wq);
 	flush_workqueue(ocfs2_wq);
 
@@ -482,8 +478,6 @@ out:
 
 	if (alloc_copy)
 		kfree(alloc_copy);
-
-	mlog_exit_void();
 }
 
 /*
@@ -502,7 +496,7 @@ int ocfs2_begin_local_alloc_recovery(struct ocfs2_super *osb,
 	struct inode *inode = NULL;
 	struct ocfs2_dinode *alloc;
 
-	mlog_entry("(slot_num = %d)\n", slot_num);
+	trace_ocfs2_begin_local_alloc_recovery(slot_num);
 
 	*alloc_copy = NULL;
 
@@ -552,7 +546,8 @@ bail:
 		iput(inode);
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -570,8 +565,6 @@ int ocfs2_complete_local_alloc_recovery(struct ocfs2_super *osb,
 	struct buffer_head *main_bm_bh = NULL;
 	struct inode *main_bm_inode;
 
-	mlog_entry_void();
-
 	main_bm_inode = ocfs2_get_system_file_inode(osb,
 						    GLOBAL_BITMAP_SYSTEM_INODE,
 						    OCFS2_INVALID_SLOT);
@@ -620,7 +613,8 @@ out_mutex:
 out:
 	if (!status)
 		ocfs2_init_steal_slots(osb);
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -640,8 +634,6 @@ int ocfs2_reserve_local_alloc_bits(struct ocfs2_super *osb,
 	struct inode *local_alloc_inode;
 	unsigned int free_bits;
 
-	mlog_entry_void();
-
 	BUG_ON(!ac);
 
 	local_alloc_inode =
@@ -712,10 +704,6 @@ int ocfs2_reserve_local_alloc_bits(struct ocfs2_super *osb,
 			goto bail;
 	}
 
-	if (ac->ac_max_block)
-		mlog(0, "Calling in_range for max block %llu\n",
-		     (unsigned long long)ac->ac_max_block);
-
 	ac->ac_inode = local_alloc_inode;
 	/* We should never use localalloc from another slot */
 	ac->ac_alloc_slot = osb->slot_num;
@@ -729,10 +717,12 @@ bail:
 		iput(local_alloc_inode);
 	}
 
-	mlog(0, "bits=%d, slot=%d, ret=%d\n", bits_wanted, osb->slot_num,
-	     status);
+	trace_ocfs2_reserve_local_alloc_bits(
+		(unsigned long long)ac->ac_max_block,
+		bits_wanted, osb->slot_num, status);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -749,7 +739,6 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb,
 	struct ocfs2_dinode *alloc;
 	struct ocfs2_local_alloc *la;
 
-	mlog_entry_void();
 	BUG_ON(ac->ac_which != OCFS2_AC_USE_LOCAL);
 
 	local_alloc_inode = ac->ac_inode;
@@ -788,7 +777,8 @@ int ocfs2_claim_local_alloc_bits(struct ocfs2_super *osb,
 	ocfs2_journal_dirty(handle, osb->local_alloc_bh);
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -799,13 +789,11 @@ static u32 ocfs2_local_alloc_count_bits(struct ocfs2_dinode *alloc)
 	u32 count = 0;
 	struct ocfs2_local_alloc *la = OCFS2_LOCAL_ALLOC(alloc);
 
-	mlog_entry_void();
-
 	buffer = la->la_bitmap;
 	for (i = 0; i < le16_to_cpu(la->la_size); i++)
 		count += hweight8(buffer[i]);
 
-	mlog_exit(count);
+	trace_ocfs2_local_alloc_count_bits(count);
 	return count;
 }
 
@@ -820,10 +808,7 @@ static int ocfs2_local_alloc_find_clear_bits(struct ocfs2_super *osb,
 	void *bitmap = NULL;
 	struct ocfs2_reservation_map *resmap = &osb->osb_la_resmap;
 
-	mlog_entry("(numbits wanted = %u)\n", *numbits);
-
 	if (!alloc->id1.bitmap1.i_total) {
-		mlog(0, "No bits in my window!\n");
 		bitoff = -1;
 		goto bail;
 	}
@@ -883,8 +868,7 @@ static int ocfs2_local_alloc_find_clear_bits(struct ocfs2_super *osb,
 		}
 	}
 
-	mlog(0, "Exiting loop, bitoff = %d, numfound = %d\n", bitoff,
-	     numfound);
+	trace_ocfs2_local_alloc_find_clear_bits_search_bitmap(bitoff, numfound);
 
 	if (numfound == *numbits)
 		bitoff = startoff - numfound;
@@ -895,7 +879,10 @@ bail:
 	if (local_resv)
 		ocfs2_resv_discard(resmap, resv);
 
-	mlog_exit(bitoff);
+	trace_ocfs2_local_alloc_find_clear_bits(*numbits,
+		le32_to_cpu(alloc->id1.bitmap1.i_total),
+		bitoff, numfound);
+
 	return bitoff;
 }
 
@@ -903,15 +890,12 @@ static void ocfs2_clear_local_alloc(struct ocfs2_dinode *alloc)
 {
 	struct ocfs2_local_alloc *la = OCFS2_LOCAL_ALLOC(alloc);
 	int i;
-	mlog_entry_void();
 
 	alloc->id1.bitmap1.i_total = 0;
 	alloc->id1.bitmap1.i_used = 0;
 	la->la_bm_off = 0;
 	for(i = 0; i < le16_to_cpu(la->la_size); i++)
 		la->la_bitmap[i] = 0;
-
-	mlog_exit_void();
 }
 
 #if 0
@@ -952,18 +936,16 @@ static int ocfs2_sync_local_to_main(struct ocfs2_super *osb,
 	void *bitmap;
 	struct ocfs2_local_alloc *la = OCFS2_LOCAL_ALLOC(alloc);
 
-	mlog_entry("total = %u, used = %u\n",
-		   le32_to_cpu(alloc->id1.bitmap1.i_total),
-		   le32_to_cpu(alloc->id1.bitmap1.i_used));
+	trace_ocfs2_sync_local_to_main(
+	     le32_to_cpu(alloc->id1.bitmap1.i_total),
+	     le32_to_cpu(alloc->id1.bitmap1.i_used));
 
 	if (!alloc->id1.bitmap1.i_total) {
-		mlog(0, "nothing to sync!\n");
 		goto bail;
 	}
 
 	if (le32_to_cpu(alloc->id1.bitmap1.i_used) ==
 	    le32_to_cpu(alloc->id1.bitmap1.i_total)) {
-		mlog(0, "all bits were taken!\n");
 		goto bail;
 	}
 
@@ -985,8 +967,7 @@ static int ocfs2_sync_local_to_main(struct ocfs2_super *osb,
 				ocfs2_clusters_to_blocks(osb->sb,
 							 start - count);
 
-			mlog(0, "freeing %u bits starting at local alloc bit "
-			     "%u (la_start_blk = %llu, blkno = %llu)\n",
+			trace_ocfs2_sync_local_to_main_free(
 			     count, start - count,
 			     (unsigned long long)la_start_blk,
 			     (unsigned long long)blkno);
@@ -1007,7 +988,8 @@ static int ocfs2_sync_local_to_main(struct ocfs2_super *osb,
 	}
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1132,7 +1114,8 @@ bail:
 		*ac = NULL;
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1148,17 +1131,12 @@ static int ocfs2_local_alloc_new_window(struct ocfs2_super *osb,
 	struct ocfs2_dinode *alloc = NULL;
 	struct ocfs2_local_alloc *la;
 
-	mlog_entry_void();
-
 	alloc = (struct ocfs2_dinode *) osb->local_alloc_bh->b_data;
 	la = OCFS2_LOCAL_ALLOC(alloc);
 
-	if (alloc->id1.bitmap1.i_total)
-		mlog(0, "asking me to alloc a new window over a non-empty "
-		     "one\n");
-
-	mlog(0, "Allocating %u clusters for a new window.\n",
-	     osb->local_alloc_bits);
+	trace_ocfs2_local_alloc_new_window(
+		le32_to_cpu(alloc->id1.bitmap1.i_total),
+		osb->local_alloc_bits);
 
 	/* Instruct the allocation code to try the most recently used
 	 * cluster group. We'll re-record the group used this pass
@@ -1220,13 +1198,13 @@ retry_enospc:
 	ocfs2_resmap_restart(&osb->osb_la_resmap, cluster_count,
 			     OCFS2_LOCAL_ALLOC(alloc)->la_bitmap);
 
-	mlog(0, "New window allocated:\n");
-	mlog(0, "window la_bm_off = %u\n",
-	     OCFS2_LOCAL_ALLOC(alloc)->la_bm_off);
-	mlog(0, "window bits = %u\n", le32_to_cpu(alloc->id1.bitmap1.i_total));
+	trace_ocfs2_local_alloc_new_window_result(
+		OCFS2_LOCAL_ALLOC(alloc)->la_bm_off,
+		le32_to_cpu(alloc->id1.bitmap1.i_total));
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1243,8 +1221,6 @@ static int ocfs2_local_alloc_slide_window(struct ocfs2_super *osb,
 	struct ocfs2_dinode *alloc_copy = NULL;
 	struct ocfs2_alloc_context *ac = NULL;
 
-	mlog_entry_void();
-
 	ocfs2_recalc_la_window(osb, OCFS2_LA_EVENT_SLIDE);
 
 	/* This will lock the main bitmap for us. */
@@ -1324,7 +1300,8 @@ bail:
 	if (ac)
 		ocfs2_free_alloc_context(ac);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
diff --git a/fs/ocfs2/locks.c b/fs/ocfs2/locks.c
index b5cb3ede940..e57c804069e 100644
--- a/fs/ocfs2/locks.c
+++ b/fs/ocfs2/locks.c
@@ -26,7 +26,6 @@
 #include <linux/fs.h>
 #include <linux/fcntl.h>
 
-#define MLOG_MASK_PREFIX ML_INODE
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c
index 7e32db9c2c9..3e9393ca39e 100644
--- a/fs/ocfs2/mmap.c
+++ b/fs/ocfs2/mmap.c
@@ -31,7 +31,6 @@
 #include <linux/signal.h>
 #include <linux/rbtree.h>
 
-#define MLOG_MASK_PREFIX ML_FILE_IO
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -42,6 +41,7 @@
 #include "inode.h"
 #include "mmap.h"
 #include "super.h"
+#include "ocfs2_trace.h"
 
 
 static int ocfs2_fault(struct vm_area_struct *area, struct vm_fault *vmf)
@@ -49,13 +49,12 @@ static int ocfs2_fault(struct vm_area_struct *area, struct vm_fault *vmf)
 	sigset_t oldset;
 	int ret;
 
-	mlog_entry("(area=%p, page offset=%lu)\n", area, vmf->pgoff);
-
 	ocfs2_block_signals(&oldset);
 	ret = filemap_fault(area, vmf);
 	ocfs2_unblock_signals(&oldset);
 
-	mlog_exit_ptr(vmf->page);
+	trace_ocfs2_fault(OCFS2_I(area->vm_file->f_mapping->host)->ip_blkno,
+			  area, vmf->page, vmf->pgoff);
 	return ret;
 }
 
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 849fb4a2e81..b6fa9eb2d29 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -42,7 +42,6 @@
 #include <linux/highmem.h>
 #include <linux/quotaops.h>
 
-#define MLOG_MASK_PREFIX ML_NAMEI
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -63,6 +62,7 @@
 #include "uptodate.h"
 #include "xattr.h"
 #include "acl.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -106,17 +106,15 @@ static struct dentry *ocfs2_lookup(struct inode *dir, struct dentry *dentry,
 	struct dentry *ret;
 	struct ocfs2_inode_info *oi;
 
-	mlog_entry("(0x%p, 0x%p, '%.*s')\n", dir, dentry,
-		   dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_lookup(dir, dentry, dentry->d_name.len,
+			   dentry->d_name.name,
+			   (unsigned long long)OCFS2_I(dir)->ip_blkno, 0);
 
 	if (dentry->d_name.len > OCFS2_MAX_FILENAME_LEN) {
 		ret = ERR_PTR(-ENAMETOOLONG);
 		goto bail;
 	}
 
-	mlog(0, "find name %.*s in directory %llu\n", dentry->d_name.len,
-	     dentry->d_name.name, (unsigned long long)OCFS2_I(dir)->ip_blkno);
-
 	status = ocfs2_inode_lock_nested(dir, NULL, 0, OI_LS_PARENT);
 	if (status < 0) {
 		if (status != -ENOENT)
@@ -182,7 +180,7 @@ bail_unlock:
 
 bail:
 
-	mlog_exit_ptr(ret);
+	trace_ocfs2_lookup_ret(ret);
 
 	return ret;
 }
@@ -235,9 +233,9 @@ static int ocfs2_mknod(struct inode *dir,
 	sigset_t oldset;
 	int did_block_signals = 0;
 
-	mlog_entry("(0x%p, 0x%p, %d, %lu, '%.*s')\n", dir, dentry, mode,
-		   (unsigned long)dev, dentry->d_name.len,
-		   dentry->d_name.name);
+	trace_ocfs2_mknod(dir, dentry, dentry->d_name.len, dentry->d_name.name,
+			  (unsigned long long)OCFS2_I(dir)->ip_blkno,
+			  (unsigned long)dev, mode);
 
 	dquot_initialize(dir);
 
@@ -354,10 +352,6 @@ static int ocfs2_mknod(struct inode *dir,
 		goto leave;
 	did_quota_inode = 1;
 
-	mlog_entry("(0x%p, 0x%p, %d, %lu, '%.*s')\n", dir, dentry,
-		   inode->i_mode, (unsigned long)dev, dentry->d_name.len,
-		   dentry->d_name.name);
-
 	/* do the real work now. */
 	status = ocfs2_mknod_locked(osb, dir, inode, dev,
 				    &new_fe_bh, parent_fe_bh, handle,
@@ -436,9 +430,6 @@ leave:
 	if (did_block_signals)
 		ocfs2_unblock_signals(&oldset);
 
-	if (status == -ENOSPC)
-		mlog(0, "Disk is full\n");
-
 	brelse(new_fe_bh);
 	brelse(parent_fe_bh);
 	kfree(si.name);
@@ -466,7 +457,8 @@ leave:
 		iput(inode);
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 
 	return status;
 }
@@ -577,7 +569,8 @@ leave:
 		}
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -615,10 +608,11 @@ static int ocfs2_mkdir(struct inode *dir,
 {
 	int ret;
 
-	mlog_entry("(0x%p, 0x%p, %d, '%.*s')\n", dir, dentry, mode,
-		   dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_mkdir(dir, dentry, dentry->d_name.len, dentry->d_name.name,
+			  OCFS2_I(dir)->ip_blkno, mode);
 	ret = ocfs2_mknod(dir, dentry, mode | S_IFDIR, 0);
-	mlog_exit(ret);
+	if (ret)
+		mlog_errno(ret);
 
 	return ret;
 }
@@ -630,10 +624,11 @@ static int ocfs2_create(struct inode *dir,
 {
 	int ret;
 
-	mlog_entry("(0x%p, 0x%p, %d, '%.*s')\n", dir, dentry, mode,
-		   dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_create(dir, dentry, dentry->d_name.len, dentry->d_name.name,
+			   (unsigned long long)OCFS2_I(dir)->ip_blkno, mode);
 	ret = ocfs2_mknod(dir, dentry, mode | S_IFREG, 0);
-	mlog_exit(ret);
+	if (ret)
+		mlog_errno(ret);
 
 	return ret;
 }
@@ -652,9 +647,9 @@ static int ocfs2_link(struct dentry *old_dentry,
 	struct ocfs2_dir_lookup_result lookup = { NULL, };
 	sigset_t oldset;
 
-	mlog_entry("(inode=%lu, old='%.*s' new='%.*s')\n", inode->i_ino,
-		   old_dentry->d_name.len, old_dentry->d_name.name,
-		   dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_link((unsigned long long)OCFS2_I(inode)->ip_blkno,
+			 old_dentry->d_name.len, old_dentry->d_name.name,
+			 dentry->d_name.len, dentry->d_name.name);
 
 	if (S_ISDIR(inode->i_mode))
 		return -EPERM;
@@ -757,7 +752,8 @@ out:
 
 	ocfs2_free_dir_lookup_result(&lookup);
 
-	mlog_exit(err);
+	if (err)
+		mlog_errno(err);
 
 	return err;
 }
@@ -809,19 +805,17 @@ static int ocfs2_unlink(struct inode *dir,
 	struct ocfs2_dir_lookup_result lookup = { NULL, };
 	struct ocfs2_dir_lookup_result orphan_insert = { NULL, };
 
-	mlog_entry("(0x%p, 0x%p, '%.*s')\n", dir, dentry,
-		   dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_unlink(dir, dentry, dentry->d_name.len,
+			   dentry->d_name.name,
+			   (unsigned long long)OCFS2_I(dir)->ip_blkno,
+			   (unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 	dquot_initialize(dir);
 
 	BUG_ON(dentry->d_parent->d_inode != dir);
 
-	mlog(0, "ino = %llu\n", (unsigned long long)OCFS2_I(inode)->ip_blkno);
-
-	if (inode == osb->root_inode) {
-		mlog(0, "Cannot delete the root directory\n");
+	if (inode == osb->root_inode)
 		return -EPERM;
-	}
 
 	status = ocfs2_inode_lock_nested(dir, &parent_node_bh, 1,
 					 OI_LS_PARENT);
@@ -843,9 +837,10 @@ static int ocfs2_unlink(struct inode *dir,
 	if (OCFS2_I(inode)->ip_blkno != blkno) {
 		status = -ENOENT;
 
-		mlog(0, "ip_blkno %llu != dirent blkno %llu ip_flags = %x\n",
-		     (unsigned long long)OCFS2_I(inode)->ip_blkno,
-		     (unsigned long long)blkno, OCFS2_I(inode)->ip_flags);
+		trace_ocfs2_unlink_noent(
+				(unsigned long long)OCFS2_I(inode)->ip_blkno,
+				(unsigned long long)blkno,
+				OCFS2_I(inode)->ip_flags);
 		goto leave;
 	}
 
@@ -954,7 +949,8 @@ leave:
 	ocfs2_free_dir_lookup_result(&orphan_insert);
 	ocfs2_free_dir_lookup_result(&lookup);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 
 	return status;
 }
@@ -975,9 +971,8 @@ static int ocfs2_double_lock(struct ocfs2_super *osb,
 	struct buffer_head **tmpbh;
 	struct inode *tmpinode;
 
-	mlog_entry("(inode1 = %llu, inode2 = %llu)\n",
-		   (unsigned long long)oi1->ip_blkno,
-		   (unsigned long long)oi2->ip_blkno);
+	trace_ocfs2_double_lock((unsigned long long)oi1->ip_blkno,
+				(unsigned long long)oi2->ip_blkno);
 
 	if (*bh1)
 		*bh1 = NULL;
@@ -988,7 +983,6 @@ static int ocfs2_double_lock(struct ocfs2_super *osb,
 	if (oi1->ip_blkno != oi2->ip_blkno) {
 		if (oi1->ip_blkno < oi2->ip_blkno) {
 			/* switch id1 and id2 around */
-			mlog(0, "switching them around...\n");
 			tmpbh = bh2;
 			bh2 = bh1;
 			bh1 = tmpbh;
@@ -1024,8 +1018,13 @@ static int ocfs2_double_lock(struct ocfs2_super *osb,
 			mlog_errno(status);
 	}
 
+	trace_ocfs2_double_lock_end(
+			(unsigned long long)OCFS2_I(inode1)->ip_blkno,
+			(unsigned long long)OCFS2_I(inode2)->ip_blkno);
+
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1067,10 +1066,9 @@ static int ocfs2_rename(struct inode *old_dir,
 	/* At some point it might be nice to break this function up a
 	 * bit. */
 
-	mlog_entry("(0x%p, 0x%p, 0x%p, 0x%p, from='%.*s' to='%.*s')\n",
-		   old_dir, old_dentry, new_dir, new_dentry,
-		   old_dentry->d_name.len, old_dentry->d_name.name,
-		   new_dentry->d_name.len, new_dentry->d_name.name);
+	trace_ocfs2_rename(old_dir, old_dentry, new_dir, new_dentry,
+			   old_dentry->d_name.len, old_dentry->d_name.name,
+			   new_dentry->d_name.len, new_dentry->d_name.name);
 
 	dquot_initialize(old_dir);
 	dquot_initialize(new_dir);
@@ -1227,16 +1225,15 @@ static int ocfs2_rename(struct inode *old_dir,
 		if (!new_inode) {
 			status = -EACCES;
 
-			mlog(0, "We found an inode for name %.*s but VFS "
-			     "didn't give us one.\n", new_dentry->d_name.len,
-			     new_dentry->d_name.name);
+			trace_ocfs2_rename_target_exists(new_dentry->d_name.len,
+						new_dentry->d_name.name);
 			goto bail;
 		}
 
 		if (OCFS2_I(new_inode)->ip_blkno != newfe_blkno) {
 			status = -EACCES;
 
-			mlog(0, "Inode %llu and dir %llu disagree. flags = %x\n",
+			trace_ocfs2_rename_disagree(
 			     (unsigned long long)OCFS2_I(new_inode)->ip_blkno,
 			     (unsigned long long)newfe_blkno,
 			     OCFS2_I(new_inode)->ip_flags);
@@ -1259,8 +1256,7 @@ static int ocfs2_rename(struct inode *old_dir,
 
 		newfe = (struct ocfs2_dinode *) newfe_bh->b_data;
 
-		mlog(0, "aha rename over existing... new_blkno=%llu "
-		     "newfebh=%p bhblocknr=%llu\n",
+		trace_ocfs2_rename_over_existing(
 		     (unsigned long long)newfe_blkno, newfe_bh, newfe_bh ?
 		     (unsigned long long)newfe_bh->b_blocknr : 0ULL);
 
@@ -1476,7 +1472,8 @@ bail:
 	brelse(old_dir_bh);
 	brelse(new_dir_bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 
 	return status;
 }
@@ -1501,9 +1498,8 @@ static int ocfs2_create_symlink_data(struct ocfs2_super *osb,
 	 * write i_size + 1 bytes. */
 	blocks = (bytes_left + sb->s_blocksize - 1) >> sb->s_blocksize_bits;
 
-	mlog_entry("i_blocks = %llu, i_size = %llu, blocks = %d\n",
-			(unsigned long long)inode->i_blocks,
-			i_size_read(inode), blocks);
+	trace_ocfs2_create_symlink_data((unsigned long long)inode->i_blocks,
+					i_size_read(inode), blocks);
 
 	/* Sanity check -- make sure we're going to fit. */
 	if (bytes_left >
@@ -1579,7 +1575,8 @@ bail:
 		kfree(bhs);
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1610,8 +1607,8 @@ static int ocfs2_symlink(struct inode *dir,
 	sigset_t oldset;
 	int did_block_signals = 0;
 
-	mlog_entry("(0x%p, 0x%p, symname='%s' actual='%.*s')\n", dir,
-		   dentry, symname, dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_symlink_begin(dir, dentry, symname,
+				  dentry->d_name.len, dentry->d_name.name);
 
 	dquot_initialize(dir);
 
@@ -1713,9 +1710,10 @@ static int ocfs2_symlink(struct inode *dir,
 		goto bail;
 	did_quota_inode = 1;
 
-	mlog_entry("(0x%p, 0x%p, %d, '%.*s')\n", dir, dentry,
-		   inode->i_mode, dentry->d_name.len,
-		   dentry->d_name.name);
+	trace_ocfs2_symlink_create(dir, dentry, dentry->d_name.len,
+				   dentry->d_name.name,
+				   (unsigned long long)OCFS2_I(dir)->ip_blkno,
+				   inode->i_mode);
 
 	status = ocfs2_mknod_locked(osb, dir, inode,
 				    0, &new_fe_bh, parent_fe_bh, handle,
@@ -1835,7 +1833,8 @@ bail:
 		iput(inode);
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 
 	return status;
 }
@@ -1844,8 +1843,6 @@ static int ocfs2_blkno_stringify(u64 blkno, char *name)
 {
 	int status, namelen;
 
-	mlog_entry_void();
-
 	namelen = snprintf(name, OCFS2_ORPHAN_NAMELEN + 1, "%016llx",
 			   (long long)blkno);
 	if (namelen <= 0) {
@@ -1862,12 +1859,12 @@ static int ocfs2_blkno_stringify(u64 blkno, char *name)
 		goto bail;
 	}
 
-	mlog(0, "built filename '%s' for orphan dir (len=%d)\n", name,
-	     namelen);
+	trace_ocfs2_blkno_stringify(blkno, name, namelen);
 
 	status = 0;
 bail:
-	mlog_exit(status);
+	if (status < 0)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1980,7 +1977,8 @@ out:
 		iput(orphan_dir_inode);
 	}
 
-	mlog_exit(ret);
+	if (ret)
+		mlog_errno(ret);
 	return ret;
 }
 
@@ -1997,7 +1995,8 @@ static int ocfs2_orphan_add(struct ocfs2_super *osb,
 	struct ocfs2_dinode *orphan_fe;
 	struct ocfs2_dinode *fe = (struct ocfs2_dinode *) fe_bh->b_data;
 
-	mlog_entry("(inode->i_ino = %lu)\n", inode->i_ino);
+	trace_ocfs2_orphan_add_begin(
+				(unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 	status = ocfs2_read_inode_block(orphan_dir_inode, &orphan_dir_bh);
 	if (status < 0) {
@@ -2056,13 +2055,14 @@ static int ocfs2_orphan_add(struct ocfs2_super *osb,
 
 	ocfs2_journal_dirty(handle, fe_bh);
 
-	mlog(0, "Inode %llu orphaned in slot %d\n",
-	     (unsigned long long)OCFS2_I(inode)->ip_blkno, osb->slot_num);
+	trace_ocfs2_orphan_add_end((unsigned long long)OCFS2_I(inode)->ip_blkno,
+				   osb->slot_num);
 
 leave:
 	brelse(orphan_dir_bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2078,17 +2078,15 @@ int ocfs2_orphan_del(struct ocfs2_super *osb,
 	int status = 0;
 	struct ocfs2_dir_lookup_result lookup = { NULL, };
 
-	mlog_entry_void();
-
 	status = ocfs2_blkno_stringify(OCFS2_I(inode)->ip_blkno, name);
 	if (status < 0) {
 		mlog_errno(status);
 		goto leave;
 	}
 
-	mlog(0, "removing '%s' from orphan dir %llu (namelen=%d)\n",
-	     name, (unsigned long long)OCFS2_I(orphan_dir_inode)->ip_blkno,
-	     OCFS2_ORPHAN_NAMELEN);
+	trace_ocfs2_orphan_del(
+	     (unsigned long long)OCFS2_I(orphan_dir_inode)->ip_blkno,
+	     name, OCFS2_ORPHAN_NAMELEN);
 
 	/* find it's spot in the orphan directory */
 	status = ocfs2_find_entry(name, OCFS2_ORPHAN_NAMELEN, orphan_dir_inode,
@@ -2124,7 +2122,8 @@ int ocfs2_orphan_del(struct ocfs2_super *osb,
 leave:
 	ocfs2_free_dir_lookup_result(&lookup);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2321,9 +2320,6 @@ leave:
 		iput(orphan_dir);
 	}
 
-	if (status == -ENOSPC)
-		mlog(0, "Disk is full\n");
-
 	if ((status < 0) && inode) {
 		clear_nlink(inode);
 		iput(inode);
@@ -2358,8 +2354,10 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
 	struct buffer_head *di_bh = NULL;
 	struct ocfs2_dir_lookup_result lookup = { NULL, };
 
-	mlog_entry("(0x%p, 0x%p, %.*s')\n", dir, dentry,
-		   dentry->d_name.len, dentry->d_name.name);
+	trace_ocfs2_mv_orphaned_inode_to_new(dir, dentry,
+				dentry->d_name.len, dentry->d_name.name,
+				(unsigned long long)OCFS2_I(dir)->ip_blkno,
+				(unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 	status = ocfs2_inode_lock(dir, &parent_di_bh, 1);
 	if (status < 0) {
@@ -2476,7 +2474,8 @@ leave:
 
 	ocfs2_free_dir_lookup_result(&lookup);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 
 	return status;
 }
diff --git a/fs/ocfs2/ocfs2_trace.h b/fs/ocfs2/ocfs2_trace.h
new file mode 100644
index 00000000000..a1dae5bb54a
--- /dev/null
+++ b/fs/ocfs2/ocfs2_trace.h
@@ -0,0 +1,2739 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM ocfs2
+
+#if !defined(_TRACE_OCFS2_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_OCFS2_H
+
+#include <linux/tracepoint.h>
+
+DECLARE_EVENT_CLASS(ocfs2__int,
+	TP_PROTO(int num),
+	TP_ARGS(num),
+	TP_STRUCT__entry(
+		__field(int, num)
+	),
+	TP_fast_assign(
+		__entry->num = num;
+	),
+	TP_printk("%d", __entry->num)
+);
+
+#define DEFINE_OCFS2_INT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__int, name,	\
+	TP_PROTO(int num),	\
+	TP_ARGS(num))
+
+DECLARE_EVENT_CLASS(ocfs2__uint,
+	TP_PROTO(unsigned int num),
+	TP_ARGS(num),
+	TP_STRUCT__entry(
+		__field(	unsigned int,	num		)
+	),
+	TP_fast_assign(
+		__entry->num	= 	num;
+	),
+	TP_printk("%u", __entry->num)
+);
+
+#define DEFINE_OCFS2_UINT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__uint, name,	\
+	TP_PROTO(unsigned int num),	\
+	TP_ARGS(num))
+
+DECLARE_EVENT_CLASS(ocfs2__ull,
+	TP_PROTO(unsigned long long blkno),
+	TP_ARGS(blkno),
+	TP_STRUCT__entry(
+		__field(unsigned long long, blkno)
+	),
+	TP_fast_assign(
+		__entry->blkno = blkno;
+	),
+	TP_printk("%llu", __entry->blkno)
+);
+
+#define DEFINE_OCFS2_ULL_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull, name,	\
+	TP_PROTO(unsigned long long num),	\
+	TP_ARGS(num))
+
+DECLARE_EVENT_CLASS(ocfs2__pointer,
+	TP_PROTO(void *pointer),
+	TP_ARGS(pointer),
+	TP_STRUCT__entry(
+		__field(void *, pointer)
+	),
+	TP_fast_assign(
+		__entry->pointer = pointer;
+	),
+	TP_printk("%p", __entry->pointer)
+);
+
+#define DEFINE_OCFS2_POINTER_EVENT(name)	\
+DEFINE_EVENT(ocfs2__pointer, name,	\
+	TP_PROTO(void *pointer),	\
+	TP_ARGS(pointer))
+
+DECLARE_EVENT_CLASS(ocfs2__string,
+	TP_PROTO(const char *name),
+	TP_ARGS(name),
+	TP_STRUCT__entry(
+		__string(name,name)
+	),
+	TP_fast_assign(
+		__assign_str(name, name);
+	),
+	TP_printk("%s", __get_str(name))
+);
+
+#define DEFINE_OCFS2_STRING_EVENT(name)	\
+DEFINE_EVENT(ocfs2__string, name,	\
+	TP_PROTO(const char *name),	\
+	TP_ARGS(name))
+
+DECLARE_EVENT_CLASS(ocfs2__int_int,
+	TP_PROTO(int value1, int value2),
+	TP_ARGS(value1, value2),
+	TP_STRUCT__entry(
+		__field(int, value1)
+		__field(int, value2)
+	),
+	TP_fast_assign(
+		__entry->value1	= value1;
+		__entry->value2	= value2;
+	),
+	TP_printk("%d %d", __entry->value1, __entry->value2)
+);
+
+#define DEFINE_OCFS2_INT_INT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__int_int, name,	\
+	TP_PROTO(int val1, int val2),	\
+	TP_ARGS(val1, val2))
+
+DECLARE_EVENT_CLASS(ocfs2__uint_int,
+	TP_PROTO(unsigned int value1, int value2),
+	TP_ARGS(value1, value2),
+	TP_STRUCT__entry(
+		__field(unsigned int, value1)
+		__field(int, value2)
+	),
+	TP_fast_assign(
+		__entry->value1	= value1;
+		__entry->value2	= value2;
+	),
+	TP_printk("%u %d", __entry->value1, __entry->value2)
+);
+
+#define DEFINE_OCFS2_UINT_INT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__uint_int, name,	\
+	TP_PROTO(unsigned int val1, int val2),	\
+	TP_ARGS(val1, val2))
+
+DECLARE_EVENT_CLASS(ocfs2__uint_uint,
+	TP_PROTO(unsigned int value1, unsigned int value2),
+	TP_ARGS(value1, value2),
+	TP_STRUCT__entry(
+		__field(unsigned int, value1)
+		__field(unsigned int, value2)
+	),
+	TP_fast_assign(
+		__entry->value1 = value1;
+		__entry->value2 = value2;
+	),
+	TP_printk("%u %u", __entry->value1, __entry->value2)
+);
+
+#define DEFINE_OCFS2_UINT_UINT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__uint_uint, name,	\
+	TP_PROTO(unsigned int val1, unsigned int val2),	\
+	TP_ARGS(val1, val2))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_uint,
+	TP_PROTO(unsigned long long value1, unsigned int value2),
+	TP_ARGS(value1, value2),
+	TP_STRUCT__entry(
+		__field(unsigned long long, value1)
+		__field(unsigned int, value2)
+	),
+	TP_fast_assign(
+		__entry->value1 = value1;
+		__entry->value2 = value2;
+	),
+	TP_printk("%llu %u", __entry->value1, __entry->value2)
+);
+
+#define DEFINE_OCFS2_ULL_UINT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_uint, name,	\
+	TP_PROTO(unsigned long long val1, unsigned int val2),	\
+	TP_ARGS(val1, val2))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_int,
+	TP_PROTO(unsigned long long value1, int value2),
+	TP_ARGS(value1, value2),
+	TP_STRUCT__entry(
+		__field(unsigned long long, value1)
+		__field(int, value2)
+	),
+	TP_fast_assign(
+		__entry->value1	= value1;
+		__entry->value2	= value2;
+	),
+	TP_printk("%llu %d", __entry->value1, __entry->value2)
+);
+
+#define DEFINE_OCFS2_ULL_INT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_int, name,	\
+	TP_PROTO(unsigned long long val1, int val2),	\
+	TP_ARGS(val1, val2))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_ull,
+	TP_PROTO(unsigned long long value1, unsigned long long value2),
+	TP_ARGS(value1, value2),
+	TP_STRUCT__entry(
+		__field(unsigned long long, value1)
+		__field(unsigned long long, value2)
+	),
+	TP_fast_assign(
+		__entry->value1 = value1;
+		__entry->value2 = value2;
+	),
+	TP_printk("%llu %llu", __entry->value1, __entry->value2)
+);
+
+#define DEFINE_OCFS2_ULL_ULL_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_ull, name,	\
+	TP_PROTO(unsigned long long val1, unsigned long long val2),	\
+	TP_ARGS(val1, val2))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_ull_uint,
+	TP_PROTO(unsigned long long value1,
+		 unsigned long long value2, unsigned int value3),
+	TP_ARGS(value1, value2, value3),
+	TP_STRUCT__entry(
+		__field(unsigned long long, value1)
+		__field(unsigned long long, value2)
+		__field(unsigned int, value3)
+	),
+	TP_fast_assign(
+		__entry->value1 = value1;
+		__entry->value2 = value2;
+		__entry->value3 = value3;
+	),
+	TP_printk("%llu %llu %u",
+		  __entry->value1, __entry->value2, __entry->value3)
+);
+
+#define DEFINE_OCFS2_ULL_ULL_UINT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_ull_uint, name,	\
+	TP_PROTO(unsigned long long val1,	\
+		 unsigned long long val2, unsigned int val3),	\
+	TP_ARGS(val1, val2, val3))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_uint_uint,
+	TP_PROTO(unsigned long long value1,
+		 unsigned int value2, unsigned int value3),
+	TP_ARGS(value1, value2, value3),
+	TP_STRUCT__entry(
+		__field(unsigned long long, value1)
+		__field(unsigned int, value2)
+		__field(unsigned int, value3)
+	),
+	TP_fast_assign(
+		__entry->value1 = value1;
+		__entry->value2 = value2;
+		__entry->value3	= value3;
+	),
+	TP_printk("%llu %u %u", __entry->value1,
+		  __entry->value2, __entry->value3)
+);
+
+#define DEFINE_OCFS2_ULL_UINT_UINT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_uint_uint, name,	\
+	TP_PROTO(unsigned long long val1,	\
+		 unsigned int val2, unsigned int val3),	\
+	TP_ARGS(val1, val2, val3))
+
+DECLARE_EVENT_CLASS(ocfs2__uint_uint_uint,
+	TP_PROTO(unsigned int value1, unsigned int value2,
+		 unsigned int value3),
+	TP_ARGS(value1, value2, value3),
+	TP_STRUCT__entry(
+		__field(	unsigned int,	value1		)
+		__field(	unsigned int,	value2		)
+		__field(	unsigned int,	value3		)
+	),
+	TP_fast_assign(
+		__entry->value1	= 	value1;
+		__entry->value2	= 	value2;
+		__entry->value3	= 	value3;
+	),
+	TP_printk("%u %u %u", __entry->value1, __entry->value2, __entry->value3)
+);
+
+#define DEFINE_OCFS2_UINT_UINT_UINT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__uint_uint_uint, name,	\
+	TP_PROTO(unsigned int value1, unsigned int value2,	\
+		 unsigned int value3),	\
+	TP_ARGS(value1, value2, value3))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_ull_ull,
+	TP_PROTO(unsigned long long value1,
+		 unsigned long long value2, unsigned long long value3),
+	TP_ARGS(value1, value2, value3),
+	TP_STRUCT__entry(
+		__field(unsigned long long, value1)
+		__field(unsigned long long, value2)
+		__field(unsigned long long, value3)
+	),
+	TP_fast_assign(
+		__entry->value1 = value1;
+		__entry->value2 = value2;
+		__entry->value3 = value3;
+	),
+	TP_printk("%llu %llu %llu",
+		  __entry->value1, __entry->value2, __entry->value3)
+);
+
+#define DEFINE_OCFS2_ULL_ULL_ULL_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_ull_ull, name,	\
+	TP_PROTO(unsigned long long value1, unsigned long long value2,	\
+		 unsigned long long value3),	\
+	TP_ARGS(value1, value2, value3))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_int_int_int,
+	TP_PROTO(unsigned long long ull, int value1, int value2, int value3),
+	TP_ARGS(ull, value1, value2, value3),
+	TP_STRUCT__entry(
+		__field(	unsigned long long,	ull	)
+		__field(	int,	value1			)
+		__field(	int,	value2			)
+		__field(	int,	value3			)
+	),
+	TP_fast_assign(
+		__entry->ull		= ull;
+		__entry->value1		= value1;
+		__entry->value2		= value2;
+		__entry->value3		= value3;
+	),
+	TP_printk("%llu %d %d %d",
+		  __entry->ull, __entry->value1,
+		  __entry->value2, __entry->value3)
+);
+
+#define DEFINE_OCFS2_ULL_INT_INT_INT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_int_int_int, name,	\
+	TP_PROTO(unsigned long long ull, int value1,	\
+		 int value2, int value3),	\
+	TP_ARGS(ull, value1, value2, value3))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_uint_uint_uint,
+	TP_PROTO(unsigned long long ull, unsigned int value1,
+		 unsigned int value2, unsigned int value3),
+	TP_ARGS(ull, value1, value2, value3),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ull)
+		__field(unsigned int, value1)
+		__field(unsigned int, value2)
+		__field(unsigned int, value3)
+	),
+	TP_fast_assign(
+		__entry->ull = ull;
+		__entry->value1 = value1;
+		__entry->value2	= value2;
+		__entry->value3	= value3;
+	),
+	TP_printk("%llu %u %u %u",
+		  __entry->ull, __entry->value1,
+		  __entry->value2, __entry->value3)
+);
+
+#define DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_uint_uint_uint, name,	\
+	TP_PROTO(unsigned long long ull, unsigned int value1,	\
+		 unsigned int value2, unsigned int value3),	\
+	TP_ARGS(ull, value1, value2, value3))
+
+DECLARE_EVENT_CLASS(ocfs2__ull_ull_uint_uint,
+	TP_PROTO(unsigned long long value1, unsigned long long value2,
+		 unsigned int value3, unsigned int value4),
+	TP_ARGS(value1, value2, value3, value4),
+	TP_STRUCT__entry(
+		__field(unsigned long long, value1)
+		__field(unsigned long long, value2)
+		__field(unsigned int, value3)
+		__field(unsigned int, value4)
+	),
+	TP_fast_assign(
+		__entry->value1 = value1;
+		__entry->value2 = value2;
+		__entry->value3 = value3;
+		__entry->value4 = value4;
+	),
+	TP_printk("%llu %llu %u %u",
+		  __entry->value1, __entry->value2,
+		  __entry->value3, __entry->value4)
+);
+
+#define DEFINE_OCFS2_ULL_ULL_UINT_UINT_EVENT(name)	\
+DEFINE_EVENT(ocfs2__ull_ull_uint_uint, name,	\
+	TP_PROTO(unsigned long long ull, unsigned long long ull1,	\
+		 unsigned int value2, unsigned int value3),	\
+	TP_ARGS(ull, ull1, value2, value3))
+
+/* Trace events for fs/ocfs2/alloc.c. */
+DECLARE_EVENT_CLASS(ocfs2__btree_ops,
+	TP_PROTO(unsigned long long owner,\
+		 unsigned int value1, unsigned int value2),
+	TP_ARGS(owner, value1, value2),
+	TP_STRUCT__entry(
+		__field(unsigned long long, owner)
+		__field(unsigned int, value1)
+		__field(unsigned int, value2)
+	),
+	TP_fast_assign(
+		__entry->owner = owner;
+		__entry->value1 = value1;
+		__entry->value2	= value2;
+	),
+	TP_printk("%llu %u %u",
+		  __entry->owner, __entry->value1, __entry->value2)
+);
+
+#define DEFINE_OCFS2_BTREE_EVENT(name)	\
+DEFINE_EVENT(ocfs2__btree_ops, name,	\
+	TP_PROTO(unsigned long long owner,	\
+		 unsigned int value1, unsigned int value2),	\
+	TP_ARGS(owner, value1, value2))
+
+DEFINE_OCFS2_BTREE_EVENT(ocfs2_adjust_rightmost_branch);
+
+DEFINE_OCFS2_BTREE_EVENT(ocfs2_rotate_tree_right);
+
+DEFINE_OCFS2_BTREE_EVENT(ocfs2_append_rec_to_path);
+
+DEFINE_OCFS2_BTREE_EVENT(ocfs2_insert_extent_start);
+
+DEFINE_OCFS2_BTREE_EVENT(ocfs2_add_clusters_in_btree);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_num_free_extents);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_complete_edge_insert);
+
+TRACE_EVENT(ocfs2_grow_tree,
+	TP_PROTO(unsigned long long owner, int depth),
+	TP_ARGS(owner, depth),
+	TP_STRUCT__entry(
+		__field(unsigned long long, owner)
+		__field(int, depth)
+	),
+	TP_fast_assign(
+		__entry->owner = owner;
+		__entry->depth = depth;
+	),
+	TP_printk("%llu %d", __entry->owner, __entry->depth)
+);
+
+TRACE_EVENT(ocfs2_rotate_subtree,
+	TP_PROTO(int subtree_root, unsigned long long blkno,
+		 int depth),
+	TP_ARGS(subtree_root, blkno, depth),
+	TP_STRUCT__entry(
+		__field(int, subtree_root)
+		__field(unsigned long long, blkno)
+		__field(int, depth)
+	),
+	TP_fast_assign(
+		__entry->subtree_root = subtree_root;
+		__entry->blkno = blkno;
+		__entry->depth = depth;
+	),
+	TP_printk("%d %llu %d", __entry->subtree_root,
+		  __entry->blkno, __entry->depth)
+);
+
+TRACE_EVENT(ocfs2_insert_extent,
+	TP_PROTO(unsigned int ins_appending, unsigned int ins_contig,
+		 int ins_contig_index, int free_records, int ins_tree_depth),
+	TP_ARGS(ins_appending, ins_contig, ins_contig_index, free_records,
+		ins_tree_depth),
+	TP_STRUCT__entry(
+		__field(unsigned int, ins_appending)
+		__field(unsigned int, ins_contig)
+		__field(int, ins_contig_index)
+		__field(int, free_records)
+		__field(int, ins_tree_depth)
+	),
+	TP_fast_assign(
+		__entry->ins_appending = ins_appending;
+		__entry->ins_contig = ins_contig;
+		__entry->ins_contig_index = ins_contig_index;
+		__entry->free_records = free_records;
+		__entry->ins_tree_depth = ins_tree_depth;
+	),
+	TP_printk("%u %u %d %d %d",
+		  __entry->ins_appending, __entry->ins_contig,
+		  __entry->ins_contig_index, __entry->free_records,
+		  __entry->ins_tree_depth)
+);
+
+TRACE_EVENT(ocfs2_split_extent,
+	TP_PROTO(int split_index, unsigned int c_contig_type,
+		 unsigned int c_has_empty_extent,
+		 unsigned int c_split_covers_rec),
+	TP_ARGS(split_index, c_contig_type,
+		c_has_empty_extent, c_split_covers_rec),
+	TP_STRUCT__entry(
+		__field(int, split_index)
+		__field(unsigned int, c_contig_type)
+		__field(unsigned int, c_has_empty_extent)
+		__field(unsigned int, c_split_covers_rec)
+	),
+	TP_fast_assign(
+		__entry->split_index = split_index;
+		__entry->c_contig_type = c_contig_type;
+		__entry->c_has_empty_extent = c_has_empty_extent;
+		__entry->c_split_covers_rec = c_split_covers_rec;
+	),
+	TP_printk("%d %u %u %u", __entry->split_index, __entry->c_contig_type,
+		  __entry->c_has_empty_extent, __entry->c_split_covers_rec)
+);
+
+TRACE_EVENT(ocfs2_remove_extent,
+	TP_PROTO(unsigned long long owner, unsigned int cpos,
+		 unsigned int len, int index,
+		 unsigned int e_cpos, unsigned int clusters),
+	TP_ARGS(owner, cpos, len, index, e_cpos, clusters),
+	TP_STRUCT__entry(
+		__field(unsigned long long, owner)
+		__field(unsigned int, cpos)
+		__field(unsigned int, len)
+		__field(int, index)
+		__field(unsigned int, e_cpos)
+		__field(unsigned int, clusters)
+	),
+	TP_fast_assign(
+		__entry->owner = owner;
+		__entry->cpos = cpos;
+		__entry->len = len;
+		__entry->index = index;
+		__entry->e_cpos = e_cpos;
+		__entry->clusters = clusters;
+	),
+	TP_printk("%llu %u %u %d %u %u",
+		  __entry->owner, __entry->cpos, __entry->len, __entry->index,
+		  __entry->e_cpos, __entry->clusters)
+);
+
+TRACE_EVENT(ocfs2_commit_truncate,
+	TP_PROTO(unsigned long long ino, unsigned int new_cpos,
+		 unsigned int clusters, unsigned int depth),
+	TP_ARGS(ino, new_cpos, clusters, depth),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned int, new_cpos)
+		__field(unsigned int, clusters)
+		__field(unsigned int, depth)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->new_cpos = new_cpos;
+		__entry->clusters = clusters;
+		__entry->depth = depth;
+	),
+	TP_printk("%llu %u %u %u",
+		  __entry->ino, __entry->new_cpos,
+		  __entry->clusters, __entry->depth)
+);
+
+TRACE_EVENT(ocfs2_validate_extent_block,
+	TP_PROTO(unsigned long long blkno),
+	TP_ARGS(blkno),
+	TP_STRUCT__entry(
+		__field(unsigned long long, blkno)
+	),
+	TP_fast_assign(
+		__entry->blkno = blkno;
+	),
+	TP_printk("%llu ", __entry->blkno)
+);
+
+TRACE_EVENT(ocfs2_rotate_leaf,
+	TP_PROTO(unsigned int insert_cpos, int insert_index,
+		 int has_empty, int next_free,
+		 unsigned int l_count),
+	TP_ARGS(insert_cpos, insert_index, has_empty,
+		next_free, l_count),
+	TP_STRUCT__entry(
+		__field(unsigned int, insert_cpos)
+		__field(int, insert_index)
+		__field(int, has_empty)
+		__field(int, next_free)
+		__field(unsigned int, l_count)
+	),
+	TP_fast_assign(
+		__entry->insert_cpos = insert_cpos;
+		__entry->insert_index = insert_index;
+		__entry->has_empty = has_empty;
+		__entry->next_free = next_free;
+		__entry->l_count = l_count;
+	),
+	TP_printk("%u %d %d %d %u", __entry->insert_cpos,
+		  __entry->insert_index, __entry->has_empty,
+		  __entry->next_free, __entry->l_count)
+);
+
+TRACE_EVENT(ocfs2_add_clusters_in_btree_ret,
+	TP_PROTO(int status, int reason, int err),
+	TP_ARGS(status, reason, err),
+	TP_STRUCT__entry(
+		__field(int, status)
+		__field(int, reason)
+		__field(int, err)
+	),
+	TP_fast_assign(
+		__entry->status = status;
+		__entry->reason = reason;
+		__entry->err = err;
+	),
+	TP_printk("%d %d %d", __entry->status,
+		  __entry->reason, __entry->err)
+);
+
+TRACE_EVENT(ocfs2_mark_extent_written,
+	TP_PROTO(unsigned long long owner, unsigned int cpos,
+		 unsigned int len, unsigned int phys),
+	TP_ARGS(owner, cpos, len, phys),
+	TP_STRUCT__entry(
+		__field(unsigned long long, owner)
+		__field(unsigned int, cpos)
+		__field(unsigned int, len)
+		__field(unsigned int, phys)
+	),
+	TP_fast_assign(
+		__entry->owner = owner;
+		__entry->cpos = cpos;
+		__entry->len = len;
+		__entry->phys = phys;
+	),
+	TP_printk("%llu %u %u %u",
+		  __entry->owner, __entry->cpos,
+		  __entry->len, __entry->phys)
+);
+
+DECLARE_EVENT_CLASS(ocfs2__truncate_log_ops,
+	TP_PROTO(unsigned long long blkno, int index,
+		 unsigned int start, unsigned int num),
+	TP_ARGS(blkno, index, start, num),
+	TP_STRUCT__entry(
+		__field(unsigned long long, blkno)
+		__field(int, index)
+		__field(unsigned int, start)
+		__field(unsigned int, num)
+	),
+	TP_fast_assign(
+		__entry->blkno = blkno;
+		__entry->index = index;
+		__entry->start = start;
+		__entry->num = num;
+	),
+	TP_printk("%llu %d %u %u",
+		  __entry->blkno, __entry->index,
+		  __entry->start, __entry->num)
+);
+
+#define DEFINE_OCFS2_TRUNCATE_LOG_OPS_EVENT(name)	\
+DEFINE_EVENT(ocfs2__truncate_log_ops, name,	\
+	TP_PROTO(unsigned long long blkno, int index,	\
+		 unsigned int start, unsigned int num),	\
+	TP_ARGS(blkno, index, start, num))
+
+DEFINE_OCFS2_TRUNCATE_LOG_OPS_EVENT(ocfs2_truncate_log_append);
+
+DEFINE_OCFS2_TRUNCATE_LOG_OPS_EVENT(ocfs2_replay_truncate_records);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_flush_truncate_log);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_begin_truncate_log_recovery);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_truncate_log_recovery_num);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_complete_truncate_log_recovery);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_free_cached_blocks);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_cache_cluster_dealloc);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_run_deallocs);
+
+TRACE_EVENT(ocfs2_cache_block_dealloc,
+	TP_PROTO(int type, int slot, unsigned long long suballoc,
+		 unsigned long long blkno, unsigned int bit),
+	TP_ARGS(type, slot, suballoc, blkno, bit),
+	TP_STRUCT__entry(
+		__field(int, type)
+		__field(int, slot)
+		__field(unsigned long long, suballoc)
+		__field(unsigned long long, blkno)
+		__field(unsigned int, bit)
+	),
+	TP_fast_assign(
+		__entry->type = type;
+		__entry->slot = slot;
+		__entry->suballoc = suballoc;
+		__entry->blkno = blkno;
+		__entry->bit = bit;
+	),
+	TP_printk("%d %d %llu %llu %u",
+		  __entry->type, __entry->slot, __entry->suballoc,
+		  __entry->blkno, __entry->bit)
+);
+
+/* End of trace events for fs/ocfs2/alloc.c. */
+
+/* Trace events for fs/ocfs2/localalloc.c. */
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_la_set_sizes);
+
+DEFINE_OCFS2_ULL_INT_INT_INT_EVENT(ocfs2_alloc_should_use_local);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_load_local_alloc);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_begin_local_alloc_recovery);
+
+DEFINE_OCFS2_ULL_INT_INT_INT_EVENT(ocfs2_reserve_local_alloc_bits);
+
+DEFINE_OCFS2_UINT_EVENT(ocfs2_local_alloc_count_bits);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_local_alloc_find_clear_bits_search_bitmap);
+
+DEFINE_OCFS2_ULL_INT_INT_INT_EVENT(ocfs2_local_alloc_find_clear_bits);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_sync_local_to_main);
+
+TRACE_EVENT(ocfs2_sync_local_to_main_free,
+	TP_PROTO(int count, int bit, unsigned long long start_blk,
+		 unsigned long long blkno),
+	TP_ARGS(count, bit, start_blk, blkno),
+	TP_STRUCT__entry(
+		__field(int, count)
+		__field(int, bit)
+		__field(unsigned long long, start_blk)
+		__field(unsigned long long, blkno)
+	),
+	TP_fast_assign(
+		__entry->count = count;
+		__entry->bit = bit;
+		__entry->start_blk = start_blk;
+		__entry->blkno = blkno;
+	),
+	TP_printk("%d %d %llu %llu",
+		  __entry->count, __entry->bit, __entry->start_blk,
+		  __entry->blkno)
+);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_local_alloc_new_window);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_local_alloc_new_window_result);
+
+/* End of trace events for fs/ocfs2/localalloc.c. */
+
+/* Trace events for fs/ocfs2/resize.c. */
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_update_last_group_and_inode);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_group_extend);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_group_add);
+
+/* End of trace events for fs/ocfs2/resize.c. */
+
+/* Trace events for fs/ocfs2/suballoc.c. */
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_validate_group_descriptor);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_block_group_alloc_contig);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_block_group_alloc_discontig);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_block_group_alloc);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_reserve_suballoc_bits_nospc);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_reserve_suballoc_bits_no_new_group);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_reserve_new_inode_new_group);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_block_group_set_bits);
+
+TRACE_EVENT(ocfs2_relink_block_group,
+	TP_PROTO(unsigned long long i_blkno, unsigned int chain,
+		 unsigned long long bg_blkno,
+		 unsigned long long prev_blkno),
+	TP_ARGS(i_blkno, chain, bg_blkno, prev_blkno),
+	TP_STRUCT__entry(
+		__field(unsigned long long, i_blkno)
+		__field(unsigned int, chain)
+		__field(unsigned long long, bg_blkno)
+		__field(unsigned long long, prev_blkno)
+	),
+	TP_fast_assign(
+		__entry->i_blkno = i_blkno;
+		__entry->chain = chain;
+		__entry->bg_blkno = bg_blkno;
+		__entry->prev_blkno = prev_blkno;
+	),
+	TP_printk("%llu %u %llu %llu",
+		  __entry->i_blkno, __entry->chain, __entry->bg_blkno,
+		  __entry->prev_blkno)
+);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_cluster_group_search_wrong_max_bits);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_cluster_group_search_max_block);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_block_group_search_max_block);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_search_chain_begin);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_search_chain_succ);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_search_chain_end);
+
+DEFINE_OCFS2_UINT_EVENT(ocfs2_claim_suballoc_bits);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_claim_new_inode_at_loc);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_block_group_clear_bits);
+
+TRACE_EVENT(ocfs2_free_suballoc_bits,
+	TP_PROTO(unsigned long long inode, unsigned long long group,
+		 unsigned int start_bit, unsigned int count),
+	TP_ARGS(inode, group, start_bit, count),
+	TP_STRUCT__entry(
+		__field(unsigned long long, inode)
+		__field(unsigned long long, group)
+		__field(unsigned int, start_bit)
+		__field(unsigned int, count)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->group = group;
+		__entry->start_bit = start_bit;
+		__entry->count = count;
+	),
+	TP_printk("%llu %llu %u %u", __entry->inode, __entry->group,
+		  __entry->start_bit, __entry->count)
+);
+
+TRACE_EVENT(ocfs2_free_clusters,
+	TP_PROTO(unsigned long long bg_blkno, unsigned long long start_blk,
+		 unsigned int start_bit, unsigned int count),
+	TP_ARGS(bg_blkno, start_blk, start_bit, count),
+	TP_STRUCT__entry(
+		__field(unsigned long long, bg_blkno)
+		__field(unsigned long long, start_blk)
+		__field(unsigned int, start_bit)
+		__field(unsigned int, count)
+	),
+	TP_fast_assign(
+		__entry->bg_blkno = bg_blkno;
+		__entry->start_blk = start_blk;
+		__entry->start_bit = start_bit;
+		__entry->count = count;
+	),
+	TP_printk("%llu %llu %u %u", __entry->bg_blkno, __entry->start_blk,
+		  __entry->start_bit, __entry->count)
+);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_get_suballoc_slot_bit);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_test_suballoc_bit);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_test_inode_bit);
+
+/* End of trace events for fs/ocfs2/suballoc.c. */
+
+/* Trace events for fs/ocfs2/refcounttree.c. */
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_validate_refcount_block);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_purge_refcount_trees);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_create_refcount_tree);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_create_refcount_tree_blkno);
+
+DEFINE_OCFS2_ULL_INT_INT_INT_EVENT(ocfs2_change_refcount_rec);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_expand_inline_ref_root);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_divide_leaf_refcount_block);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_new_leaf_refcount_block);
+
+DECLARE_EVENT_CLASS(ocfs2__refcount_tree_ops,
+	TP_PROTO(unsigned long long blkno, int index,
+		 unsigned long long cpos,
+		 unsigned int clusters, unsigned int refcount),
+	TP_ARGS(blkno, index, cpos, clusters, refcount),
+	TP_STRUCT__entry(
+		__field(unsigned long long, blkno)
+		__field(int, index)
+		__field(unsigned long long, cpos)
+		__field(unsigned int, clusters)
+		__field(unsigned int, refcount)
+	),
+	TP_fast_assign(
+		__entry->blkno = blkno;
+		__entry->index = index;
+		__entry->cpos = cpos;
+		__entry->clusters = clusters;
+		__entry->refcount = refcount;
+	),
+	TP_printk("%llu %d %llu %u %u", __entry->blkno, __entry->index,
+		  __entry->cpos, __entry->clusters, __entry->refcount)
+);
+
+#define DEFINE_OCFS2_REFCOUNT_TREE_OPS_EVENT(name)	\
+DEFINE_EVENT(ocfs2__refcount_tree_ops, name,		\
+	TP_PROTO(unsigned long long blkno, int index,	\
+		 unsigned long long cpos,		\
+		 unsigned int count, unsigned int refcount),	\
+	TP_ARGS(blkno, index, cpos, count, refcount))
+
+DEFINE_OCFS2_REFCOUNT_TREE_OPS_EVENT(ocfs2_insert_refcount_rec);
+
+TRACE_EVENT(ocfs2_split_refcount_rec,
+	TP_PROTO(unsigned long long cpos,
+		 unsigned int clusters, unsigned int refcount,
+		 unsigned long long split_cpos,
+		 unsigned int split_clusters, unsigned int split_refcount),
+	TP_ARGS(cpos, clusters, refcount,
+		split_cpos, split_clusters, split_refcount),
+	TP_STRUCT__entry(
+		__field(unsigned long long, cpos)
+		__field(unsigned int, clusters)
+		__field(unsigned int, refcount)
+		__field(unsigned long long, split_cpos)
+		__field(unsigned int, split_clusters)
+		__field(unsigned int, split_refcount)
+	),
+	TP_fast_assign(
+		__entry->cpos = cpos;
+		__entry->clusters = clusters;
+		__entry->refcount = refcount;
+		__entry->split_cpos = split_cpos;
+		__entry->split_clusters = split_clusters;
+		__entry->split_refcount	= split_refcount;
+	),
+	TP_printk("%llu %u %u %llu %u %u",
+		  __entry->cpos, __entry->clusters, __entry->refcount,
+		  __entry->split_cpos, __entry->split_clusters,
+		  __entry->split_refcount)
+);
+
+DEFINE_OCFS2_REFCOUNT_TREE_OPS_EVENT(ocfs2_split_refcount_rec_insert);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_increase_refcount_begin);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_increase_refcount_change);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_increase_refcount_insert);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_increase_refcount_split);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_remove_refcount_extent);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_restore_refcount_block);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_decrease_refcount_rec);
+
+TRACE_EVENT(ocfs2_decrease_refcount,
+	TP_PROTO(unsigned long long owner,
+		 unsigned long long cpos,
+		 unsigned int len, int delete),
+	TP_ARGS(owner, cpos, len, delete),
+	TP_STRUCT__entry(
+		__field(unsigned long long, owner)
+		__field(unsigned long long, cpos)
+		__field(unsigned int, len)
+		__field(int, delete)
+	),
+	TP_fast_assign(
+		__entry->owner = owner;
+		__entry->cpos = cpos;
+		__entry->len = len;
+		__entry->delete = delete;
+	),
+	TP_printk("%llu %llu %u %d",
+		  __entry->owner, __entry->cpos, __entry->len, __entry->delete)
+);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_mark_extent_refcounted);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_calc_refcount_meta_credits);
+
+TRACE_EVENT(ocfs2_calc_refcount_meta_credits_iterate,
+	TP_PROTO(int recs_add, unsigned long long cpos,
+		 unsigned int clusters, unsigned long long r_cpos,
+		 unsigned int r_clusters, unsigned int refcount, int index),
+	TP_ARGS(recs_add, cpos, clusters, r_cpos, r_clusters, refcount, index),
+	TP_STRUCT__entry(
+		__field(int, recs_add)
+		__field(unsigned long long, cpos)
+		__field(unsigned int, clusters)
+		__field(unsigned long long, r_cpos)
+		__field(unsigned int, r_clusters)
+		__field(unsigned int, refcount)
+		__field(int, index)
+	),
+	TP_fast_assign(
+		__entry->recs_add = recs_add;
+		__entry->cpos = cpos;
+		__entry->clusters = clusters;
+		__entry->r_cpos = r_cpos;
+		__entry->r_clusters = r_clusters;
+		__entry->refcount = refcount;
+		__entry->index = index;
+	),
+	TP_printk("%d %llu %u %llu %u %u %d",
+		  __entry->recs_add, __entry->cpos, __entry->clusters,
+		  __entry->r_cpos, __entry->r_clusters,
+		  __entry->refcount, __entry->index)
+);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_add_refcount_flag);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_prepare_refcount_change_for_del);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_lock_refcount_allocators);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_duplicate_clusters_by_page);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_duplicate_clusters_by_jbd);
+
+TRACE_EVENT(ocfs2_clear_ext_refcount,
+	TP_PROTO(unsigned long long ino, unsigned int cpos,
+		 unsigned int len, unsigned int p_cluster,
+		 unsigned int ext_flags),
+	TP_ARGS(ino, cpos, len, p_cluster, ext_flags),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned int, cpos)
+		__field(unsigned int, len)
+		__field(unsigned int, p_cluster)
+		__field(unsigned int, ext_flags)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->cpos = cpos;
+		__entry->len = len;
+		__entry->p_cluster = p_cluster;
+		__entry->ext_flags = ext_flags;
+	),
+	TP_printk("%llu %u %u %u %u",
+		  __entry->ino, __entry->cpos, __entry->len,
+		  __entry->p_cluster, __entry->ext_flags)
+);
+
+TRACE_EVENT(ocfs2_replace_clusters,
+	TP_PROTO(unsigned long long ino, unsigned int cpos,
+		 unsigned int old, unsigned int new, unsigned int len,
+		 unsigned int ext_flags),
+	TP_ARGS(ino, cpos, old, new, len, ext_flags),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned int, cpos)
+		__field(unsigned int, old)
+		__field(unsigned int, new)
+		__field(unsigned int, len)
+		__field(unsigned int, ext_flags)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->cpos = cpos;
+		__entry->old = old;
+		__entry->new = new;
+		__entry->len = len;
+		__entry->ext_flags = ext_flags;
+	),
+	TP_printk("%llu %u %u %u %u %u",
+		  __entry->ino, __entry->cpos, __entry->old, __entry->new,
+		  __entry->len, __entry->ext_flags)
+);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_make_clusters_writable);
+
+TRACE_EVENT(ocfs2_refcount_cow_hunk,
+	TP_PROTO(unsigned long long ino, unsigned int cpos,
+		 unsigned int write_len, unsigned int max_cpos,
+		 unsigned int cow_start, unsigned int cow_len),
+	TP_ARGS(ino, cpos, write_len, max_cpos, cow_start, cow_len),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned int, cpos)
+		__field(unsigned int, write_len)
+		__field(unsigned int, max_cpos)
+		__field(unsigned int, cow_start)
+		__field(unsigned int, cow_len)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->cpos = cpos;
+		__entry->write_len = write_len;
+		__entry->max_cpos = max_cpos;
+		__entry->cow_start = cow_start;
+		__entry->cow_len = cow_len;
+	),
+	TP_printk("%llu %u %u %u %u %u",
+		  __entry->ino, __entry->cpos, __entry->write_len,
+		  __entry->max_cpos, __entry->cow_start, __entry->cow_len)
+);
+
+/* End of trace events for fs/ocfs2/refcounttree.c. */
+
+/* Trace events for fs/ocfs2/aops.c. */
+
+DECLARE_EVENT_CLASS(ocfs2__get_block,
+	TP_PROTO(unsigned long long ino, unsigned long long iblock,
+		 void *bh_result, int create),
+	TP_ARGS(ino, iblock, bh_result, create),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned long long, iblock)
+		__field(void *, bh_result)
+		__field(int, create)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->iblock = iblock;
+		__entry->bh_result = bh_result;
+		__entry->create = create;
+	),
+	TP_printk("%llu %llu %p %d",
+		  __entry->ino, __entry->iblock,
+		  __entry->bh_result, __entry->create)
+);
+
+#define DEFINE_OCFS2_GET_BLOCK_EVENT(name)	\
+DEFINE_EVENT(ocfs2__get_block, name,	\
+	TP_PROTO(unsigned long long ino, unsigned long long iblock,	\
+		 void *bh_result, int create),	\
+	TP_ARGS(ino, iblock, bh_result, create))
+
+DEFINE_OCFS2_GET_BLOCK_EVENT(ocfs2_symlink_get_block);
+
+DEFINE_OCFS2_GET_BLOCK_EVENT(ocfs2_get_block);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_get_block_end);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_readpage);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_writepage);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_bmap);
+
+TRACE_EVENT(ocfs2_try_to_write_inline_data,
+	TP_PROTO(unsigned long long ino, unsigned int len,
+		 unsigned long long pos, unsigned int flags),
+	TP_ARGS(ino, len, pos, flags),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned int, len)
+		__field(unsigned long long, pos)
+		__field(unsigned int, flags)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->len = len;
+		__entry->pos = pos;
+		__entry->flags = flags;
+	),
+	TP_printk("%llu %u %llu 0x%x",
+		  __entry->ino, __entry->len, __entry->pos, __entry->flags)
+);
+
+TRACE_EVENT(ocfs2_write_begin_nolock,
+	TP_PROTO(unsigned long long ino,
+		 long long i_size, unsigned int i_clusters,
+		 unsigned long long pos, unsigned int len,
+		 unsigned int flags, void *page,
+		 unsigned int clusters, unsigned int extents_to_split),
+	TP_ARGS(ino, i_size, i_clusters, pos, len, flags,
+		page, clusters, extents_to_split),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(long long, i_size)
+		__field(unsigned int, i_clusters)
+		__field(unsigned long long, pos)
+		__field(unsigned int, len)
+		__field(unsigned int, flags)
+		__field(void *, page)
+		__field(unsigned int, clusters)
+		__field(unsigned int, extents_to_split)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->i_size = i_size;
+		__entry->i_clusters = i_clusters;
+		__entry->pos = pos;
+		__entry->len = len;
+		__entry->flags = flags;
+		__entry->page = page;
+		__entry->clusters = clusters;
+		__entry->extents_to_split = extents_to_split;
+	),
+	TP_printk("%llu %lld %u %llu %u %u %p %u %u",
+		  __entry->ino, __entry->i_size, __entry->i_clusters,
+		  __entry->pos, __entry->len,
+		  __entry->flags, __entry->page, __entry->clusters,
+		  __entry->extents_to_split)
+);
+
+TRACE_EVENT(ocfs2_write_end_inline,
+	TP_PROTO(unsigned long long ino,
+		 unsigned long long pos, unsigned int copied,
+		 unsigned int id_count, unsigned int features),
+	TP_ARGS(ino, pos, copied, id_count, features),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned long long, pos)
+		__field(unsigned int, copied)
+		__field(unsigned int, id_count)
+		__field(unsigned int, features)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->pos = pos;
+		__entry->copied = copied;
+		__entry->id_count = id_count;
+		__entry->features = features;
+	),
+	TP_printk("%llu %llu %u %u %u",
+		  __entry->ino, __entry->pos, __entry->copied,
+		  __entry->id_count, __entry->features)
+);
+
+/* End of trace events for fs/ocfs2/aops.c. */
+
+/* Trace events for fs/ocfs2/mmap.c. */
+
+TRACE_EVENT(ocfs2_fault,
+	TP_PROTO(unsigned long long ino,
+		 void *area, void *page, unsigned long pgoff),
+	TP_ARGS(ino, area, page, pgoff),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(void *, area)
+		__field(void *, page)
+		__field(unsigned long, pgoff)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->area = area;
+		__entry->page = page;
+		__entry->pgoff = pgoff;
+	),
+	TP_printk("%llu %p %p %lu",
+		  __entry->ino, __entry->area, __entry->page, __entry->pgoff)
+);
+
+/* End of trace events for fs/ocfs2/mmap.c. */
+
+/* Trace events for fs/ocfs2/file.c. */
+
+DECLARE_EVENT_CLASS(ocfs2__file_ops,
+	TP_PROTO(void *inode, void *file, void *dentry,
+		 unsigned long long ino,
+		 unsigned int d_len, const unsigned char *d_name,
+		 unsigned long long para),
+	TP_ARGS(inode, file, dentry, ino, d_len, d_name, para),
+	TP_STRUCT__entry(
+		__field(void *, inode)
+		__field(void *, file)
+		__field(void *, dentry)
+		__field(unsigned long long, ino)
+		__field(unsigned int, d_len)
+		__string(d_name, d_name)
+		__field(unsigned long long, para)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->file = file;
+		__entry->dentry = dentry;
+		__entry->ino = ino;
+		__entry->d_len = d_len;
+		__assign_str(d_name, d_name);
+		__entry->para = para;
+	),
+	TP_printk("%p %p %p %llu %llu %.*s", __entry->inode, __entry->file,
+		  __entry->dentry, __entry->ino, __entry->para,
+		  __entry->d_len, __get_str(d_name))
+);
+
+#define DEFINE_OCFS2_FILE_OPS(name)				\
+DEFINE_EVENT(ocfs2__file_ops, name,				\
+TP_PROTO(void *inode, void *file, void *dentry,			\
+	 unsigned long long ino,				\
+	 unsigned int d_len, const unsigned char *d_name,	\
+	 unsigned long long mode),				\
+	TP_ARGS(inode, file, dentry, ino, d_len, d_name, mode))
+
+DEFINE_OCFS2_FILE_OPS(ocfs2_file_open);
+
+DEFINE_OCFS2_FILE_OPS(ocfs2_file_release);
+
+DEFINE_OCFS2_FILE_OPS(ocfs2_sync_file);
+
+DEFINE_OCFS2_FILE_OPS(ocfs2_file_aio_write);
+
+DEFINE_OCFS2_FILE_OPS(ocfs2_file_splice_write);
+
+DEFINE_OCFS2_FILE_OPS(ocfs2_file_splice_read);
+
+DEFINE_OCFS2_FILE_OPS(ocfs2_file_aio_read);
+
+DEFINE_OCFS2_ULL_ULL_ULL_EVENT(ocfs2_truncate_file);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_truncate_file_error);
+
+TRACE_EVENT(ocfs2_extend_allocation,
+	TP_PROTO(unsigned long long ip_blkno, unsigned long long size,
+		 unsigned int clusters, unsigned int clusters_to_add,
+		 int why, int restart_func),
+	TP_ARGS(ip_blkno, size, clusters, clusters_to_add, why, restart_func),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ip_blkno)
+		__field(unsigned long long, size)
+		__field(unsigned int, clusters)
+		__field(unsigned int, clusters_to_add)
+		__field(int, why)
+		__field(int, restart_func)
+	),
+	TP_fast_assign(
+		__entry->ip_blkno = ip_blkno;
+		__entry->size = size;
+		__entry->clusters = clusters;
+		__entry->clusters_to_add = clusters_to_add;
+		__entry->why = why;
+		__entry->restart_func = restart_func;
+	),
+	TP_printk("%llu %llu %u %u %d %d",
+		  __entry->ip_blkno, __entry->size, __entry->clusters,
+		  __entry->clusters_to_add, __entry->why, __entry->restart_func)
+);
+
+TRACE_EVENT(ocfs2_extend_allocation_end,
+	TP_PROTO(unsigned long long ino,
+		 unsigned int di_clusters, unsigned long long di_size,
+		 unsigned int ip_clusters, unsigned long long i_size),
+	TP_ARGS(ino, di_clusters, di_size, ip_clusters, i_size),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned int, di_clusters)
+		__field(unsigned long long, di_size)
+		__field(unsigned int, ip_clusters)
+		__field(unsigned long long, i_size)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->di_clusters = di_clusters;
+		__entry->di_size = di_size;
+		__entry->ip_clusters = ip_clusters;
+		__entry->i_size = i_size;
+	),
+	TP_printk("%llu %u %llu %u %llu", __entry->ino, __entry->di_clusters,
+		  __entry->di_size, __entry->ip_clusters, __entry->i_size)
+);
+
+TRACE_EVENT(ocfs2_write_zero_page,
+	TP_PROTO(unsigned long long ino,
+		 unsigned long long abs_from, unsigned long long abs_to,
+		 unsigned long index, unsigned int zero_from,
+		 unsigned int zero_to),
+	TP_ARGS(ino, abs_from, abs_to, index, zero_from, zero_to),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned long long, abs_from)
+		__field(unsigned long long, abs_to)
+		__field(unsigned long, index)
+		__field(unsigned int, zero_from)
+		__field(unsigned int, zero_to)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->abs_from = abs_from;
+		__entry->abs_to = abs_to;
+		__entry->index = index;
+		__entry->zero_from = zero_from;
+		__entry->zero_to = zero_to;
+	),
+	TP_printk("%llu %llu %llu %lu %u %u", __entry->ino,
+		  __entry->abs_from, __entry->abs_to,
+		  __entry->index, __entry->zero_from, __entry->zero_to)
+);
+
+DEFINE_OCFS2_ULL_ULL_ULL_EVENT(ocfs2_zero_extend_range);
+
+DEFINE_OCFS2_ULL_ULL_ULL_EVENT(ocfs2_zero_extend);
+
+TRACE_EVENT(ocfs2_setattr,
+	TP_PROTO(void *inode, void *dentry,
+		 unsigned long long ino,
+		 unsigned int d_len, const unsigned char *d_name,
+		 unsigned int ia_valid, unsigned int ia_mode,
+		 unsigned int ia_uid, unsigned int ia_gid),
+	TP_ARGS(inode, dentry, ino, d_len, d_name,
+		ia_valid, ia_mode, ia_uid, ia_gid),
+	TP_STRUCT__entry(
+		__field(void *, inode)
+		__field(void *, dentry)
+		__field(unsigned long long, ino)
+		__field(unsigned int, d_len)
+		__string(d_name, d_name)
+		__field(unsigned int, ia_valid)
+		__field(unsigned int, ia_mode)
+		__field(unsigned int, ia_uid)
+		__field(unsigned int, ia_gid)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->dentry = dentry;
+		__entry->ino = ino;
+		__entry->d_len = d_len;
+		__assign_str(d_name, d_name);
+		__entry->ia_valid = ia_valid;
+		__entry->ia_mode = ia_mode;
+		__entry->ia_uid = ia_uid;
+		__entry->ia_gid = ia_gid;
+	),
+	TP_printk("%p %p %llu %.*s %u %u %u %u", __entry->inode,
+		  __entry->dentry, __entry->ino, __entry->d_len,
+		  __get_str(d_name), __entry->ia_valid, __entry->ia_mode,
+		  __entry->ia_uid, __entry->ia_gid)
+);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_write_remove_suid);
+
+DEFINE_OCFS2_ULL_ULL_ULL_EVENT(ocfs2_zero_partial_clusters);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_zero_partial_clusters_range1);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_zero_partial_clusters_range2);
+
+DEFINE_OCFS2_ULL_ULL_ULL_EVENT(ocfs2_remove_inode_range);
+
+TRACE_EVENT(ocfs2_prepare_inode_for_write,
+	TP_PROTO(unsigned long long ino, unsigned long long saved_pos,
+		 int appending, unsigned long count,
+		 int *direct_io, int *has_refcount),
+	TP_ARGS(ino, saved_pos, appending, count, direct_io, has_refcount),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned long long, saved_pos)
+		__field(int, appending)
+		__field(unsigned long, count)
+		__field(int, direct_io)
+		__field(int, has_refcount)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->saved_pos = saved_pos;
+		__entry->appending = appending;
+		__entry->count = count;
+		__entry->direct_io = direct_io ? *direct_io : -1;
+		__entry->has_refcount = has_refcount ? *has_refcount : -1;
+	),
+	TP_printk("%llu %llu %d %lu %d %d", __entry->ino,
+		  __entry->saved_pos, __entry->appending, __entry->count,
+		  __entry->direct_io, __entry->has_refcount)
+);
+
+DEFINE_OCFS2_INT_EVENT(generic_file_aio_read_ret);
+
+/* End of trace events for fs/ocfs2/file.c. */
+
+/* Trace events for fs/ocfs2/inode.c. */
+
+TRACE_EVENT(ocfs2_iget_begin,
+	TP_PROTO(unsigned long long ino, unsigned int flags, int sysfile_type),
+	TP_ARGS(ino, flags, sysfile_type),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(unsigned int, flags)
+		__field(int, sysfile_type)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->flags = flags;
+		__entry->sysfile_type = sysfile_type;
+	),
+	TP_printk("%llu %u %d", __entry->ino,
+		  __entry->flags, __entry->sysfile_type)
+);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_iget5_locked);
+
+TRACE_EVENT(ocfs2_iget_end,
+	TP_PROTO(void *inode, unsigned long long ino),
+	TP_ARGS(inode, ino),
+	TP_STRUCT__entry(
+		__field(void *, inode)
+		__field(unsigned long long, ino)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->ino = ino;
+	),
+	TP_printk("%p %llu", __entry->inode, __entry->ino)
+);
+
+TRACE_EVENT(ocfs2_find_actor,
+	TP_PROTO(void *inode, unsigned long long ino,
+		 void *args,  unsigned long long fi_blkno),
+	TP_ARGS(inode, ino, args, fi_blkno),
+	TP_STRUCT__entry(
+		__field(void *, inode)
+		__field(unsigned long long, ino)
+		__field(void *, args)
+		__field(unsigned long long, fi_blkno)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->ino = ino;
+		__entry->args = args;
+		__entry->fi_blkno = fi_blkno;
+	),
+	TP_printk("%p %llu %p %llu", __entry->inode, __entry->ino,
+		  __entry->args, __entry->fi_blkno)
+);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_populate_inode);
+
+DEFINE_OCFS2_ULL_INT_EVENT(ocfs2_read_locked_inode);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_check_orphan_recovery_state);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_validate_inode_block);
+
+TRACE_EVENT(ocfs2_inode_is_valid_to_delete,
+	TP_PROTO(void *task, void *dc_task, unsigned long long ino,
+		 unsigned int flags),
+	TP_ARGS(task, dc_task, ino, flags),
+	TP_STRUCT__entry(
+		__field(void *, task)
+		__field(void *, dc_task)
+		__field(unsigned long long, ino)
+		__field(unsigned int, flags)
+	),
+	TP_fast_assign(
+		__entry->task = task;
+		__entry->dc_task = dc_task;
+		__entry->ino = ino;
+		__entry->flags = flags;
+	),
+	TP_printk("%p %p %llu %u", __entry->task, __entry->dc_task,
+		  __entry->ino, __entry->flags)
+);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_query_inode_wipe_begin);
+
+DEFINE_OCFS2_UINT_EVENT(ocfs2_query_inode_wipe_succ);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_query_inode_wipe_end);
+
+DEFINE_OCFS2_ULL_INT_EVENT(ocfs2_cleanup_delete_inode);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_delete_inode);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_clear_inode);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_drop_inode);
+
+TRACE_EVENT(ocfs2_inode_revalidate,
+	TP_PROTO(void *inode, unsigned long long ino,
+		 unsigned int flags),
+	TP_ARGS(inode, ino, flags),
+	TP_STRUCT__entry(
+		__field(void *, inode)
+		__field(unsigned long long, ino)
+		__field(unsigned int, flags)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->ino = ino;
+		__entry->flags = flags;
+	),
+	TP_printk("%p %llu %u", __entry->inode, __entry->ino, __entry->flags)
+);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_mark_inode_dirty);
+
+/* End of trace events for fs/ocfs2/inode.c. */
+
+/* Trace events for fs/ocfs2/extent_map.c. */
+
+TRACE_EVENT(ocfs2_read_virt_blocks,
+	TP_PROTO(void *inode, unsigned long long vblock, int nr,
+		 void *bhs, unsigned int flags, void *validate),
+	TP_ARGS(inode, vblock, nr, bhs, flags, validate),
+	TP_STRUCT__entry(
+		__field(void *, inode)
+		__field(unsigned long long, vblock)
+		__field(int, nr)
+		__field(void *, bhs)
+		__field(unsigned int, flags)
+		__field(void *, validate)
+	),
+	TP_fast_assign(
+		__entry->inode = inode;
+		__entry->vblock = vblock;
+		__entry->nr = nr;
+		__entry->bhs = bhs;
+		__entry->flags = flags;
+		__entry->validate = validate;
+	),
+	TP_printk("%p %llu %d %p %x %p", __entry->inode, __entry->vblock,
+		  __entry->nr, __entry->bhs, __entry->flags, __entry->validate)
+);
+
+/* End of trace events for fs/ocfs2/extent_map.c. */
+
+/* Trace events for fs/ocfs2/slot_map.c. */
+
+DEFINE_OCFS2_UINT_EVENT(ocfs2_refresh_slot_info);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_map_slot_buffers);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_map_slot_buffers_block);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_find_slot);
+
+/* End of trace events for fs/ocfs2/slot_map.c. */
+
+/* Trace events for fs/ocfs2/heartbeat.c. */
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_do_node_down);
+
+/* End of trace events for fs/ocfs2/heartbeat.c. */
+
+/* Trace events for fs/ocfs2/super.c. */
+
+TRACE_EVENT(ocfs2_remount,
+	TP_PROTO(unsigned long s_flags, unsigned long osb_flags, int flags),
+	TP_ARGS(s_flags, osb_flags, flags),
+	TP_STRUCT__entry(
+		__field(unsigned long, s_flags)
+		__field(unsigned long, osb_flags)
+		__field(int, flags)
+	),
+	TP_fast_assign(
+		__entry->s_flags = s_flags;
+		__entry->osb_flags = osb_flags;
+		__entry->flags = flags;
+	),
+	TP_printk("%lu %lu %d", __entry->s_flags,
+		  __entry->osb_flags, __entry->flags)
+);
+
+TRACE_EVENT(ocfs2_fill_super,
+	TP_PROTO(void *sb, void *data, int silent),
+	TP_ARGS(sb, data, silent),
+	TP_STRUCT__entry(
+		__field(void *, sb)
+		__field(void *, data)
+		__field(int, silent)
+	),
+	TP_fast_assign(
+		__entry->sb = sb;
+		__entry->data = data;
+		__entry->silent = silent;
+	),
+	TP_printk("%p %p %d", __entry->sb,
+		  __entry->data, __entry->silent)
+);
+
+TRACE_EVENT(ocfs2_parse_options,
+	TP_PROTO(int is_remount, char *options),
+	TP_ARGS(is_remount, options),
+	TP_STRUCT__entry(
+		__field(int, is_remount)
+		__string(options, options)
+	),
+	TP_fast_assign(
+		__entry->is_remount = is_remount;
+		__assign_str(options, options);
+	),
+	TP_printk("%d %s", __entry->is_remount, __get_str(options))
+);
+
+DEFINE_OCFS2_POINTER_EVENT(ocfs2_put_super);
+
+TRACE_EVENT(ocfs2_statfs,
+	TP_PROTO(void *sb, void *buf),
+	TP_ARGS(sb, buf),
+	TP_STRUCT__entry(
+		__field(void *, sb)
+		__field(void *, buf)
+	),
+	TP_fast_assign(
+		__entry->sb = sb;
+		__entry->buf = buf;
+	),
+	TP_printk("%p %p", __entry->sb, __entry->buf)
+);
+
+DEFINE_OCFS2_POINTER_EVENT(ocfs2_dismount_volume);
+
+TRACE_EVENT(ocfs2_initialize_super,
+	TP_PROTO(char *label, char *uuid_str, unsigned long long root_dir,
+		 unsigned long long system_dir, int cluster_bits),
+	TP_ARGS(label, uuid_str, root_dir, system_dir, cluster_bits),
+	TP_STRUCT__entry(
+		__string(label, label)
+		__string(uuid_str, uuid_str)
+		__field(unsigned long long, root_dir)
+		__field(unsigned long long, system_dir)
+		__field(int, cluster_bits)
+	),
+	TP_fast_assign(
+		__assign_str(label, label);
+		__assign_str(uuid_str, uuid_str);
+		__entry->root_dir = root_dir;
+		__entry->system_dir = system_dir;
+		__entry->cluster_bits = cluster_bits;
+	),
+	TP_printk("%s %s %llu %llu %d", __get_str(label), __get_str(uuid_str),
+		  __entry->root_dir, __entry->system_dir, __entry->cluster_bits)
+);
+
+/* End of trace events for fs/ocfs2/super.c. */
+
+/* Trace events for fs/ocfs2/xattr.c. */
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_validate_xattr_block);
+
+DEFINE_OCFS2_UINT_EVENT(ocfs2_xattr_extend_allocation);
+
+TRACE_EVENT(ocfs2_init_xattr_set_ctxt,
+	TP_PROTO(const char *name, int meta, int clusters, int credits),
+	TP_ARGS(name, meta, clusters, credits),
+	TP_STRUCT__entry(
+		__string(name, name)
+		__field(int, meta)
+		__field(int, clusters)
+		__field(int, credits)
+	),
+	TP_fast_assign(
+		__assign_str(name, name);
+		__entry->meta = meta;
+		__entry->clusters = clusters;
+		__entry->credits = credits;
+	),
+	TP_printk("%s %d %d %d", __get_str(name), __entry->meta,
+		  __entry->clusters, __entry->credits)
+);
+
+DECLARE_EVENT_CLASS(ocfs2__xattr_find,
+	TP_PROTO(unsigned long long ino, const char *name, int name_index,
+		 unsigned int hash, unsigned long long location,
+		 int xe_index),
+	TP_ARGS(ino, name, name_index, hash, location, xe_index),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__string(name, name)
+		__field(int, name_index)
+		__field(unsigned int, hash)
+		__field(unsigned long long, location)
+		__field(int, xe_index)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__assign_str(name, name);
+		__entry->name_index = name_index;
+		__entry->hash = hash;
+		__entry->location = location;
+		__entry->xe_index = xe_index;
+	),
+	TP_printk("%llu %s %d %u %llu %d", __entry->ino, __get_str(name),
+		  __entry->name_index, __entry->hash, __entry->location,
+		  __entry->xe_index)
+);
+
+#define DEFINE_OCFS2_XATTR_FIND_EVENT(name)					\
+DEFINE_EVENT(ocfs2__xattr_find, name,					\
+TP_PROTO(unsigned long long ino, const char *name, int name_index,	\
+	 unsigned int hash, unsigned long long bucket,			\
+	 int xe_index),							\
+	TP_ARGS(ino, name, name_index, hash, bucket, xe_index))
+
+DEFINE_OCFS2_XATTR_FIND_EVENT(ocfs2_xattr_bucket_find);
+
+DEFINE_OCFS2_XATTR_FIND_EVENT(ocfs2_xattr_index_block_find);
+
+DEFINE_OCFS2_XATTR_FIND_EVENT(ocfs2_xattr_index_block_find_rec);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_iterate_xattr_buckets);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_iterate_xattr_bucket);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_cp_xattr_block_to_bucket_begin);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_cp_xattr_block_to_bucket_end);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_xattr_create_index_block_begin);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_xattr_create_index_block);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_defrag_xattr_bucket);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_mv_xattr_bucket_cross_cluster);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_divide_xattr_bucket_begin);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_divide_xattr_bucket_move);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_cp_xattr_bucket);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_mv_xattr_buckets);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_adjust_xattr_cross_cluster);
+
+DEFINE_OCFS2_ULL_ULL_UINT_UINT_EVENT(ocfs2_add_new_xattr_cluster_begin);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_add_new_xattr_cluster);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_add_new_xattr_cluster_insert);
+
+DEFINE_OCFS2_ULL_ULL_UINT_UINT_EVENT(ocfs2_extend_xattr_bucket);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_add_new_xattr_bucket);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_xattr_bucket_value_truncate);
+
+DEFINE_OCFS2_ULL_ULL_UINT_UINT_EVENT(ocfs2_rm_xattr_cluster);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_reflink_xattr_header);
+
+DEFINE_OCFS2_ULL_INT_EVENT(ocfs2_create_empty_xattr_block);
+
+DEFINE_OCFS2_STRING_EVENT(ocfs2_xattr_set_entry_bucket);
+
+DEFINE_OCFS2_STRING_EVENT(ocfs2_xattr_set_entry_index_block);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_xattr_bucket_value_refcount);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_reflink_xattr_buckets);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_reflink_xattr_rec);
+
+/* End of trace events for fs/ocfs2/xattr.c. */
+
+/* Trace events for fs/ocfs2/reservations.c. */
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_resv_insert);
+
+DEFINE_OCFS2_ULL_UINT_UINT_UINT_EVENT(ocfs2_resmap_find_free_bits_begin);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_resmap_find_free_bits_end);
+
+TRACE_EVENT(ocfs2_resv_find_window_begin,
+	TP_PROTO(unsigned int r_start, unsigned int r_end, unsigned int goal,
+		 unsigned int wanted, int empty_root),
+	TP_ARGS(r_start, r_end, goal, wanted, empty_root),
+	TP_STRUCT__entry(
+		__field(unsigned int, r_start)
+		__field(unsigned int, r_end)
+		__field(unsigned int, goal)
+		__field(unsigned int, wanted)
+		__field(int, empty_root)
+	),
+	TP_fast_assign(
+		__entry->r_start = r_start;
+		__entry->r_end = r_end;
+		__entry->goal = goal;
+		__entry->wanted = wanted;
+		__entry->empty_root = empty_root;
+	),
+	TP_printk("%u %u %u %u %d", __entry->r_start, __entry->r_end,
+		  __entry->goal, __entry->wanted, __entry->empty_root)
+);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_resv_find_window_prev);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_resv_find_window_next);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_cannibalize_resv_begin);
+
+TRACE_EVENT(ocfs2_cannibalize_resv_end,
+	TP_PROTO(unsigned int start, unsigned int end, unsigned int len,
+		 unsigned int last_start, unsigned int last_len),
+	TP_ARGS(start, end, len, last_start, last_len),
+	TP_STRUCT__entry(
+		__field(unsigned int, start)
+		__field(unsigned int, end)
+		__field(unsigned int, len)
+		__field(unsigned int, last_start)
+		__field(unsigned int, last_len)
+	),
+	TP_fast_assign(
+		__entry->start = start;
+		__entry->end = end;
+		__entry->len = len;
+		__entry->last_start = last_start;
+		__entry->last_len = last_len;
+	),
+	TP_printk("%u %u %u %u %u", __entry->start, __entry->end,
+		  __entry->len, __entry->last_start, __entry->last_len)
+);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_resmap_resv_bits);
+
+TRACE_EVENT(ocfs2_resmap_claimed_bits_begin,
+	TP_PROTO(unsigned int cstart, unsigned int cend, unsigned int clen,
+		 unsigned int r_start, unsigned int r_end, unsigned int r_len,
+		 unsigned int last_start, unsigned int last_len),
+	TP_ARGS(cstart, cend, clen, r_start, r_end,
+		r_len, last_start, last_len),
+	TP_STRUCT__entry(
+		__field(unsigned int, cstart)
+		__field(unsigned int, cend)
+		__field(unsigned int, clen)
+		__field(unsigned int, r_start)
+		__field(unsigned int, r_end)
+		__field(unsigned int, r_len)
+		__field(unsigned int, last_start)
+		__field(unsigned int, last_len)
+	),
+	TP_fast_assign(
+		__entry->cstart = cstart;
+		__entry->cend = cend;
+		__entry->clen = clen;
+		__entry->r_start = r_start;
+		__entry->r_end = r_end;
+		__entry->r_len = r_len;
+		__entry->last_start = last_start;
+		__entry->last_len = last_len;
+	),
+	TP_printk("%u %u %u %u %u %u %u %u",
+		  __entry->cstart, __entry->cend, __entry->clen,
+		  __entry->r_start, __entry->r_end, __entry->r_len,
+		  __entry->last_start, __entry->last_len)
+);
+
+TRACE_EVENT(ocfs2_resmap_claimed_bits_end,
+	TP_PROTO(unsigned int start, unsigned int end, unsigned int len,
+		 unsigned int last_start, unsigned int last_len),
+	TP_ARGS(start, end, len, last_start, last_len),
+	TP_STRUCT__entry(
+		__field(unsigned int, start)
+		__field(unsigned int, end)
+		__field(unsigned int, len)
+		__field(unsigned int, last_start)
+		__field(unsigned int, last_len)
+	),
+	TP_fast_assign(
+		__entry->start = start;
+		__entry->end = end;
+		__entry->len = len;
+		__entry->last_start = last_start;
+		__entry->last_len = last_len;
+	),
+	TP_printk("%u %u %u %u %u", __entry->start, __entry->end,
+		  __entry->len, __entry->last_start, __entry->last_len)
+);
+
+/* End of trace events for fs/ocfs2/reservations.c. */
+
+/* Trace events for fs/ocfs2/quota_local.c. */
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_recover_local_quota_file);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_finish_quota_recovery);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(olq_set_dquot);
+
+/* End of trace events for fs/ocfs2/quota_local.c. */
+
+/* Trace events for fs/ocfs2/quota_global.c. */
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_validate_quota_block);
+
+TRACE_EVENT(ocfs2_sync_dquot,
+	TP_PROTO(unsigned int dq_id, long long dqb_curspace,
+		 long long spacechange, long long curinodes,
+		 long long inodechange),
+	TP_ARGS(dq_id, dqb_curspace, spacechange, curinodes, inodechange),
+	TP_STRUCT__entry(
+		__field(unsigned int, dq_id)
+		__field(long long, dqb_curspace)
+		__field(long long, spacechange)
+		__field(long long, curinodes)
+		__field(long long, inodechange)
+	),
+	TP_fast_assign(
+		__entry->dq_id = dq_id;
+		__entry->dqb_curspace = dqb_curspace;
+		__entry->spacechange = spacechange;
+		__entry->curinodes = curinodes;
+		__entry->inodechange = inodechange;
+	),
+	TP_printk("%u %lld %lld %lld %lld", __entry->dq_id,
+		  __entry->dqb_curspace, __entry->spacechange,
+		  __entry->curinodes, __entry->inodechange)
+);
+
+TRACE_EVENT(ocfs2_sync_dquot_helper,
+	TP_PROTO(unsigned int dq_id, unsigned int dq_type, unsigned long type,
+		 const char *s_id),
+	TP_ARGS(dq_id, dq_type, type, s_id),
+
+	TP_STRUCT__entry(
+		__field(unsigned int, dq_id)
+		__field(unsigned int, dq_type)
+		__field(unsigned long, type)
+		__string(s_id, s_id)
+	),
+	TP_fast_assign(
+		__entry->dq_id = dq_id;
+		__entry->dq_type = dq_type;
+		__entry->type = type;
+		__assign_str(s_id, s_id);
+	),
+	TP_printk("%u %u %lu %s", __entry->dq_id, __entry->dq_type,
+		  __entry->type, __get_str(s_id))
+);
+
+DEFINE_OCFS2_UINT_INT_EVENT(ocfs2_write_dquot);
+
+DEFINE_OCFS2_UINT_INT_EVENT(ocfs2_release_dquot);
+
+DEFINE_OCFS2_UINT_INT_EVENT(ocfs2_acquire_dquot);
+
+DEFINE_OCFS2_UINT_INT_EVENT(ocfs2_mark_dquot_dirty);
+
+/* End of trace events for fs/ocfs2/quota_global.c. */
+
+/* Trace events for fs/ocfs2/dir.c. */
+DEFINE_OCFS2_INT_EVENT(ocfs2_search_dirblock);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_validate_dir_block);
+
+DEFINE_OCFS2_POINTER_EVENT(ocfs2_find_entry_el);
+
+TRACE_EVENT(ocfs2_dx_dir_search,
+	TP_PROTO(unsigned long long ino, int namelen, const char *name,
+		 unsigned int major_hash, unsigned int minor_hash,
+		 unsigned long long blkno),
+	TP_ARGS(ino, namelen, name, major_hash, minor_hash, blkno),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(int, namelen)
+		__string(name, name)
+		__field(unsigned int, major_hash)
+		__field(unsigned int,minor_hash)
+		__field(unsigned long long, blkno)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->namelen = namelen;
+		__assign_str(name, name);
+		__entry->major_hash = major_hash;
+		__entry->minor_hash = minor_hash;
+		__entry->blkno = blkno;
+	),
+	TP_printk("%llu %.*s %u %u %llu", __entry->ino,
+		   __entry->namelen, __get_str(name),
+		  __entry->major_hash, __entry->minor_hash, __entry->blkno)
+);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_dx_dir_search_leaf_info);
+
+DEFINE_OCFS2_ULL_INT_EVENT(ocfs2_delete_entry_dx);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_readdir);
+
+TRACE_EVENT(ocfs2_find_files_on_disk,
+	TP_PROTO(int namelen, const char *name, void *blkno,
+		 unsigned long long dir),
+	TP_ARGS(namelen, name, blkno, dir),
+	TP_STRUCT__entry(
+		__field(int, namelen)
+		__string(name, name)
+		__field(void *, blkno)
+		__field(unsigned long long, dir)
+	),
+	TP_fast_assign(
+		__entry->namelen = namelen;
+		__assign_str(name, name);
+		__entry->blkno = blkno;
+		__entry->dir = dir;
+	),
+	TP_printk("%.*s %p %llu", __entry->namelen, __get_str(name),
+		  __entry->blkno, __entry->dir)
+);
+
+TRACE_EVENT(ocfs2_check_dir_for_entry,
+	TP_PROTO(unsigned long long dir, int namelen, const char *name),
+	TP_ARGS(dir, namelen, name),
+	TP_STRUCT__entry(
+		__field(unsigned long long, dir)
+		__field(int, namelen)
+		__string(name, name)
+	),
+	TP_fast_assign(
+		__entry->dir = dir;
+		__entry->namelen = namelen;
+		__assign_str(name, name);
+	),
+	TP_printk("%llu %.*s", __entry->dir,
+		  __entry->namelen, __get_str(name))
+);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_dx_dir_attach_index);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_dx_dir_format_cluster);
+
+TRACE_EVENT(ocfs2_dx_dir_index_root_block,
+	TP_PROTO(unsigned long long dir,
+		 unsigned int major_hash, unsigned int minor_hash,
+		 int namelen, const char *name, unsigned int num_used),
+	TP_ARGS(dir, major_hash, minor_hash, namelen, name, num_used),
+	TP_STRUCT__entry(
+		__field(unsigned long long, dir)
+		__field(unsigned int, major_hash)
+		__field(unsigned int, minor_hash)
+		__field(int, namelen)
+		__string(name, name)
+		__field(unsigned int, num_used)
+	),
+	TP_fast_assign(
+		__entry->dir = dir;
+		__entry->major_hash = major_hash;
+		__entry->minor_hash = minor_hash;
+		__entry->namelen = namelen;
+		__assign_str(name, name);
+		__entry->num_used = num_used;
+	),
+	TP_printk("%llu %x %x %.*s %u", __entry->dir,
+		  __entry->major_hash, __entry->minor_hash,
+		   __entry->namelen, __get_str(name), __entry->num_used)
+);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_extend_dir);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_dx_dir_rebalance);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_dx_dir_rebalance_split);
+
+DEFINE_OCFS2_ULL_INT_EVENT(ocfs2_prepare_dir_for_insert);
+
+/* End of trace events for fs/ocfs2/dir.c. */
+
+/* Trace events for fs/ocfs2/namei.c. */
+
+DECLARE_EVENT_CLASS(ocfs2__dentry_ops,
+	TP_PROTO(void *dir, void *dentry, int name_len, const char *name,
+		 unsigned long long dir_blkno, unsigned long long extra),
+	TP_ARGS(dir, dentry, name_len, name, dir_blkno, extra),
+	TP_STRUCT__entry(
+		__field(void *, dir)
+		__field(void *, dentry)
+		__field(int, name_len)
+		__string(name, name)
+		__field(unsigned long long, dir_blkno)
+		__field(unsigned long long, extra)
+	),
+	TP_fast_assign(
+		__entry->dir = dir;
+		__entry->dentry = dentry;
+		__entry->name_len = name_len;
+		__assign_str(name, name);
+		__entry->dir_blkno = dir_blkno;
+		__entry->extra = extra;
+	),
+	TP_printk("%p %p %.*s %llu %llu", __entry->dir, __entry->dentry,
+		  __entry->name_len, __get_str(name),
+		  __entry->dir_blkno, __entry->extra)
+);
+
+#define DEFINE_OCFS2_DENTRY_OPS(name)					\
+DEFINE_EVENT(ocfs2__dentry_ops, name,					\
+TP_PROTO(void *dir, void *dentry, int name_len, const char *name,	\
+	 unsigned long long dir_blkno, unsigned long long extra),	\
+	TP_ARGS(dir, dentry, name_len, name, dir_blkno, extra))
+
+DEFINE_OCFS2_DENTRY_OPS(ocfs2_lookup);
+
+DEFINE_OCFS2_DENTRY_OPS(ocfs2_mkdir);
+
+DEFINE_OCFS2_DENTRY_OPS(ocfs2_create);
+
+DEFINE_OCFS2_DENTRY_OPS(ocfs2_unlink);
+
+DEFINE_OCFS2_DENTRY_OPS(ocfs2_symlink_create);
+
+DEFINE_OCFS2_DENTRY_OPS(ocfs2_mv_orphaned_inode_to_new);
+
+DEFINE_OCFS2_POINTER_EVENT(ocfs2_lookup_ret);
+
+TRACE_EVENT(ocfs2_mknod,
+	TP_PROTO(void *dir, void *dentry, int name_len, const char *name,
+		 unsigned long long dir_blkno, unsigned long dev, int mode),
+	TP_ARGS(dir, dentry, name_len, name, dir_blkno, dev, mode),
+	TP_STRUCT__entry(
+		__field(void *, dir)
+		__field(void *, dentry)
+		__field(int, name_len)
+		__string(name, name)
+		__field(unsigned long long, dir_blkno)
+		__field(unsigned long, dev)
+		__field(int, mode)
+	),
+	TP_fast_assign(
+		__entry->dir = dir;
+		__entry->dentry = dentry;
+		__entry->name_len = name_len;
+		__assign_str(name, name);
+		__entry->dir_blkno = dir_blkno;
+		__entry->dev = dev;
+		__entry->mode = mode;
+	),
+	TP_printk("%p %p %.*s %llu %lu %d", __entry->dir, __entry->dentry,
+		  __entry->name_len, __get_str(name),
+		  __entry->dir_blkno, __entry->dev, __entry->mode)
+);
+
+TRACE_EVENT(ocfs2_link,
+	TP_PROTO(unsigned long long ino, int old_len, const char *old_name,
+		 int name_len, const char *name),
+	TP_ARGS(ino, old_len, old_name, name_len, name),
+	TP_STRUCT__entry(
+		__field(unsigned long long, ino)
+		__field(int, old_len)
+		__string(old_name, old_name)
+		__field(int, name_len)
+		__string(name, name)
+	),
+	TP_fast_assign(
+		__entry->ino = ino;
+		__entry->old_len = old_len;
+		__assign_str(old_name, old_name);
+		__entry->name_len = name_len;
+		__assign_str(name, name);
+	),
+	TP_printk("%llu %.*s %.*s", __entry->ino,
+		  __entry->old_len, __get_str(old_name),
+		  __entry->name_len, __get_str(name))
+);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_unlink_noent);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_double_lock);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_double_lock_end);
+
+TRACE_EVENT(ocfs2_rename,
+	TP_PROTO(void *old_dir, void *old_dentry,
+		 void *new_dir, void *new_dentry,
+		 int old_len, const char *old_name,
+		 int new_len, const char *new_name),
+	TP_ARGS(old_dir, old_dentry, new_dir, new_dentry,
+		old_len, old_name, new_len, new_name),
+	TP_STRUCT__entry(
+		__field(void *, old_dir)
+		__field(void *, old_dentry)
+		__field(void *, new_dir)
+		__field(void *, new_dentry)
+		__field(int, old_len)
+		__string(old_name, old_name)
+		__field(int, new_len)
+		__string(new_name, new_name)
+	),
+	TP_fast_assign(
+		__entry->old_dir = old_dir;
+		__entry->old_dentry = old_dentry;
+		__entry->new_dir = new_dir;
+		__entry->new_dentry = new_dentry;
+		__entry->old_len = old_len;
+		__assign_str(old_name, old_name);
+		__entry->new_len = new_len;
+		__assign_str(new_name, new_name);
+	),
+	TP_printk("%p %p %p %p %.*s %.*s",
+		  __entry->old_dir, __entry->old_dentry,
+		  __entry->new_dir, __entry->new_dentry,
+		  __entry->old_len, __get_str(old_name),
+		  __entry->new_len, __get_str(new_name))
+);
+
+TRACE_EVENT(ocfs2_rename_target_exists,
+	TP_PROTO(int new_len, const char *new_name),
+	TP_ARGS(new_len, new_name),
+	TP_STRUCT__entry(
+		__field(int, new_len)
+		__string(new_name, new_name)
+	),
+	TP_fast_assign(
+		__entry->new_len = new_len;
+		__assign_str(new_name, new_name);
+	),
+	TP_printk("%.*s", __entry->new_len, __get_str(new_name))
+);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_rename_disagree);
+
+TRACE_EVENT(ocfs2_rename_over_existing,
+	TP_PROTO(unsigned long long new_blkno, void *new_bh,
+		 unsigned long long newdi_blkno),
+	TP_ARGS(new_blkno, new_bh, newdi_blkno),
+	TP_STRUCT__entry(
+		__field(unsigned long long, new_blkno)
+		__field(void *, new_bh)
+		__field(unsigned long long, newdi_blkno)
+	),
+	TP_fast_assign(
+		__entry->new_blkno = new_blkno;
+		__entry->new_bh = new_bh;
+		__entry->newdi_blkno = newdi_blkno;
+	),
+	TP_printk("%llu %p %llu", __entry->new_blkno, __entry->new_bh,
+		  __entry->newdi_blkno)
+);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_create_symlink_data);
+
+TRACE_EVENT(ocfs2_symlink_begin,
+	TP_PROTO(void *dir, void *dentry, const char *symname,
+		 int len, const char *name),
+	TP_ARGS(dir, dentry, symname, len, name),
+	TP_STRUCT__entry(
+		__field(void *, dir)
+		__field(void *, dentry)
+		__field(const char *, symname)
+		__field(int, len)
+		__string(name, name)
+	),
+	TP_fast_assign(
+		__entry->dir = dir;
+		__entry->dentry = dentry;
+		__entry->symname = symname;
+		__entry->len = len;
+		__assign_str(name, name);
+	),
+	TP_printk("%p %p %s %.*s", __entry->dir, __entry->dentry,
+		  __entry->symname, __entry->len, __get_str(name))
+);
+
+TRACE_EVENT(ocfs2_blkno_stringify,
+	TP_PROTO(unsigned long long blkno, const char *name, int namelen),
+	TP_ARGS(blkno, name, namelen),
+	TP_STRUCT__entry(
+		__field(unsigned long long, blkno)
+		__string(name, name)
+		__field(int, namelen)
+	),
+	TP_fast_assign(
+		__entry->blkno = blkno;
+		__assign_str(name, name);
+		__entry->namelen = namelen;
+	),
+	TP_printk("%llu %s %d", __entry->blkno, __get_str(name),
+		  __entry->namelen)
+);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_orphan_add_begin);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_orphan_add_end);
+
+TRACE_EVENT(ocfs2_orphan_del,
+	TP_PROTO(unsigned long long dir, const char *name, int namelen),
+	TP_ARGS(dir, name, namelen),
+	TP_STRUCT__entry(
+		__field(unsigned long long, dir)
+		__string(name, name)
+		__field(int, namelen)
+	),
+	TP_fast_assign(
+		__entry->dir = dir;
+		__assign_str(name, name);
+		__entry->namelen = namelen;
+	),
+	TP_printk("%llu %s %d", __entry->dir, __get_str(name),
+		  __entry->namelen)
+);
+
+/* End of trace events for fs/ocfs2/namei.c. */
+
+/* Trace events for fs/ocfs2/dcache.c. */
+
+TRACE_EVENT(ocfs2_dentry_revalidate,
+	TP_PROTO(void *dentry, int len, const char *name),
+	TP_ARGS(dentry, len, name),
+	TP_STRUCT__entry(
+		__field(void *, dentry)
+		__field(int, len)
+		__string(name, name)
+	),
+	TP_fast_assign(
+		__entry->dentry = dentry;
+		__entry->len = len;
+		__assign_str(name, name);
+	),
+	TP_printk("%p %.*s", __entry->dentry, __entry->len, __get_str(name))
+);
+
+TRACE_EVENT(ocfs2_dentry_revalidate_negative,
+	TP_PROTO(int len, const char *name, unsigned long pgen,
+		 unsigned long gen),
+	TP_ARGS(len, name, pgen, gen),
+	TP_STRUCT__entry(
+		__field(int, len)
+		__string(name, name)
+		__field(unsigned long, pgen)
+		__field(unsigned long, gen)
+	),
+	TP_fast_assign(
+		__entry->len = len;
+		__assign_str(name, name);
+		__entry->pgen = pgen;
+		__entry->gen = gen;
+	),
+	TP_printk("%.*s %lu %lu", __entry->len, __get_str(name),
+		  __entry->pgen, __entry->gen)
+);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_dentry_revalidate_delete);
+
+DEFINE_OCFS2_ULL_INT_EVENT(ocfs2_dentry_revalidate_orphaned);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_dentry_revalidate_nofsdata);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_dentry_revalidate_ret);
+
+TRACE_EVENT(ocfs2_find_local_alias,
+	TP_PROTO(int len, const char *name),
+	TP_ARGS(len, name),
+	TP_STRUCT__entry(
+		__field(int, len)
+		__string(name, name)
+	),
+	TP_fast_assign(
+		__entry->len = len;
+		__assign_str(name, name);
+	),
+	TP_printk("%.*s", __entry->len, __get_str(name))
+);
+
+TRACE_EVENT(ocfs2_dentry_attach_lock,
+	TP_PROTO(int len, const char *name,
+		 unsigned long long parent, void *fsdata),
+	TP_ARGS(len, name, parent, fsdata),
+	TP_STRUCT__entry(
+		__field(int, len)
+		__string(name, name)
+		__field(unsigned long long, parent)
+		__field(void *, fsdata)
+	),
+	TP_fast_assign(
+		__entry->len = len;
+		__assign_str(name, name);
+		__entry->parent = parent;
+		__entry->fsdata = fsdata;
+	),
+	TP_printk("%.*s %llu %p", __entry->len, __get_str(name),
+		  __entry->parent, __entry->fsdata)
+);
+
+TRACE_EVENT(ocfs2_dentry_attach_lock_found,
+	TP_PROTO(const char *name, unsigned long long parent,
+		 unsigned long long ino),
+	TP_ARGS(name, parent, ino),
+	TP_STRUCT__entry(
+		__string(name, name)
+		__field(unsigned long long, parent)
+		__field(unsigned long long, ino)
+	),
+	TP_fast_assign(
+		__assign_str(name, name);
+		__entry->parent = parent;
+		__entry->ino = ino;
+	),
+	TP_printk("%s %llu %llu", __get_str(name), __entry->parent, __entry->ino)
+);
+/* End of trace events for fs/ocfs2/dcache.c. */
+
+/* Trace events for fs/ocfs2/export.c. */
+
+TRACE_EVENT(ocfs2_get_dentry_begin,
+	TP_PROTO(void *sb, void *handle, unsigned long long blkno),
+	TP_ARGS(sb, handle, blkno),
+	TP_STRUCT__entry(
+		__field(void *, sb)
+		__field(void *, handle)
+		__field(unsigned long long, blkno)
+	),
+	TP_fast_assign(
+		__entry->sb = sb;
+		__entry->handle = handle;
+		__entry->blkno = blkno;
+	),
+	TP_printk("%p %p %llu", __entry->sb, __entry->handle, __entry->blkno)
+);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_get_dentry_test_bit);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_get_dentry_stale);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_get_dentry_generation);
+
+DEFINE_OCFS2_POINTER_EVENT(ocfs2_get_dentry_end);
+
+TRACE_EVENT(ocfs2_get_parent,
+	TP_PROTO(void *child, int len, const char *name,
+		 unsigned long long ino),
+	TP_ARGS(child, len, name, ino),
+	TP_STRUCT__entry(
+		__field(void *,	child)
+		__field(int, len)
+		__string(name, name)
+		__field(unsigned long long, ino)
+	),
+	TP_fast_assign(
+		__entry->child = child;
+		__entry->len = len;
+		__assign_str(name, name);
+		__entry->ino = ino;
+	),
+	TP_printk("%p %.*s %llu", __entry->child, __entry->len,
+		  __get_str(name), __entry->ino)
+);
+
+DEFINE_OCFS2_POINTER_EVENT(ocfs2_get_parent_end);
+
+TRACE_EVENT(ocfs2_encode_fh_begin,
+	TP_PROTO(void *dentry, int name_len, const char *name,
+		 void *fh, int len, int connectable),
+	TP_ARGS(dentry, name_len, name, fh, len, connectable),
+	TP_STRUCT__entry(
+		__field(void *, dentry)
+		__field(int, name_len)
+		__string(name, name)
+		__field(void *, fh)
+		__field(int, len)
+		__field(int, connectable)
+	),
+	TP_fast_assign(
+		__entry->dentry = dentry;
+		__entry->name_len = name_len;
+		__assign_str(name, name);
+		__entry->fh = fh;
+		__entry->len = len;
+		__entry->connectable = connectable;
+	),
+	TP_printk("%p %.*s %p %d %d", __entry->dentry, __entry->name_len,
+		  __get_str(name), __entry->fh, __entry->len,
+		  __entry->connectable)
+);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_encode_fh_self);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_encode_fh_parent);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_encode_fh_type);
+
+/* End of trace events for fs/ocfs2/export.c. */
+
+/* Trace events for fs/ocfs2/journal.c. */
+
+DEFINE_OCFS2_UINT_EVENT(ocfs2_commit_cache_begin);
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_commit_cache_end);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_extend_trans);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_extend_trans_restart);
+
+DEFINE_OCFS2_ULL_ULL_UINT_UINT_EVENT(ocfs2_journal_access);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_journal_dirty);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_journal_init);
+
+DEFINE_OCFS2_UINT_EVENT(ocfs2_journal_init_maxlen);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_journal_shutdown);
+
+DEFINE_OCFS2_POINTER_EVENT(ocfs2_journal_shutdown_wait);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_complete_recovery);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_complete_recovery_end);
+
+TRACE_EVENT(ocfs2_complete_recovery_slot,
+	TP_PROTO(int slot, unsigned long long la_ino,
+		 unsigned long long tl_ino, void *qrec),
+	TP_ARGS(slot, la_ino, tl_ino, qrec),
+	TP_STRUCT__entry(
+		__field(int, slot)
+		__field(unsigned long long, la_ino)
+		__field(unsigned long long, tl_ino)
+		__field(void *, qrec)
+	),
+	TP_fast_assign(
+		__entry->slot = slot;
+		__entry->la_ino = la_ino;
+		__entry->tl_ino = tl_ino;
+		__entry->qrec = qrec;
+	),
+	TP_printk("%d %llu %llu %p", __entry->slot, __entry->la_ino,
+		  __entry->tl_ino, __entry->qrec)
+);
+
+DEFINE_OCFS2_INT_INT_EVENT(ocfs2_recovery_thread_node);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_recovery_thread_end);
+
+TRACE_EVENT(ocfs2_recovery_thread,
+	TP_PROTO(int node_num, int osb_node_num, int disable,
+		 void *recovery_thread, int map_set),
+	TP_ARGS(node_num, osb_node_num, disable, recovery_thread, map_set),
+	TP_STRUCT__entry(
+		__field(int, node_num)
+		__field(int, osb_node_num)
+		__field(int,disable)
+		__field(void *, recovery_thread)
+		__field(int,map_set)
+	),
+	TP_fast_assign(
+		__entry->node_num = node_num;
+		__entry->osb_node_num = osb_node_num;
+		__entry->disable = disable;
+		__entry->recovery_thread = recovery_thread;
+		__entry->map_set = map_set;
+	),
+	TP_printk("%d %d %d %p %d", __entry->node_num,
+		   __entry->osb_node_num, __entry->disable,
+		   __entry->recovery_thread, __entry->map_set)
+);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_replay_journal_recovered);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_replay_journal_lock_err);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_replay_journal_skip);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_recover_node);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_recover_node_skip);
+
+DEFINE_OCFS2_UINT_UINT_EVENT(ocfs2_mark_dead_nodes);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_queue_orphan_scan_begin);
+
+DEFINE_OCFS2_UINT_UINT_UINT_EVENT(ocfs2_queue_orphan_scan_end);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_orphan_filldir);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_recover_orphans);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_recover_orphans_iput);
+
+DEFINE_OCFS2_INT_EVENT(ocfs2_wait_on_mount);
+
+/* End of trace events for fs/ocfs2/journal.c. */
+
+/* Trace events for fs/ocfs2/buffer_head_io.c. */
+
+DEFINE_OCFS2_ULL_UINT_EVENT(ocfs2_read_blocks_sync);
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_read_blocks_sync_jbd);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_read_blocks_from_disk);
+
+DEFINE_OCFS2_ULL_INT_INT_INT_EVENT(ocfs2_read_blocks_bh);
+
+DEFINE_OCFS2_ULL_INT_INT_INT_EVENT(ocfs2_read_blocks_end);
+
+TRACE_EVENT(ocfs2_write_block,
+	TP_PROTO(unsigned long long block, void *ci),
+	TP_ARGS(block, ci),
+	TP_STRUCT__entry(
+		__field(unsigned long long, block)
+		__field(void *, ci)
+	),
+	TP_fast_assign(
+		__entry->block = block;
+		__entry->ci = ci;
+	),
+	TP_printk("%llu %p", __entry->block, __entry->ci)
+);
+
+TRACE_EVENT(ocfs2_read_blocks_begin,
+	TP_PROTO(void *ci, unsigned long long block,
+		 unsigned int nr, int flags),
+	TP_ARGS(ci, block, nr, flags),
+	TP_STRUCT__entry(
+		__field(void *, ci)
+		__field(unsigned long long, block)
+		__field(unsigned int, nr)
+		__field(int, flags)
+	),
+	TP_fast_assign(
+		__entry->ci = ci;
+		__entry->block = block;
+		__entry->nr = nr;
+		__entry->flags = flags;
+	),
+	TP_printk("%p %llu %u %d", __entry->ci, __entry->block,
+		  __entry->nr, __entry->flags)
+);
+
+/* End of trace events for fs/ocfs2/buffer_head_io.c. */
+
+/* Trace events for fs/ocfs2/uptodate.c. */
+
+DEFINE_OCFS2_ULL_EVENT(ocfs2_purge_copied_metadata_tree);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_metadata_cache_purge);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_buffer_cached_begin);
+
+TRACE_EVENT(ocfs2_buffer_cached_end,
+	TP_PROTO(int index, void *item),
+	TP_ARGS(index, item),
+	TP_STRUCT__entry(
+		__field(int, index)
+		__field(void *, item)
+	),
+	TP_fast_assign(
+		__entry->index = index;
+		__entry->item = item;
+	),
+	TP_printk("%d %p", __entry->index, __entry->item)
+);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_append_cache_array);
+
+DEFINE_OCFS2_ULL_ULL_UINT_EVENT(ocfs2_insert_cache_tree);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_expand_cache);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_set_buffer_uptodate);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_set_buffer_uptodate_begin);
+
+DEFINE_OCFS2_ULL_UINT_UINT_EVENT(ocfs2_remove_metadata_array);
+
+DEFINE_OCFS2_ULL_ULL_EVENT(ocfs2_remove_metadata_tree);
+
+DEFINE_OCFS2_ULL_ULL_UINT_UINT_EVENT(ocfs2_remove_block_from_cache);
+
+/* End of trace events for fs/ocfs2/uptodate.c. */
+#endif /* _TRACE_OCFS2_H */
+
+/* This part must be outside protection */
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE ocfs2_trace
+#include <trace/define_trace.h>
diff --git a/fs/ocfs2/quota_global.c b/fs/ocfs2/quota_global.c
index 4607923eb24..64346f836c0 100644
--- a/fs/ocfs2/quota_global.c
+++ b/fs/ocfs2/quota_global.c
@@ -11,7 +11,6 @@
 #include <linux/writeback.h>
 #include <linux/workqueue.h>
 
-#define MLOG_MASK_PREFIX ML_QUOTA
 #include <cluster/masklog.h>
 
 #include "ocfs2_fs.h"
@@ -27,6 +26,7 @@
 #include "super.h"
 #include "buffer_head_io.h"
 #include "quota.h"
+#include "ocfs2_trace.h"
 
 /*
  * Locking of quotas with OCFS2 is rather complex. Here are rules that
@@ -132,8 +132,7 @@ int ocfs2_validate_quota_block(struct super_block *sb, struct buffer_head *bh)
 	struct ocfs2_disk_dqtrailer *dqt =
 		ocfs2_block_dqtrailer(sb->s_blocksize, bh->b_data);
 
-	mlog(0, "Validating quota block %llu\n",
-	     (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_validate_quota_block((unsigned long long)bh->b_blocknr);
 
 	BUG_ON(!buffer_uptodate(bh));
 
@@ -343,8 +342,6 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
 	u64 pcount;
 	int status;
 
-	mlog_entry_void();
-
 	/* Read global header */
 	gqinode = ocfs2_get_system_file_inode(OCFS2_SB(sb), ino[type],
 			OCFS2_INVALID_SLOT);
@@ -404,7 +401,8 @@ int ocfs2_global_read_info(struct super_block *sb, int type)
 			   msecs_to_jiffies(oinfo->dqi_syncms));
 
 out_err:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 out_unlock:
 	ocfs2_unlock_global_qf(oinfo, 0);
@@ -510,9 +508,10 @@ int __ocfs2_sync_dquot(struct dquot *dquot, int freeing)
 	olditime = dquot->dq_dqb.dqb_itime;
 	oldbtime = dquot->dq_dqb.dqb_btime;
 	ocfs2_global_disk2memdqb(dquot, &dqblk);
-	mlog(0, "Syncing global dquot %u space %lld+%lld, inodes %lld+%lld\n",
-	     dquot->dq_id, dquot->dq_dqb.dqb_curspace, (long long)spacechange,
-	     dquot->dq_dqb.dqb_curinodes, (long long)inodechange);
+	trace_ocfs2_sync_dquot(dquot->dq_id, dquot->dq_dqb.dqb_curspace,
+			       (long long)spacechange,
+			       dquot->dq_dqb.dqb_curinodes,
+			       (long long)inodechange);
 	if (!test_bit(DQ_LASTSET_B + QIF_SPACE_B, &dquot->dq_flags))
 		dquot->dq_dqb.dqb_curspace += spacechange;
 	if (!test_bit(DQ_LASTSET_B + QIF_INODES_B, &dquot->dq_flags))
@@ -596,8 +595,8 @@ static int ocfs2_sync_dquot_helper(struct dquot *dquot, unsigned long type)
 	struct ocfs2_super *osb = OCFS2_SB(sb);
 	int status = 0;
 
-	mlog_entry("id=%u qtype=%u type=%lu device=%s\n", dquot->dq_id,
-		   dquot->dq_type, type, sb->s_id);
+	trace_ocfs2_sync_dquot_helper(dquot->dq_id, dquot->dq_type,
+				      type, sb->s_id);
 	if (type != dquot->dq_type)
 		goto out;
 	status = ocfs2_lock_global_qf(oinfo, 1);
@@ -623,7 +622,6 @@ static int ocfs2_sync_dquot_helper(struct dquot *dquot, unsigned long type)
 out_ilock:
 	ocfs2_unlock_global_qf(oinfo, 1);
 out:
-	mlog_exit(status);
 	return status;
 }
 
@@ -649,7 +647,7 @@ static int ocfs2_write_dquot(struct dquot *dquot)
 	struct ocfs2_super *osb = OCFS2_SB(dquot->dq_sb);
 	int status = 0;
 
-	mlog_entry("id=%u, type=%d", dquot->dq_id, dquot->dq_type);
+	trace_ocfs2_write_dquot(dquot->dq_id, dquot->dq_type);
 
 	handle = ocfs2_start_trans(osb, OCFS2_QWRITE_CREDITS);
 	if (IS_ERR(handle)) {
@@ -662,7 +660,6 @@ static int ocfs2_write_dquot(struct dquot *dquot)
 	mutex_unlock(&sb_dqopt(dquot->dq_sb)->dqio_mutex);
 	ocfs2_commit_trans(osb, handle);
 out:
-	mlog_exit(status);
 	return status;
 }
 
@@ -688,7 +685,7 @@ static int ocfs2_release_dquot(struct dquot *dquot)
 	struct ocfs2_super *osb = OCFS2_SB(dquot->dq_sb);
 	int status = 0;
 
-	mlog_entry("id=%u, type=%d", dquot->dq_id, dquot->dq_type);
+	trace_ocfs2_release_dquot(dquot->dq_id, dquot->dq_type);
 
 	mutex_lock(&dquot->dq_lock);
 	/* Check whether we are not racing with some other dqget() */
@@ -724,7 +721,8 @@ out_ilock:
 	ocfs2_unlock_global_qf(oinfo, 1);
 out:
 	mutex_unlock(&dquot->dq_lock);
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -745,7 +743,7 @@ static int ocfs2_acquire_dquot(struct dquot *dquot)
 	int need_alloc = ocfs2_global_qinit_alloc(sb, type);
 	handle_t *handle;
 
-	mlog_entry("id=%u, type=%d", dquot->dq_id, type);
+	trace_ocfs2_acquire_dquot(dquot->dq_id, type);
 	mutex_lock(&dquot->dq_lock);
 	/*
 	 * We need an exclusive lock, because we're going to update use count
@@ -811,7 +809,8 @@ out_dq:
 	set_bit(DQ_ACTIVE_B, &dquot->dq_flags);
 out:
 	mutex_unlock(&dquot->dq_lock);
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -831,7 +830,7 @@ static int ocfs2_mark_dquot_dirty(struct dquot *dquot)
 	handle_t *handle;
 	struct ocfs2_super *osb = OCFS2_SB(sb);
 
-	mlog_entry("id=%u, type=%d", dquot->dq_id, type);
+	trace_ocfs2_mark_dquot_dirty(dquot->dq_id, type);
 
 	/* In case user set some limits, sync dquot immediately to global
 	 * quota file so that information propagates quicker */
@@ -868,7 +867,8 @@ out_dlock:
 out_ilock:
 	ocfs2_unlock_global_qf(oinfo, 1);
 out:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -879,8 +879,6 @@ static int ocfs2_write_info(struct super_block *sb, int type)
 	int status = 0;
 	struct ocfs2_mem_dqinfo *oinfo = sb_dqinfo(sb, type)->dqi_priv;
 
-	mlog_entry_void();
-
 	status = ocfs2_lock_global_qf(oinfo, 1);
 	if (status < 0)
 		goto out;
@@ -895,7 +893,8 @@ static int ocfs2_write_info(struct super_block *sb, int type)
 out_ilock:
 	ocfs2_unlock_global_qf(oinfo, 1);
 out:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c
index dc78764ccc4..dc8007fc924 100644
--- a/fs/ocfs2/quota_local.c
+++ b/fs/ocfs2/quota_local.c
@@ -8,7 +8,6 @@
 #include <linux/quotaops.h>
 #include <linux/module.h>
 
-#define MLOG_MASK_PREFIX ML_QUOTA
 #include <cluster/masklog.h>
 
 #include "ocfs2_fs.h"
@@ -23,6 +22,7 @@
 #include "quota.h"
 #include "uptodate.h"
 #include "super.h"
+#include "ocfs2_trace.h"
 
 /* Number of local quota structures per block */
 static inline unsigned int ol_quota_entries_per_block(struct super_block *sb)
@@ -475,7 +475,7 @@ static int ocfs2_recover_local_quota_file(struct inode *lqinode,
 	struct ocfs2_recovery_chunk *rchunk, *next;
 	qsize_t spacechange, inodechange;
 
-	mlog_entry("ino=%lu type=%u", (unsigned long)lqinode->i_ino, type);
+	trace_ocfs2_recover_local_quota_file((unsigned long)lqinode->i_ino, type);
 
 	list_for_each_entry_safe(rchunk, next, &(rec->r_list[type]), rc_list) {
 		chunk = rchunk->rc_chunk;
@@ -575,7 +575,8 @@ out_put_bh:
 	}
 	if (status < 0)
 		free_recovery_list(&(rec->r_list[type]));
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -600,7 +601,7 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb,
 	for (type = 0; type < MAXQUOTAS; type++) {
 		if (list_empty(&(rec->r_list[type])))
 			continue;
-		mlog(0, "Recovering quota in slot %d\n", slot_num);
+		trace_ocfs2_finish_quota_recovery(slot_num);
 		lqinode = ocfs2_get_system_file_inode(osb, ino[type], slot_num);
 		if (!lqinode) {
 			status = -ENOENT;
@@ -882,9 +883,10 @@ static void olq_set_dquot(struct buffer_head *bh, void *private)
 	dqblk->dqb_inodemod = cpu_to_le64(od->dq_dquot.dq_dqb.dqb_curinodes -
 					  od->dq_originodes);
 	spin_unlock(&dq_data_lock);
-	mlog(0, "Writing local dquot %u space %lld inodes %lld\n",
-	     od->dq_dquot.dq_id, (long long)le64_to_cpu(dqblk->dqb_spacemod),
-	     (long long)le64_to_cpu(dqblk->dqb_inodemod));
+	trace_olq_set_dquot(
+		(unsigned long long)le64_to_cpu(dqblk->dqb_spacemod),
+		(unsigned long long)le64_to_cpu(dqblk->dqb_inodemod),
+		od->dq_dquot.dq_id);
 }
 
 /* Write dquot to local quota file */
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index b5f9160e93e..93a81f853e6 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -16,7 +16,6 @@
  */
 
 #include <linux/sort.h>
-#define MLOG_MASK_PREFIX ML_REFCOUNT
 #include <cluster/masklog.h>
 #include "ocfs2.h"
 #include "inode.h"
@@ -34,6 +33,7 @@
 #include "aops.h"
 #include "xattr.h"
 #include "namei.h"
+#include "ocfs2_trace.h"
 
 #include <linux/bio.h>
 #include <linux/blkdev.h>
@@ -84,8 +84,7 @@ static int ocfs2_validate_refcount_block(struct super_block *sb,
 	struct ocfs2_refcount_block *rb =
 		(struct ocfs2_refcount_block *)bh->b_data;
 
-	mlog(0, "Validating refcount block %llu\n",
-	     (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_validate_refcount_block((unsigned long long)bh->b_blocknr);
 
 	BUG_ON(!buffer_uptodate(bh));
 
@@ -545,8 +544,8 @@ void ocfs2_purge_refcount_trees(struct ocfs2_super *osb)
 	while ((node = rb_last(root)) != NULL) {
 		tree = rb_entry(node, struct ocfs2_refcount_tree, rf_node);
 
-		mlog(0, "Purge tree %llu\n",
-		     (unsigned long long) tree->rf_blkno);
+		trace_ocfs2_purge_refcount_trees(
+				(unsigned long long) tree->rf_blkno);
 
 		rb_erase(&tree->rf_node, root);
 		ocfs2_free_refcount_tree(tree);
@@ -575,7 +574,8 @@ static int ocfs2_create_refcount_tree(struct inode *inode,
 
 	BUG_ON(oi->ip_dyn_features & OCFS2_HAS_REFCOUNT_FL);
 
-	mlog(0, "create tree for inode %lu\n", inode->i_ino);
+	trace_ocfs2_create_refcount_tree(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno);
 
 	ret = ocfs2_reserve_new_metadata_blocks(osb, 1, &meta_ac);
 	if (ret) {
@@ -646,8 +646,7 @@ static int ocfs2_create_refcount_tree(struct inode *inode,
 	di->i_refcount_loc = cpu_to_le64(first_blkno);
 	spin_unlock(&oi->ip_lock);
 
-	mlog(0, "created tree for inode %lu, refblock %llu\n",
-	     inode->i_ino, (unsigned long long)first_blkno);
+	trace_ocfs2_create_refcount_tree_blkno((unsigned long long)first_blkno);
 
 	ocfs2_journal_dirty(handle, di_bh);
 
@@ -1256,8 +1255,9 @@ static int ocfs2_change_refcount_rec(handle_t *handle,
 		goto out;
 	}
 
-	mlog(0, "change index %d, old count %u, change %d\n", index,
-	     le32_to_cpu(rec->r_refcount), change);
+	trace_ocfs2_change_refcount_rec(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		index, le32_to_cpu(rec->r_refcount), change);
 	le32_add_cpu(&rec->r_refcount, change);
 
 	if (!rec->r_refcount) {
@@ -1353,8 +1353,8 @@ static int ocfs2_expand_inline_ref_root(handle_t *handle,
 
 	ocfs2_journal_dirty(handle, ref_root_bh);
 
-	mlog(0, "new leaf block %llu, used %u\n", (unsigned long long)blkno,
-	     le16_to_cpu(new_rb->rf_records.rl_used));
+	trace_ocfs2_expand_inline_ref_root((unsigned long long)blkno,
+		le16_to_cpu(new_rb->rf_records.rl_used));
 
 	*ref_leaf_bh = new_bh;
 	new_bh = NULL;
@@ -1466,9 +1466,9 @@ static int ocfs2_divide_leaf_refcount_block(struct buffer_head *ref_leaf_bh,
 			(struct ocfs2_refcount_block *)new_bh->b_data;
 	struct ocfs2_refcount_list *new_rl = &new_rb->rf_records;
 
-	mlog(0, "split old leaf refcount block %llu, count = %u, used = %u\n",
-	     (unsigned long long)ref_leaf_bh->b_blocknr,
-	     le32_to_cpu(rl->rl_count), le32_to_cpu(rl->rl_used));
+	trace_ocfs2_divide_leaf_refcount_block(
+		(unsigned long long)ref_leaf_bh->b_blocknr,
+		le32_to_cpu(rl->rl_count), le32_to_cpu(rl->rl_used));
 
 	/*
 	 * XXX: Improvement later.
@@ -1601,8 +1601,8 @@ static int ocfs2_new_leaf_refcount_block(handle_t *handle,
 
 	ocfs2_init_refcount_extent_tree(&ref_et, ci, ref_root_bh);
 
-	mlog(0, "insert new leaf block %llu at %u\n",
-	     (unsigned long long)new_bh->b_blocknr, new_cpos);
+	trace_ocfs2_new_leaf_refcount_block(
+			(unsigned long long)new_bh->b_blocknr, new_cpos);
 
 	/* Insert the new leaf block with the specific offset cpos. */
 	ret = ocfs2_insert_extent(handle, &ref_et, new_cpos, new_bh->b_blocknr,
@@ -1794,11 +1794,10 @@ static int ocfs2_insert_refcount_rec(handle_t *handle,
 			(le16_to_cpu(rf_list->rl_used) - index) *
 			 sizeof(struct ocfs2_refcount_rec));
 
-	mlog(0, "insert refcount record start %llu, len %u, count %u "
-	     "to leaf block %llu at index %d\n",
-	     (unsigned long long)le64_to_cpu(rec->r_cpos),
-	     le32_to_cpu(rec->r_clusters), le32_to_cpu(rec->r_refcount),
-	     (unsigned long long)ref_leaf_bh->b_blocknr, index);
+	trace_ocfs2_insert_refcount_rec(
+		(unsigned long long)ref_leaf_bh->b_blocknr, index,
+		(unsigned long long)le64_to_cpu(rec->r_cpos),
+		le32_to_cpu(rec->r_clusters), le32_to_cpu(rec->r_refcount));
 
 	rf_list->rl_recs[index] = *rec;
 
@@ -1850,10 +1849,12 @@ static int ocfs2_split_refcount_rec(handle_t *handle,
 
 	BUG_ON(le32_to_cpu(rb->rf_flags) & OCFS2_REFCOUNT_TREE_FL);
 
-	mlog(0, "original r_pos %llu, cluster %u, split %llu, cluster %u\n",
-	     le64_to_cpu(orig_rec->r_cpos), le32_to_cpu(orig_rec->r_clusters),
-	     le64_to_cpu(split_rec->r_cpos),
-	     le32_to_cpu(split_rec->r_clusters));
+	trace_ocfs2_split_refcount_rec(le64_to_cpu(orig_rec->r_cpos),
+		le32_to_cpu(orig_rec->r_clusters),
+		le32_to_cpu(orig_rec->r_refcount),
+		le64_to_cpu(split_rec->r_cpos),
+		le32_to_cpu(split_rec->r_clusters),
+		le32_to_cpu(split_rec->r_refcount));
 
 	/*
 	 * If we just need to split the header or tail clusters,
@@ -1967,12 +1968,11 @@ static int ocfs2_split_refcount_rec(handle_t *handle,
 
 	if (split_rec->r_refcount) {
 		rf_list->rl_recs[index] = *split_rec;
-		mlog(0, "insert refcount record start %llu, len %u, count %u "
-		     "to leaf block %llu at index %d\n",
-		     (unsigned long long)le64_to_cpu(split_rec->r_cpos),
-		     le32_to_cpu(split_rec->r_clusters),
-		     le32_to_cpu(split_rec->r_refcount),
-		     (unsigned long long)ref_leaf_bh->b_blocknr, index);
+		trace_ocfs2_split_refcount_rec_insert(
+			(unsigned long long)ref_leaf_bh->b_blocknr, index,
+			(unsigned long long)le64_to_cpu(split_rec->r_cpos),
+			le32_to_cpu(split_rec->r_clusters),
+			le32_to_cpu(split_rec->r_refcount));
 
 		if (merge)
 			ocfs2_refcount_rec_merge(rb, index);
@@ -1997,7 +1997,7 @@ static int __ocfs2_increase_refcount(handle_t *handle,
 	struct ocfs2_refcount_rec rec;
 	unsigned int set_len = 0;
 
-	mlog(0, "Tree owner %llu, add refcount start %llu, len %u\n",
+	trace_ocfs2_increase_refcount_begin(
 	     (unsigned long long)ocfs2_metadata_cache_owner(ci),
 	     (unsigned long long)cpos, len);
 
@@ -2024,9 +2024,9 @@ static int __ocfs2_increase_refcount(handle_t *handle,
 		 */
 		if (rec.r_refcount && le64_to_cpu(rec.r_cpos) == cpos &&
 		    set_len <= len) {
-			mlog(0, "increase refcount rec, start %llu, len %u, "
-			     "count %u\n", (unsigned long long)cpos, set_len,
-			     le32_to_cpu(rec.r_refcount));
+			trace_ocfs2_increase_refcount_change(
+				(unsigned long long)cpos, set_len,
+				le32_to_cpu(rec.r_refcount));
 			ret = ocfs2_change_refcount_rec(handle, ci,
 							ref_leaf_bh, index,
 							merge, 1);
@@ -2037,7 +2037,7 @@ static int __ocfs2_increase_refcount(handle_t *handle,
 		} else if (!rec.r_refcount) {
 			rec.r_refcount = cpu_to_le32(1);
 
-			mlog(0, "insert refcount rec, start %llu, len %u\n",
+			trace_ocfs2_increase_refcount_insert(
 			     (unsigned long long)le64_to_cpu(rec.r_cpos),
 			     set_len);
 			ret = ocfs2_insert_refcount_rec(handle, ci, ref_root_bh,
@@ -2055,8 +2055,7 @@ static int __ocfs2_increase_refcount(handle_t *handle,
 			rec.r_clusters = cpu_to_le32(set_len);
 			le32_add_cpu(&rec.r_refcount, 1);
 
-			mlog(0, "split refcount rec, start %llu, "
-			     "len %u, count %u\n",
+			trace_ocfs2_increase_refcount_split(
 			     (unsigned long long)le64_to_cpu(rec.r_cpos),
 			     set_len, le32_to_cpu(rec.r_refcount));
 			ret = ocfs2_split_refcount_rec(handle, ci,
@@ -2095,6 +2094,11 @@ static int ocfs2_remove_refcount_extent(handle_t *handle,
 
 	BUG_ON(rb->rf_records.rl_used);
 
+	trace_ocfs2_remove_refcount_extent(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)ref_leaf_bh->b_blocknr,
+		le32_to_cpu(rb->rf_cpos));
+
 	ocfs2_init_refcount_extent_tree(&et, ci, ref_root_bh);
 	ret = ocfs2_remove_extent(handle, &et, le32_to_cpu(rb->rf_cpos),
 				  1, meta_ac, dealloc);
@@ -2137,7 +2141,7 @@ static int ocfs2_remove_refcount_extent(handle_t *handle,
 	if (!rb->rf_list.l_next_free_rec) {
 		BUG_ON(rb->rf_clusters);
 
-		mlog(0, "reset refcount tree root %llu to be a record block.\n",
+		trace_ocfs2_restore_refcount_block(
 		     (unsigned long long)ref_root_bh->b_blocknr);
 
 		rb->rf_flags = 0;
@@ -2184,6 +2188,10 @@ static int ocfs2_decrease_refcount_rec(handle_t *handle,
 	BUG_ON(cpos + len >
 	       le64_to_cpu(rec->r_cpos) + le32_to_cpu(rec->r_clusters));
 
+	trace_ocfs2_decrease_refcount_rec(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)cpos, len);
+
 	if (cpos == le64_to_cpu(rec->r_cpos) &&
 	    len == le32_to_cpu(rec->r_clusters))
 		ret = ocfs2_change_refcount_rec(handle, ci,
@@ -2195,12 +2203,6 @@ static int ocfs2_decrease_refcount_rec(handle_t *handle,
 
 		le32_add_cpu(&split.r_refcount, -1);
 
-		mlog(0, "split refcount rec, start %llu, "
-		     "len %u, count %u, original start %llu, len %u\n",
-		     (unsigned long long)le64_to_cpu(split.r_cpos),
-		     len, le32_to_cpu(split.r_refcount),
-		     (unsigned long long)le64_to_cpu(rec->r_cpos),
-		     le32_to_cpu(rec->r_clusters));
 		ret = ocfs2_split_refcount_rec(handle, ci,
 					       ref_root_bh, ref_leaf_bh,
 					       &split, index, 1,
@@ -2239,10 +2241,9 @@ static int __ocfs2_decrease_refcount(handle_t *handle,
 	struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
 	struct buffer_head *ref_leaf_bh = NULL;
 
-	mlog(0, "Tree owner %llu, decrease refcount start %llu, "
-	     "len %u, delete %u\n",
-	     (unsigned long long)ocfs2_metadata_cache_owner(ci),
-	     (unsigned long long)cpos, len, delete);
+	trace_ocfs2_decrease_refcount(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)cpos, len, delete);
 
 	while (len) {
 		ret = ocfs2_get_refcount_rec(ci, ref_root_bh,
@@ -2352,8 +2353,8 @@ static int ocfs2_mark_extent_refcounted(struct inode *inode,
 {
 	int ret;
 
-	mlog(0, "Inode %lu refcount tree cpos %u, len %u, phys cluster %u\n",
-	     inode->i_ino, cpos, len, phys);
+	trace_ocfs2_mark_extent_refcounted(OCFS2_I(inode)->ip_blkno,
+					   cpos, len, phys);
 
 	if (!ocfs2_refcount_tree(OCFS2_SB(inode->i_sb))) {
 		ocfs2_error(inode->i_sb, "Inode %lu want to use refcount "
@@ -2392,8 +2393,6 @@ static int ocfs2_calc_refcount_meta_credits(struct super_block *sb,
 	struct buffer_head *ref_leaf_bh = NULL, *prev_bh = NULL;
 	u32 len;
 
-	mlog(0, "start_cpos %llu, clusters %u\n",
-	     (unsigned long long)start_cpos, clusters);
 	while (clusters) {
 		ret = ocfs2_get_refcount_rec(ci, ref_root_bh,
 					     cpos, clusters, &rec,
@@ -2427,12 +2426,11 @@ static int ocfs2_calc_refcount_meta_credits(struct super_block *sb,
 
 		rb = (struct ocfs2_refcount_block *)ref_leaf_bh->b_data;
 
-		mlog(0, "recs_add %d,cpos %llu, clusters %u, rec->r_cpos %llu,"
-		     "rec->r_clusters %u, rec->r_refcount %u, index %d\n",
-		     recs_add, (unsigned long long)cpos, clusters,
-		     (unsigned long long)le64_to_cpu(rec.r_cpos),
-		     le32_to_cpu(rec.r_clusters),
-		     le32_to_cpu(rec.r_refcount), index);
+		trace_ocfs2_calc_refcount_meta_credits_iterate(
+				recs_add, (unsigned long long)cpos, clusters,
+				(unsigned long long)le64_to_cpu(rec.r_cpos),
+				le32_to_cpu(rec.r_clusters),
+				le32_to_cpu(rec.r_refcount), index);
 
 		len = min((u64)cpos + clusters, le64_to_cpu(rec.r_cpos) +
 			  le32_to_cpu(rec.r_clusters)) - cpos;
@@ -2488,7 +2486,6 @@ static int ocfs2_calc_refcount_meta_credits(struct super_block *sb,
 	if (!ref_blocks)
 		goto out;
 
-	mlog(0, "we need ref_blocks %d\n", ref_blocks);
 	*meta_add += ref_blocks;
 	*credits += ref_blocks;
 
@@ -2514,6 +2511,10 @@ static int ocfs2_calc_refcount_meta_credits(struct super_block *sb,
 	}
 
 out:
+
+	trace_ocfs2_calc_refcount_meta_credits(
+		(unsigned long long)start_cpos, clusters,
+		*meta_add, *credits);
 	brelse(ref_leaf_bh);
 	brelse(prev_bh);
 	return ret;
@@ -2578,8 +2579,7 @@ int ocfs2_prepare_refcount_change_for_del(struct inode *inode,
 		goto out;
 	}
 
-	mlog(0, "reserve new metadata %d blocks, credits = %d\n",
-	     *ref_blocks, *credits);
+	trace_ocfs2_prepare_refcount_change_for_del(*ref_blocks, *credits);
 
 out:
 	brelse(ref_root_bh);
@@ -2886,8 +2886,7 @@ static int ocfs2_lock_refcount_allocators(struct super_block *sb,
 		goto out;
 	}
 
-	mlog(0, "reserve new metadata %d, clusters %u, credits = %d\n",
-	     meta_add, num_clusters, *credits);
+	trace_ocfs2_lock_refcount_allocators(meta_add, *credits);
 	ret = ocfs2_reserve_new_metadata_blocks(OCFS2_SB(sb), meta_add,
 						meta_ac);
 	if (ret) {
@@ -2937,8 +2936,8 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle,
 	loff_t offset, end, map_end;
 	struct address_space *mapping = context->inode->i_mapping;
 
-	mlog(0, "old_cluster %u, new %u, len %u at offset %u\n", old_cluster,
-	     new_cluster, new_len, cpos);
+	trace_ocfs2_duplicate_clusters_by_page(cpos, old_cluster,
+					       new_cluster, new_len);
 
 	readahead_pages =
 		(ocfs2_cow_contig_clusters(sb) <<
@@ -3031,8 +3030,8 @@ static int ocfs2_duplicate_clusters_by_jbd(handle_t *handle,
 	struct buffer_head *old_bh = NULL;
 	struct buffer_head *new_bh = NULL;
 
-	mlog(0, "old_cluster %u, new %u, len %u\n", old_cluster,
-	     new_cluster, new_len);
+	trace_ocfs2_duplicate_clusters_by_page(cpos, old_cluster,
+					       new_cluster, new_len);
 
 	for (i = 0; i < blocks; i++, old_block++, new_block++) {
 		new_bh = sb_getblk(osb->sb, new_block);
@@ -3085,8 +3084,8 @@ static int ocfs2_clear_ext_refcount(handle_t *handle,
 	struct super_block *sb = ocfs2_metadata_cache_get_super(et->et_ci);
 	u64 ino = ocfs2_metadata_cache_owner(et->et_ci);
 
-	mlog(0, "inode %llu cpos %u, len %u, p_cluster %u, ext_flags %u\n",
-	     (unsigned long long)ino, cpos, len, p_cluster, ext_flags);
+	trace_ocfs2_clear_ext_refcount((unsigned long long)ino,
+				       cpos, len, p_cluster, ext_flags);
 
 	memset(&replace_rec, 0, sizeof(replace_rec));
 	replace_rec.e_cpos = cpu_to_le32(cpos);
@@ -3141,8 +3140,8 @@ static int ocfs2_replace_clusters(handle_t *handle,
 	struct ocfs2_caching_info *ci = context->data_et.et_ci;
 	u64 ino = ocfs2_metadata_cache_owner(ci);
 
-	mlog(0, "inode %llu, cpos %u, old %u, new %u, len %u, ext_flags %u\n",
-	     (unsigned long long)ino, cpos, old, new, len, ext_flags);
+	trace_ocfs2_replace_clusters((unsigned long long)ino,
+				     cpos, old, new, len, ext_flags);
 
 	/*If the old clusters is unwritten, no need to duplicate. */
 	if (!(ext_flags & OCFS2_EXT_UNWRITTEN)) {
@@ -3228,7 +3227,7 @@ static int ocfs2_make_clusters_writable(struct super_block *sb,
 					u32 num_clusters, unsigned int e_flags)
 {
 	int ret, delete, index, credits =  0;
-	u32 new_bit, new_len;
+	u32 new_bit, new_len, orig_num_clusters;
 	unsigned int set_len;
 	struct ocfs2_super *osb = OCFS2_SB(sb);
 	handle_t *handle;
@@ -3236,8 +3235,8 @@ static int ocfs2_make_clusters_writable(struct super_block *sb,
 	struct ocfs2_caching_info *ref_ci = &context->ref_tree->rf_ci;
 	struct ocfs2_refcount_rec rec;
 
-	mlog(0, "cpos %u, p_cluster %u, num_clusters %u, e_flags %u\n",
-	     cpos, p_cluster, num_clusters, e_flags);
+	trace_ocfs2_make_clusters_writable(cpos, p_cluster,
+					   num_clusters, e_flags);
 
 	ret = ocfs2_lock_refcount_allocators(sb, p_cluster, num_clusters,
 					     &context->data_et,
@@ -3261,6 +3260,8 @@ static int ocfs2_make_clusters_writable(struct super_block *sb,
 		goto out;
 	}
 
+	orig_num_clusters = num_clusters;
+
 	while (num_clusters) {
 		ret = ocfs2_get_refcount_rec(ref_ci, context->ref_root_bh,
 					     p_cluster, num_clusters,
@@ -3348,7 +3349,8 @@ static int ocfs2_make_clusters_writable(struct super_block *sb,
 	 * in write-back mode.
 	 */
 	if (context->get_clusters == ocfs2_di_get_clusters) {
-		ret = ocfs2_cow_sync_writeback(sb, context, cpos, num_clusters);
+		ret = ocfs2_cow_sync_writeback(sb, context, cpos,
+					       orig_num_clusters);
 		if (ret)
 			mlog_errno(ret);
 	}
@@ -3472,9 +3474,9 @@ static int ocfs2_refcount_cow_hunk(struct inode *inode,
 		goto out;
 	}
 
-	mlog(0, "CoW inode %lu, cpos %u, write_len %u, cow_start %u, "
-	     "cow_len %u\n", inode->i_ino,
-	     cpos, write_len, cow_start, cow_len);
+	trace_ocfs2_refcount_cow_hunk(OCFS2_I(inode)->ip_blkno,
+				      cpos, write_len, max_cpos,
+				      cow_start, cow_len);
 
 	BUG_ON(cow_len == 0);
 
@@ -3753,8 +3755,7 @@ int ocfs2_add_refcount_flag(struct inode *inode,
 		goto out;
 	}
 
-	mlog(0, "reserve new metadata %d, credits = %d\n",
-	     ref_blocks, credits);
+	trace_ocfs2_add_refcount_flag(ref_blocks, credits);
 
 	if (ref_blocks) {
 		ret = ocfs2_reserve_new_metadata_blocks(OCFS2_SB(inode->i_sb),
@@ -4376,7 +4377,7 @@ static int ocfs2_user_path_parent(const char __user *path,
 	if (IS_ERR(s))
 		return PTR_ERR(s);
 
-	error = path_lookup(s, LOOKUP_PARENT, nd);
+	error = kern_path_parent(s, nd);
 	if (error)
 		putname(s);
 	else
diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c
index 3e78db361bc..41ffd36c689 100644
--- a/fs/ocfs2/reservations.c
+++ b/fs/ocfs2/reservations.c
@@ -30,10 +30,10 @@
 #include <linux/bitops.h>
 #include <linux/list.h>
 
-#define MLOG_MASK_PREFIX ML_RESERVATIONS
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
+#include "ocfs2_trace.h"
 
 #ifdef CONFIG_OCFS2_DEBUG_FS
 #define OCFS2_CHECK_RESERVATIONS
@@ -321,8 +321,7 @@ static void ocfs2_resv_insert(struct ocfs2_reservation_map *resmap,
 
 	assert_spin_locked(&resv_lock);
 
-	mlog(0, "Insert reservation start: %u len: %u\n", new->r_start,
-	     new->r_len);
+	trace_ocfs2_resv_insert(new->r_start, new->r_len);
 
 	while (*p) {
 		parent = *p;
@@ -423,8 +422,8 @@ static int ocfs2_resmap_find_free_bits(struct ocfs2_reservation_map *resmap,
 	unsigned int best_start, best_len = 0;
 	int offset, start, found;
 
-	mlog(0, "Find %u bits within range (%u, len %u) resmap len: %u\n",
-	     wanted, search_start, search_len, resmap->m_bitmap_len);
+	trace_ocfs2_resmap_find_free_bits_begin(search_start, search_len,
+						wanted, resmap->m_bitmap_len);
 
 	found = best_start = best_len = 0;
 
@@ -463,7 +462,7 @@ static int ocfs2_resmap_find_free_bits(struct ocfs2_reservation_map *resmap,
 	*rlen = best_len;
 	*rstart = best_start;
 
-	mlog(0, "Found start: %u len: %u\n", best_start, best_len);
+	trace_ocfs2_resmap_find_free_bits_end(best_start, best_len);
 
 	return *rlen;
 }
@@ -487,9 +486,8 @@ static void __ocfs2_resv_find_window(struct ocfs2_reservation_map *resmap,
 	 * - our window should be last in all reservations
 	 * - need to make sure we don't go past end of bitmap
 	 */
-
-	mlog(0, "resv start: %u resv end: %u goal: %u wanted: %u\n",
-	     resv->r_start, ocfs2_resv_end(resv), goal, wanted);
+	trace_ocfs2_resv_find_window_begin(resv->r_start, ocfs2_resv_end(resv),
+					   goal, wanted, RB_EMPTY_ROOT(root));
 
 	assert_spin_locked(&resv_lock);
 
@@ -498,9 +496,6 @@ static void __ocfs2_resv_find_window(struct ocfs2_reservation_map *resmap,
 		 * Easiest case - empty tree. We can just take
 		 * whatever window of free bits we want.
 		 */
-
-		mlog(0, "Empty root\n");
-
 		clen = ocfs2_resmap_find_free_bits(resmap, wanted, goal,
 						   resmap->m_bitmap_len - goal,
 						   &cstart, &clen);
@@ -524,8 +519,6 @@ static void __ocfs2_resv_find_window(struct ocfs2_reservation_map *resmap,
 	prev_resv = ocfs2_find_resv_lhs(resmap, goal);
 
 	if (prev_resv == NULL) {
-		mlog(0, "Goal on LHS of leftmost window\n");
-
 		/*
 		 * A NULL here means that the search code couldn't
 		 * find a window that starts before goal.
@@ -570,13 +563,15 @@ static void __ocfs2_resv_find_window(struct ocfs2_reservation_map *resmap,
 		next_resv = NULL;
 	}
 
+	trace_ocfs2_resv_find_window_prev(prev_resv->r_start,
+					  ocfs2_resv_end(prev_resv));
+
 	prev = &prev_resv->r_node;
 
 	/* Now we do a linear search for a window, starting at 'prev_rsv' */
 	while (1) {
 		next = rb_next(prev);
 		if (next) {
-			mlog(0, "One more resv found in linear search\n");
 			next_resv = rb_entry(next,
 					     struct ocfs2_alloc_reservation,
 					     r_node);
@@ -585,7 +580,6 @@ static void __ocfs2_resv_find_window(struct ocfs2_reservation_map *resmap,
 			gap_end = next_resv->r_start - 1;
 			gap_len = gap_end - gap_start + 1;
 		} else {
-			mlog(0, "No next node\n");
 			/*
 			 * We're at the rightmost edge of the
 			 * tree. See if a reservation between this
@@ -596,6 +590,8 @@ static void __ocfs2_resv_find_window(struct ocfs2_reservation_map *resmap,
 			gap_end = resmap->m_bitmap_len - 1;
 		}
 
+		trace_ocfs2_resv_find_window_next(next ? next_resv->r_start: -1,
+					next ? ocfs2_resv_end(next_resv) : -1);
 		/*
 		 * No need to check this gap if we have already found
 		 * a larger region of free bits.
@@ -654,8 +650,9 @@ static void ocfs2_cannibalize_resv(struct ocfs2_reservation_map *resmap,
 	lru_resv = list_first_entry(&resmap->m_lru,
 				    struct ocfs2_alloc_reservation, r_lru);
 
-	mlog(0, "lru resv: start: %u len: %u end: %u\n", lru_resv->r_start,
-	     lru_resv->r_len, ocfs2_resv_end(lru_resv));
+	trace_ocfs2_cannibalize_resv_begin(lru_resv->r_start,
+					   lru_resv->r_len,
+					   ocfs2_resv_end(lru_resv));
 
 	/*
 	 * Cannibalize (some or all) of the target reservation and
@@ -684,10 +681,9 @@ static void ocfs2_cannibalize_resv(struct ocfs2_reservation_map *resmap,
 		resv->r_len = shrink;
 	}
 
-	mlog(0, "Reservation now looks like: r_start: %u r_end: %u "
-	     "r_len: %u r_last_start: %u r_last_len: %u\n",
-	     resv->r_start, ocfs2_resv_end(resv), resv->r_len,
-	     resv->r_last_start, resv->r_last_len);
+	trace_ocfs2_cannibalize_resv_end(resv->r_start, ocfs2_resv_end(resv),
+					 resv->r_len, resv->r_last_start,
+					 resv->r_last_len);
 
 	ocfs2_resv_insert(resmap, resv);
 }
@@ -748,7 +744,6 @@ int ocfs2_resmap_resv_bits(struct ocfs2_reservation_map *resmap,
 		if ((resv->r_flags & OCFS2_RESV_FLAG_TMP) || wanted < *clen)
 			wanted = *clen;
 
-		mlog(0, "empty reservation, find new window\n");
 		/*
 		 * Try to get a window here. If it works, we must fall
 		 * through and test the bitmap . This avoids some
@@ -757,6 +752,7 @@ int ocfs2_resmap_resv_bits(struct ocfs2_reservation_map *resmap,
 		 * that inode.
 		 */
 		ocfs2_resv_find_window(resmap, resv, wanted);
+		trace_ocfs2_resmap_resv_bits(resv->r_start, resv->r_len);
 	}
 
 	BUG_ON(ocfs2_resv_empty(resv));
@@ -813,10 +809,10 @@ void ocfs2_resmap_claimed_bits(struct ocfs2_reservation_map *resmap,
 
 	spin_lock(&resv_lock);
 
-	mlog(0, "claim bits: cstart: %u cend: %u clen: %u r_start: %u "
-	     "r_end: %u r_len: %u, r_last_start: %u r_last_len: %u\n",
-	     cstart, cend, clen, resv->r_start, ocfs2_resv_end(resv),
-	     resv->r_len, resv->r_last_start, resv->r_last_len);
+	trace_ocfs2_resmap_claimed_bits_begin(cstart, cend, clen, resv->r_start,
+					      ocfs2_resv_end(resv), resv->r_len,
+					      resv->r_last_start,
+					      resv->r_last_len);
 
 	BUG_ON(cstart < resv->r_start);
 	BUG_ON(cstart > ocfs2_resv_end(resv));
@@ -833,10 +829,9 @@ void ocfs2_resmap_claimed_bits(struct ocfs2_reservation_map *resmap,
 	if (!ocfs2_resv_empty(resv))
 		ocfs2_resv_mark_lru(resmap, resv);
 
-	mlog(0, "Reservation now looks like: r_start: %u r_end: %u "
-	     "r_len: %u r_last_start: %u r_last_len: %u\n",
-	     resv->r_start, ocfs2_resv_end(resv), resv->r_len,
-	     resv->r_last_start, resv->r_last_len);
+	trace_ocfs2_resmap_claimed_bits_end(resv->r_start, ocfs2_resv_end(resv),
+					    resv->r_len, resv->r_last_start,
+					    resv->r_last_len);
 
 	ocfs2_check_resmap(resmap);
 
diff --git a/fs/ocfs2/resize.c b/fs/ocfs2/resize.c
index dacd553d861..ec55add7604 100644
--- a/fs/ocfs2/resize.c
+++ b/fs/ocfs2/resize.c
@@ -27,7 +27,6 @@
 #include <linux/fs.h>
 #include <linux/types.h>
 
-#define MLOG_MASK_PREFIX ML_DISK_ALLOC
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -39,6 +38,7 @@
 #include "super.h"
 #include "sysfile.h"
 #include "uptodate.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 #include "suballoc.h"
@@ -82,7 +82,6 @@ static u16 ocfs2_calc_new_backup_super(struct inode *inode,
 		backups++;
 	}
 
-	mlog_exit_void();
 	return backups;
 }
 
@@ -103,8 +102,8 @@ static int ocfs2_update_last_group_and_inode(handle_t *handle,
 	u16 cl_bpc = le16_to_cpu(cl->cl_bpc);
 	u16 cl_cpg = le16_to_cpu(cl->cl_cpg);
 
-	mlog_entry("(new_clusters=%d, first_new_cluster = %u)\n",
-		   new_clusters, first_new_cluster);
+	trace_ocfs2_update_last_group_and_inode(new_clusters,
+						first_new_cluster);
 
 	ret = ocfs2_journal_access_gd(handle, INODE_CACHE(bm_inode),
 				      group_bh, OCFS2_JOURNAL_ACCESS_WRITE);
@@ -176,7 +175,8 @@ out_rollback:
 		le16_add_cpu(&group->bg_free_bits_count, -1 * num_bits);
 	}
 out:
-	mlog_exit(ret);
+	if (ret)
+		mlog_errno(ret);
 	return ret;
 }
 
@@ -281,8 +281,6 @@ int ocfs2_group_extend(struct inode * inode, int new_clusters)
 	u32 first_new_cluster;
 	u64 lgd_blkno;
 
-	mlog_entry_void();
-
 	if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb))
 		return -EROFS;
 
@@ -342,7 +340,8 @@ int ocfs2_group_extend(struct inode * inode, int new_clusters)
 		goto out_unlock;
 	}
 
-	mlog(0, "extend the last group at %llu, new clusters = %d\n",
+
+	trace_ocfs2_group_extend(
 	     (unsigned long long)le64_to_cpu(group->bg_blkno), new_clusters);
 
 	handle = ocfs2_start_trans(osb, OCFS2_GROUP_EXTEND_CREDITS);
@@ -377,7 +376,6 @@ out_mutex:
 	iput(main_bm_inode);
 
 out:
-	mlog_exit_void();
 	return ret;
 }
 
@@ -472,8 +470,6 @@ int ocfs2_group_add(struct inode *inode, struct ocfs2_new_group_input *input)
 	struct ocfs2_chain_rec *cr;
 	u16 cl_bpc;
 
-	mlog_entry_void();
-
 	if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb))
 		return -EROFS;
 
@@ -520,8 +516,8 @@ int ocfs2_group_add(struct inode *inode, struct ocfs2_new_group_input *input)
 		goto out_unlock;
 	}
 
-	mlog(0, "Add a new group  %llu in chain = %u, length = %u\n",
-	     (unsigned long long)input->group, input->chain, input->clusters);
+	trace_ocfs2_group_add((unsigned long long)input->group,
+			       input->chain, input->clusters, input->frees);
 
 	handle = ocfs2_start_trans(osb, OCFS2_GROUP_ADD_CREDITS);
 	if (IS_ERR(handle)) {
@@ -589,6 +585,5 @@ out_mutex:
 	iput(main_bm_inode);
 
 out:
-	mlog_exit_void();
 	return ret;
 }
diff --git a/fs/ocfs2/slot_map.c b/fs/ocfs2/slot_map.c
index ab4e0172cc1..26fc0014d50 100644
--- a/fs/ocfs2/slot_map.c
+++ b/fs/ocfs2/slot_map.c
@@ -27,7 +27,6 @@
 #include <linux/slab.h>
 #include <linux/highmem.h>
 
-#define MLOG_MASK_PREFIX ML_SUPER
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -39,6 +38,7 @@
 #include "slot_map.h"
 #include "super.h"
 #include "sysfile.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -142,8 +142,7 @@ int ocfs2_refresh_slot_info(struct ocfs2_super *osb)
 	BUG_ON(si->si_blocks == 0);
 	BUG_ON(si->si_bh == NULL);
 
-	mlog(0, "Refreshing slot map, reading %u block(s)\n",
-	     si->si_blocks);
+	trace_ocfs2_refresh_slot_info(si->si_blocks);
 
 	/*
 	 * We pass -1 as blocknr because we expect all of si->si_bh to
@@ -381,8 +380,7 @@ static int ocfs2_map_slot_buffers(struct ocfs2_super *osb,
 	/* The size checks above should ensure this */
 	BUG_ON((osb->max_slots / si->si_slots_per_block) > blocks);
 
-	mlog(0, "Slot map needs %u buffers for %llu bytes\n",
-	     si->si_blocks, bytes);
+	trace_ocfs2_map_slot_buffers(bytes, si->si_blocks);
 
 	si->si_bh = kzalloc(sizeof(struct buffer_head *) * si->si_blocks,
 			    GFP_KERNEL);
@@ -400,8 +398,7 @@ static int ocfs2_map_slot_buffers(struct ocfs2_super *osb,
 			goto bail;
 		}
 
-		mlog(0, "Reading slot map block %u at %llu\n", i,
-		     (unsigned long long)blkno);
+		trace_ocfs2_map_slot_buffers_block((unsigned long long)blkno, i);
 
 		bh = NULL;  /* Acquire a fresh bh */
 		status = ocfs2_read_blocks(INODE_CACHE(si->si_inode), blkno,
@@ -475,8 +472,6 @@ int ocfs2_find_slot(struct ocfs2_super *osb)
 	int slot;
 	struct ocfs2_slot_info *si;
 
-	mlog_entry_void();
-
 	si = osb->slot_info;
 
 	spin_lock(&osb->osb_lock);
@@ -505,14 +500,13 @@ int ocfs2_find_slot(struct ocfs2_super *osb)
 	osb->slot_num = slot;
 	spin_unlock(&osb->osb_lock);
 
-	mlog(0, "taking node slot %d\n", osb->slot_num);
+	trace_ocfs2_find_slot(osb->slot_num);
 
 	status = ocfs2_update_disk_slot(osb, si, osb->slot_num);
 	if (status < 0)
 		mlog_errno(status);
 
 bail:
-	mlog_exit(status);
 	return status;
 }
 
diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index 71998d4d61d..ab6e2061074 100644
--- a/fs/ocfs2/suballoc.c
+++ b/fs/ocfs2/suballoc.c
@@ -29,7 +29,6 @@
 #include <linux/slab.h>
 #include <linux/highmem.h>
 
-#define MLOG_MASK_PREFIX ML_DISK_ALLOC
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -44,6 +43,7 @@
 #include "super.h"
 #include "sysfile.h"
 #include "uptodate.h"
+#include "ocfs2_trace.h"
 
 #include "buffer_head_io.h"
 
@@ -308,8 +308,8 @@ static int ocfs2_validate_group_descriptor(struct super_block *sb,
 	int rc;
 	struct ocfs2_group_desc *gd = (struct ocfs2_group_desc *)bh->b_data;
 
-	mlog(0, "Validating group descriptor %llu\n",
-	     (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_validate_group_descriptor(
+					(unsigned long long)bh->b_blocknr);
 
 	BUG_ON(!buffer_uptodate(bh));
 
@@ -389,8 +389,6 @@ static int ocfs2_block_group_fill(handle_t *handle,
 	struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) bg_bh->b_data;
 	struct super_block * sb = alloc_inode->i_sb;
 
-	mlog_entry_void();
-
 	if (((unsigned long long) bg_bh->b_blocknr) != group_blkno) {
 		ocfs2_error(alloc_inode->i_sb, "group block (%llu) != "
 			    "b_blocknr (%llu)",
@@ -436,7 +434,8 @@ static int ocfs2_block_group_fill(handle_t *handle,
 	 * allocation time. */
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -477,8 +476,8 @@ ocfs2_block_group_alloc_contig(struct ocfs2_super *osb, handle_t *handle,
 
 	/* setup the group */
 	bg_blkno = ocfs2_clusters_to_blocks(osb->sb, bit_off);
-	mlog(0, "new descriptor, record %u, at block %llu\n",
-	     alloc_rec, (unsigned long long)bg_blkno);
+	trace_ocfs2_block_group_alloc_contig(
+	     (unsigned long long)bg_blkno, alloc_rec);
 
 	bg_bh = sb_getblk(osb->sb, bg_blkno);
 	if (!bg_bh) {
@@ -657,8 +656,8 @@ ocfs2_block_group_alloc_discontig(handle_t *handle,
 
 	/* setup the group */
 	bg_blkno = ocfs2_clusters_to_blocks(osb->sb, bit_off);
-	mlog(0, "new descriptor, record %u, at block %llu\n",
-	     alloc_rec, (unsigned long long)bg_blkno);
+	trace_ocfs2_block_group_alloc_discontig(
+				(unsigned long long)bg_blkno, alloc_rec);
 
 	bg_bh = sb_getblk(osb->sb, bg_blkno);
 	if (!bg_bh) {
@@ -707,8 +706,6 @@ static int ocfs2_block_group_alloc(struct ocfs2_super *osb,
 
 	BUG_ON(ocfs2_is_cluster_bitmap(alloc_inode));
 
-	mlog_entry_void();
-
 	cl = &fe->id2.i_chain;
 	status = ocfs2_reserve_clusters_with_limit(osb,
 						   le16_to_cpu(cl->cl_cpg),
@@ -730,8 +727,8 @@ static int ocfs2_block_group_alloc(struct ocfs2_super *osb,
 	}
 
 	if (last_alloc_group && *last_alloc_group != 0) {
-		mlog(0, "use old allocation group %llu for block group alloc\n",
-		     (unsigned long long)*last_alloc_group);
+		trace_ocfs2_block_group_alloc(
+				(unsigned long long)*last_alloc_group);
 		ac->ac_last_group = *last_alloc_group;
 	}
 
@@ -796,7 +793,8 @@ bail:
 
 	brelse(bg_bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -814,8 +812,6 @@ static int ocfs2_reserve_suballoc_bits(struct ocfs2_super *osb,
 	struct ocfs2_dinode *fe;
 	u32 free_bits;
 
-	mlog_entry_void();
-
 	alloc_inode = ocfs2_get_system_file_inode(osb, type, slot);
 	if (!alloc_inode) {
 		mlog_errno(-EINVAL);
@@ -855,16 +851,15 @@ static int ocfs2_reserve_suballoc_bits(struct ocfs2_super *osb,
 	if (bits_wanted > free_bits) {
 		/* cluster bitmap never grows */
 		if (ocfs2_is_cluster_bitmap(alloc_inode)) {
-			mlog(0, "Disk Full: wanted=%u, free_bits=%u\n",
-			     bits_wanted, free_bits);
+			trace_ocfs2_reserve_suballoc_bits_nospc(bits_wanted,
+								free_bits);
 			status = -ENOSPC;
 			goto bail;
 		}
 
 		if (!(flags & ALLOC_NEW_GROUP)) {
-			mlog(0, "Alloc File %u Full: wanted=%u, free_bits=%u, "
-			     "and we don't alloc a new group for it.\n",
-			     slot, bits_wanted, free_bits);
+			trace_ocfs2_reserve_suballoc_bits_no_new_group(
+						slot, bits_wanted, free_bits);
 			status = -ENOSPC;
 			goto bail;
 		}
@@ -890,7 +885,8 @@ static int ocfs2_reserve_suballoc_bits(struct ocfs2_super *osb,
 bail:
 	brelse(bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1052,7 +1048,8 @@ bail:
 		*ac = NULL;
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1119,8 +1116,8 @@ int ocfs2_reserve_new_inode(struct ocfs2_super *osb,
 		spin_lock(&osb->osb_lock);
 		osb->osb_inode_alloc_group = alloc_group;
 		spin_unlock(&osb->osb_lock);
-		mlog(0, "after reservation, new allocation group is "
-		     "%llu\n", (unsigned long long)alloc_group);
+		trace_ocfs2_reserve_new_inode_new_group(
+			(unsigned long long)alloc_group);
 
 		/*
 		 * Some inodes must be freed by us, so try to allocate
@@ -1152,7 +1149,8 @@ bail:
 		*ac = NULL;
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1189,8 +1187,6 @@ static int ocfs2_reserve_clusters_with_limit(struct ocfs2_super *osb,
 {
 	int status;
 
-	mlog_entry_void();
-
 	*ac = kzalloc(sizeof(struct ocfs2_alloc_context), GFP_KERNEL);
 	if (!(*ac)) {
 		status = -ENOMEM;
@@ -1229,7 +1225,8 @@ bail:
 		*ac = NULL;
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1357,15 +1354,12 @@ static inline int ocfs2_block_group_set_bits(handle_t *handle,
 	void *bitmap = bg->bg_bitmap;
 	int journal_type = OCFS2_JOURNAL_ACCESS_WRITE;
 
-	mlog_entry_void();
-
 	/* All callers get the descriptor via
 	 * ocfs2_read_group_descriptor().  Any corruption is a code bug. */
 	BUG_ON(!OCFS2_IS_VALID_GROUP_DESC(bg));
 	BUG_ON(le16_to_cpu(bg->bg_free_bits_count) < num_bits);
 
-	mlog(0, "block_group_set_bits: off = %u, num = %u\n", bit_off,
-	     num_bits);
+	trace_ocfs2_block_group_set_bits(bit_off, num_bits);
 
 	if (ocfs2_is_cluster_bitmap(alloc_inode))
 		journal_type = OCFS2_JOURNAL_ACCESS_UNDO;
@@ -1394,7 +1388,8 @@ static inline int ocfs2_block_group_set_bits(handle_t *handle,
 	ocfs2_journal_dirty(handle, group_bh);
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1437,10 +1432,10 @@ static int ocfs2_relink_block_group(handle_t *handle,
 	BUG_ON(!OCFS2_IS_VALID_GROUP_DESC(bg));
 	BUG_ON(!OCFS2_IS_VALID_GROUP_DESC(prev_bg));
 
-	mlog(0, "Suballoc %llu, chain %u, move group %llu to top, prev = %llu\n",
-	     (unsigned long long)le64_to_cpu(fe->i_blkno), chain,
-	     (unsigned long long)le64_to_cpu(bg->bg_blkno),
-	     (unsigned long long)le64_to_cpu(prev_bg->bg_blkno));
+	trace_ocfs2_relink_block_group(
+		(unsigned long long)le64_to_cpu(fe->i_blkno), chain,
+		(unsigned long long)le64_to_cpu(bg->bg_blkno),
+		(unsigned long long)le64_to_cpu(prev_bg->bg_blkno));
 
 	fe_ptr = le64_to_cpu(fe->id2.i_chain.cl_recs[chain].c_blkno);
 	bg_ptr = le64_to_cpu(bg->bg_next_group);
@@ -1484,7 +1479,8 @@ out_rollback:
 		prev_bg->bg_next_group = cpu_to_le64(prev_bg_ptr);
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1525,10 +1521,10 @@ static int ocfs2_cluster_group_search(struct inode *inode,
 		if ((gd_cluster_off + max_bits) >
 		    OCFS2_I(inode)->ip_clusters) {
 			max_bits = OCFS2_I(inode)->ip_clusters - gd_cluster_off;
-			mlog(0, "Desc %llu, bg_bits %u, clusters %u, use %u\n",
-			     (unsigned long long)le64_to_cpu(gd->bg_blkno),
-			     le16_to_cpu(gd->bg_bits),
-			     OCFS2_I(inode)->ip_clusters, max_bits);
+			trace_ocfs2_cluster_group_search_wrong_max_bits(
+				(unsigned long long)le64_to_cpu(gd->bg_blkno),
+				le16_to_cpu(gd->bg_bits),
+				OCFS2_I(inode)->ip_clusters, max_bits);
 		}
 
 		ret = ocfs2_block_group_find_clear_bits(OCFS2_SB(inode->i_sb),
@@ -1542,9 +1538,9 @@ static int ocfs2_cluster_group_search(struct inode *inode,
 							  gd_cluster_off +
 							  res->sr_bit_offset +
 							  res->sr_bits);
-			mlog(0, "Checking %llu against %llu\n",
-			     (unsigned long long)blkoff,
-			     (unsigned long long)max_block);
+			trace_ocfs2_cluster_group_search_max_block(
+				(unsigned long long)blkoff,
+				(unsigned long long)max_block);
 			if (blkoff > max_block)
 				return -ENOSPC;
 		}
@@ -1588,9 +1584,9 @@ static int ocfs2_block_group_search(struct inode *inode,
 		if (!ret && max_block) {
 			blkoff = le64_to_cpu(bg->bg_blkno) +
 				res->sr_bit_offset + res->sr_bits;
-			mlog(0, "Checking %llu against %llu\n",
-			     (unsigned long long)blkoff,
-			     (unsigned long long)max_block);
+			trace_ocfs2_block_group_search_max_block(
+				(unsigned long long)blkoff,
+				(unsigned long long)max_block);
 			if (blkoff > max_block)
 				ret = -ENOSPC;
 		}
@@ -1756,9 +1752,9 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac,
 	struct ocfs2_group_desc *bg;
 
 	chain = ac->ac_chain;
-	mlog(0, "trying to alloc %u bits from chain %u, inode %llu\n",
-	     bits_wanted, chain,
-	     (unsigned long long)OCFS2_I(alloc_inode)->ip_blkno);
+	trace_ocfs2_search_chain_begin(
+		(unsigned long long)OCFS2_I(alloc_inode)->ip_blkno,
+		bits_wanted, chain);
 
 	status = ocfs2_read_group_descriptor(alloc_inode, fe,
 					     le64_to_cpu(cl->cl_recs[chain].c_blkno),
@@ -1799,8 +1795,8 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac,
 		goto bail;
 	}
 
-	mlog(0, "alloc succeeds: we give %u bits from block group %llu\n",
-	     res->sr_bits, (unsigned long long)le64_to_cpu(bg->bg_blkno));
+	trace_ocfs2_search_chain_succ(
+		(unsigned long long)le64_to_cpu(bg->bg_blkno), res->sr_bits);
 
 	res->sr_bg_blkno = le64_to_cpu(bg->bg_blkno);
 
@@ -1861,8 +1857,9 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac,
 		goto bail;
 	}
 
-	mlog(0, "Allocated %u bits from suballocator %llu\n", res->sr_bits,
-	     (unsigned long long)le64_to_cpu(fe->i_blkno));
+	trace_ocfs2_search_chain_end(
+			(unsigned long long)le64_to_cpu(fe->i_blkno),
+			res->sr_bits);
 
 out_loc_only:
 	*bits_left = le16_to_cpu(bg->bg_free_bits_count);
@@ -1870,7 +1867,8 @@ bail:
 	brelse(group_bh);
 	brelse(prev_group_bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1888,8 +1886,6 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac,
 	struct ocfs2_chain_list *cl;
 	struct ocfs2_dinode *fe;
 
-	mlog_entry_void();
-
 	BUG_ON(ac->ac_bits_given >= ac->ac_bits_wanted);
 	BUG_ON(bits_wanted > (ac->ac_bits_wanted - ac->ac_bits_given));
 	BUG_ON(!ac->ac_bh);
@@ -1945,8 +1941,7 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac,
 		goto bail;
 	}
 
-	mlog(0, "Search of victim chain %u came up with nothing, "
-	     "trying all chains now.\n", victim);
+	trace_ocfs2_claim_suballoc_bits(victim);
 
 	/* If we didn't pick a good victim, then just default to
 	 * searching each chain in order. Don't allow chain relinking
@@ -1984,7 +1979,8 @@ set_hint:
 	}
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2021,7 +2017,8 @@ int ocfs2_claim_metadata(handle_t *handle,
 	*num_bits = res.sr_bits;
 	status = 0;
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2172,8 +2169,8 @@ int ocfs2_claim_new_inode_at_loc(handle_t *handle,
 		goto out;
 	}
 
-	mlog(0, "Allocated %u bits from suballocator %llu\n", res->sr_bits,
-	     (unsigned long long)di_blkno);
+	trace_ocfs2_claim_new_inode_at_loc((unsigned long long)di_blkno,
+					   res->sr_bits);
 
 	atomic_inc(&OCFS2_SB(ac->ac_inode->i_sb)->alloc_stats.bg_allocs);
 
@@ -2201,8 +2198,6 @@ int ocfs2_claim_new_inode(handle_t *handle,
 	int status;
 	struct ocfs2_suballoc_result res;
 
-	mlog_entry_void();
-
 	BUG_ON(!ac);
 	BUG_ON(ac->ac_bits_given != 0);
 	BUG_ON(ac->ac_bits_wanted != 1);
@@ -2230,7 +2225,8 @@ int ocfs2_claim_new_inode(handle_t *handle,
 	ocfs2_save_inode_ac_group(dir, ac);
 	status = 0;
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2307,8 +2303,6 @@ int __ocfs2_claim_clusters(handle_t *handle,
 	struct ocfs2_suballoc_result res = { .sr_blkno = 0, };
 	struct ocfs2_super *osb = OCFS2_SB(ac->ac_inode->i_sb);
 
-	mlog_entry_void();
-
 	BUG_ON(ac->ac_bits_given >= ac->ac_bits_wanted);
 
 	BUG_ON(ac->ac_which != OCFS2_AC_USE_LOCAL
@@ -2363,7 +2357,8 @@ int __ocfs2_claim_clusters(handle_t *handle,
 	ac->ac_bits_given += *num_clusters;
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2392,13 +2387,11 @@ static int ocfs2_block_group_clear_bits(handle_t *handle,
 	unsigned int tmp;
 	struct ocfs2_group_desc *undo_bg = NULL;
 
-	mlog_entry_void();
-
 	/* The caller got this descriptor from
 	 * ocfs2_read_group_descriptor().  Any corruption is a code bug. */
 	BUG_ON(!OCFS2_IS_VALID_GROUP_DESC(bg));
 
-	mlog(0, "off = %u, num = %u\n", bit_off, num_bits);
+	trace_ocfs2_block_group_clear_bits(bit_off, num_bits);
 
 	BUG_ON(undo_fn && !ocfs2_is_cluster_bitmap(alloc_inode));
 	status = ocfs2_journal_access_gd(handle, INODE_CACHE(alloc_inode),
@@ -2463,8 +2456,6 @@ static int _ocfs2_free_suballoc_bits(handle_t *handle,
 	struct buffer_head *group_bh = NULL;
 	struct ocfs2_group_desc *group;
 
-	mlog_entry_void();
-
 	/* The alloc_bh comes from ocfs2_free_dinode() or
 	 * ocfs2_free_clusters().  The callers have all locked the
 	 * allocator and gotten alloc_bh from the lock call.  This
@@ -2473,9 +2464,10 @@ static int _ocfs2_free_suballoc_bits(handle_t *handle,
 	BUG_ON(!OCFS2_IS_VALID_DINODE(fe));
 	BUG_ON((count + start_bit) > ocfs2_bits_per_group(cl));
 
-	mlog(0, "%llu: freeing %u bits from group %llu, starting at %u\n",
-	     (unsigned long long)OCFS2_I(alloc_inode)->ip_blkno, count,
-	     (unsigned long long)bg_blkno, start_bit);
+	trace_ocfs2_free_suballoc_bits(
+		(unsigned long long)OCFS2_I(alloc_inode)->ip_blkno,
+		(unsigned long long)bg_blkno,
+		start_bit, count);
 
 	status = ocfs2_read_group_descriptor(alloc_inode, fe, bg_blkno,
 					     &group_bh);
@@ -2511,7 +2503,8 @@ static int _ocfs2_free_suballoc_bits(handle_t *handle,
 bail:
 	brelse(group_bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2556,11 +2549,8 @@ static int _ocfs2_free_clusters(handle_t *handle,
 
 	/* You can't ever have a contiguous set of clusters
 	 * bigger than a block group bitmap so we never have to worry
-	 * about looping on them. */
-
-	mlog_entry_void();
-
-	/* This is expensive. We can safely remove once this stuff has
+	 * about looping on them.
+	 * This is expensive. We can safely remove once this stuff has
 	 * gotten tested really well. */
 	BUG_ON(start_blk != ocfs2_clusters_to_blocks(bitmap_inode->i_sb, ocfs2_blocks_to_clusters(bitmap_inode->i_sb, start_blk)));
 
@@ -2569,10 +2559,9 @@ static int _ocfs2_free_clusters(handle_t *handle,
 	ocfs2_block_to_cluster_group(bitmap_inode, start_blk, &bg_blkno,
 				     &bg_start_bit);
 
-	mlog(0, "want to free %u clusters starting at block %llu\n",
-	     num_clusters, (unsigned long long)start_blk);
-	mlog(0, "bg_blkno = %llu, bg_start_bit = %u\n",
-	     (unsigned long long)bg_blkno, bg_start_bit);
+	trace_ocfs2_free_clusters((unsigned long long)bg_blkno,
+			(unsigned long long)start_blk,
+			bg_start_bit, num_clusters);
 
 	status = _ocfs2_free_suballoc_bits(handle, bitmap_inode, bitmap_bh,
 					   bg_start_bit, bg_blkno,
@@ -2586,7 +2575,8 @@ static int _ocfs2_free_clusters(handle_t *handle,
 					 num_clusters);
 
 out:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2756,7 +2746,7 @@ static int ocfs2_get_suballoc_slot_bit(struct ocfs2_super *osb, u64 blkno,
 	struct buffer_head *inode_bh = NULL;
 	struct ocfs2_dinode *inode_fe;
 
-	mlog_entry("blkno: %llu\n", (unsigned long long)blkno);
+	trace_ocfs2_get_suballoc_slot_bit((unsigned long long)blkno);
 
 	/* dirty read disk */
 	status = ocfs2_read_blocks_sync(osb, blkno, 1, &inode_bh);
@@ -2793,7 +2783,8 @@ static int ocfs2_get_suballoc_slot_bit(struct ocfs2_super *osb, u64 blkno,
 bail:
 	brelse(inode_bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2816,8 +2807,8 @@ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb,
 	u64 bg_blkno;
 	int status;
 
-	mlog_entry("blkno: %llu bit: %u\n", (unsigned long long)blkno,
-		   (unsigned int)bit);
+	trace_ocfs2_test_suballoc_bit((unsigned long long)blkno,
+				      (unsigned int)bit);
 
 	alloc_di = (struct ocfs2_dinode *)alloc_bh->b_data;
 	if ((bit + 1) > ocfs2_bits_per_group(&alloc_di->id2.i_chain)) {
@@ -2844,7 +2835,8 @@ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb,
 bail:
 	brelse(group_bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2869,7 +2861,7 @@ int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res)
 	struct inode *inode_alloc_inode;
 	struct buffer_head *alloc_bh = NULL;
 
-	mlog_entry("blkno: %llu", (unsigned long long)blkno);
+	trace_ocfs2_test_inode_bit((unsigned long long)blkno);
 
 	status = ocfs2_get_suballoc_slot_bit(osb, blkno, &suballoc_slot,
 					     &group_blkno, &suballoc_bit);
@@ -2910,6 +2902,7 @@ int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res)
 	iput(inode_alloc_inode);
 	brelse(alloc_bh);
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 38f986d2447..0e4083987d2 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -42,7 +42,9 @@
 #include <linux/seq_file.h>
 #include <linux/quotaops.h>
 
-#define MLOG_MASK_PREFIX ML_SUPER
+#define CREATE_TRACE_POINTS
+#include "ocfs2_trace.h"
+
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -441,8 +443,6 @@ static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb)
 	int status = 0;
 	int i;
 
-	mlog_entry_void();
-
 	new = ocfs2_iget(osb, osb->root_blkno, OCFS2_FI_FLAG_SYSFILE, 0);
 	if (IS_ERR(new)) {
 		status = PTR_ERR(new);
@@ -478,7 +478,8 @@ static int ocfs2_init_global_system_inodes(struct ocfs2_super *osb)
 	}
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -488,8 +489,6 @@ static int ocfs2_init_local_system_inodes(struct ocfs2_super *osb)
 	int status = 0;
 	int i;
 
-	mlog_entry_void();
-
 	for (i = OCFS2_LAST_GLOBAL_SYSTEM_INODE + 1;
 	     i < NUM_SYSTEM_INODES;
 	     i++) {
@@ -508,7 +507,8 @@ static int ocfs2_init_local_system_inodes(struct ocfs2_super *osb)
 	}
 
 bail:
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -517,8 +517,6 @@ static void ocfs2_release_system_inodes(struct ocfs2_super *osb)
 	int i;
 	struct inode *inode;
 
-	mlog_entry_void();
-
 	for (i = 0; i < NUM_GLOBAL_SYSTEM_INODES; i++) {
 		inode = osb->global_system_inodes[i];
 		if (inode) {
@@ -540,7 +538,7 @@ static void ocfs2_release_system_inodes(struct ocfs2_super *osb)
 	}
 
 	if (!osb->local_system_inodes)
-		goto out;
+		return;
 
 	for (i = 0; i < NUM_LOCAL_SYSTEM_INODES * osb->max_slots; i++) {
 		if (osb->local_system_inodes[i]) {
@@ -551,9 +549,6 @@ static void ocfs2_release_system_inodes(struct ocfs2_super *osb)
 
 	kfree(osb->local_system_inodes);
 	osb->local_system_inodes = NULL;
-
-out:
-	mlog_exit(0);
 }
 
 /* We're allocating fs objects, use GFP_NOFS */
@@ -684,12 +679,9 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
 		}
 
 		if (*flags & MS_RDONLY) {
-			mlog(0, "Going to ro mode.\n");
 			sb->s_flags |= MS_RDONLY;
 			osb->osb_flags |= OCFS2_OSB_SOFT_RO;
 		} else {
-			mlog(0, "Making ro filesystem writeable.\n");
-
 			if (osb->osb_flags & OCFS2_OSB_ERROR_FS) {
 				mlog(ML_ERROR, "Cannot remount RDWR "
 				     "filesystem due to previous errors.\n");
@@ -707,6 +699,7 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
 			sb->s_flags &= ~MS_RDONLY;
 			osb->osb_flags &= ~OCFS2_OSB_SOFT_RO;
 		}
+		trace_ocfs2_remount(sb->s_flags, osb->osb_flags, *flags);
 unlock_osb:
 		spin_unlock(&osb->osb_lock);
 		/* Enable quota accounting after remounting RW */
@@ -1032,7 +1025,7 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
 	char nodestr[8];
 	struct ocfs2_blockcheck_stats stats;
 
-	mlog_entry("%p, %p, %i", sb, data, silent);
+	trace_ocfs2_fill_super(sb, data, silent);
 
 	if (!ocfs2_parse_options(sb, data, &parsed_options, 0)) {
 		status = -EINVAL;
@@ -1208,7 +1201,6 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
 			mlog_errno(status);
 			atomic_set(&osb->vol_state, VOLUME_DISABLED);
 			wake_up(&osb->osb_mount_event);
-			mlog_exit(status);
 			return status;
 		}
 	}
@@ -1222,7 +1214,6 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
 	/* Start this when the mount is almost sure of being successful */
 	ocfs2_orphan_scan_start(osb);
 
-	mlog_exit(status);
 	return status;
 
 read_super_error:
@@ -1237,7 +1228,8 @@ read_super_error:
 		ocfs2_dismount_volume(sb, 1);
 	}
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -1316,12 +1308,11 @@ static int ocfs2_parse_options(struct super_block *sb,
 			       struct mount_options *mopt,
 			       int is_remount)
 {
-	int status;
+	int status, user_stack = 0;
 	char *p;
 	u32 tmp;
 
-	mlog_entry("remount: %d, options: \"%s\"\n", is_remount,
-		   options ? options : "(none)");
+	trace_ocfs2_parse_options(is_remount, options ? options : "(none)");
 
 	mopt->commit_interval = 0;
 	mopt->mount_opt = OCFS2_MOUNT_NOINTR;
@@ -1459,6 +1450,15 @@ static int ocfs2_parse_options(struct super_block *sb,
 			memcpy(mopt->cluster_stack, args[0].from,
 			       OCFS2_STACK_LABEL_LEN);
 			mopt->cluster_stack[OCFS2_STACK_LABEL_LEN] = '\0';
+			/*
+			 * Open code the memcmp here as we don't have
+			 * an osb to pass to
+			 * ocfs2_userspace_stack().
+			 */
+			if (memcmp(mopt->cluster_stack,
+				   OCFS2_CLASSIC_CLUSTER_STACK,
+				   OCFS2_STACK_LABEL_LEN))
+				user_stack = 1;
 			break;
 		case Opt_inode64:
 			mopt->mount_opt |= OCFS2_MOUNT_INODE64;
@@ -1514,19 +1514,21 @@ static int ocfs2_parse_options(struct super_block *sb,
 		}
 	}
 
-	/* Ensure only one heartbeat mode */
-	tmp = mopt->mount_opt & (OCFS2_MOUNT_HB_LOCAL | OCFS2_MOUNT_HB_GLOBAL |
-				 OCFS2_MOUNT_HB_NONE);
-	if (hweight32(tmp) != 1) {
-		mlog(ML_ERROR, "Invalid heartbeat mount options\n");
-		status = 0;
-		goto bail;
+	if (user_stack == 0) {
+		/* Ensure only one heartbeat mode */
+		tmp = mopt->mount_opt & (OCFS2_MOUNT_HB_LOCAL |
+					 OCFS2_MOUNT_HB_GLOBAL |
+					 OCFS2_MOUNT_HB_NONE);
+		if (hweight32(tmp) != 1) {
+			mlog(ML_ERROR, "Invalid heartbeat mount options\n");
+			status = 0;
+			goto bail;
+		}
 	}
 
 	status = 1;
 
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -1617,8 +1619,6 @@ static int __init ocfs2_init(void)
 {
 	int status;
 
-	mlog_entry_void();
-
 	ocfs2_print_version();
 
 	status = init_ocfs2_uptodate_cache();
@@ -1657,10 +1657,9 @@ leave:
 		ocfs2_quota_shutdown();
 		ocfs2_free_mem_caches();
 		exit_ocfs2_uptodate_cache();
+		mlog_errno(status);
 	}
 
-	mlog_exit(status);
-
 	if (status >= 0) {
 		return register_filesystem(&ocfs2_fs_type);
 	} else
@@ -1669,8 +1668,6 @@ leave:
 
 static void __exit ocfs2_exit(void)
 {
-	mlog_entry_void();
-
 	ocfs2_quota_shutdown();
 
 	if (ocfs2_wq) {
@@ -1687,18 +1684,14 @@ static void __exit ocfs2_exit(void)
 	unregister_filesystem(&ocfs2_fs_type);
 
 	exit_ocfs2_uptodate_cache();
-
-	mlog_exit_void();
 }
 
 static void ocfs2_put_super(struct super_block *sb)
 {
-	mlog_entry("(0x%p)\n", sb);
+	trace_ocfs2_put_super(sb);
 
 	ocfs2_sync_blockdev(sb);
 	ocfs2_dismount_volume(sb, 0);
-
-	mlog_exit_void();
 }
 
 static int ocfs2_statfs(struct dentry *dentry, struct kstatfs *buf)
@@ -1710,7 +1703,7 @@ static int ocfs2_statfs(struct dentry *dentry, struct kstatfs *buf)
 	struct buffer_head *bh = NULL;
 	struct inode *inode = NULL;
 
-	mlog_entry("(%p, %p)\n", dentry->d_sb, buf);
+	trace_ocfs2_statfs(dentry->d_sb, buf);
 
 	osb = OCFS2_SB(dentry->d_sb);
 
@@ -1757,7 +1750,8 @@ bail:
 	if (inode)
 		iput(inode);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 
 	return status;
 }
@@ -1877,8 +1871,6 @@ static int ocfs2_mount_volume(struct super_block *sb)
 	int unlock_super = 0;
 	struct ocfs2_super *osb = OCFS2_SB(sb);
 
-	mlog_entry_void();
-
 	if (ocfs2_is_hard_readonly(osb))
 		goto leave;
 
@@ -1923,7 +1915,6 @@ leave:
 	if (unlock_super)
 		ocfs2_super_unlock(osb, 1);
 
-	mlog_exit(status);
 	return status;
 }
 
@@ -1933,7 +1924,7 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err)
 	struct ocfs2_super *osb = NULL;
 	char nodestr[8];
 
-	mlog_entry("(0x%p)\n", sb);
+	trace_ocfs2_dismount_volume(sb);
 
 	BUG_ON(!sb);
 	osb = OCFS2_SB(sb);
@@ -2085,8 +2076,6 @@ static int ocfs2_initialize_super(struct super_block *sb,
 	struct ocfs2_super *osb;
 	u64 total_blocks;
 
-	mlog_entry_void();
-
 	osb = kzalloc(sizeof(struct ocfs2_super), GFP_KERNEL);
 	if (!osb) {
 		status = -ENOMEM;
@@ -2150,7 +2139,6 @@ static int ocfs2_initialize_super(struct super_block *sb,
 		status = -EINVAL;
 		goto bail;
 	}
-	mlog(0, "max_slots for this device: %u\n", osb->max_slots);
 
 	ocfs2_orphan_scan_init(osb);
 
@@ -2289,7 +2277,6 @@ static int ocfs2_initialize_super(struct super_block *sb,
 	osb->s_clustersize_bits =
 		le32_to_cpu(di->id2.i_super.s_clustersize_bits);
 	osb->s_clustersize = 1 << osb->s_clustersize_bits;
-	mlog(0, "clusterbits=%d\n", osb->s_clustersize_bits);
 
 	if (osb->s_clustersize < OCFS2_MIN_CLUSTERSIZE ||
 	    osb->s_clustersize > OCFS2_MAX_CLUSTERSIZE) {
@@ -2328,11 +2315,10 @@ static int ocfs2_initialize_super(struct super_block *sb,
 		le64_to_cpu(di->id2.i_super.s_first_cluster_group);
 	osb->fs_generation = le32_to_cpu(di->i_fs_generation);
 	osb->uuid_hash = le32_to_cpu(di->id2.i_super.s_uuid_hash);
-	mlog(0, "vol_label: %s\n", osb->vol_label);
-	mlog(0, "uuid: %s\n", osb->uuid_str);
-	mlog(0, "root_blkno=%llu, system_dir_blkno=%llu\n",
-	     (unsigned long long)osb->root_blkno,
-	     (unsigned long long)osb->system_dir_blkno);
+	trace_ocfs2_initialize_super(osb->vol_label, osb->uuid_str,
+				     (unsigned long long)osb->root_blkno,
+				     (unsigned long long)osb->system_dir_blkno,
+				     osb->s_clustersize_bits);
 
 	osb->osb_dlm_debug = ocfs2_new_dlm_debug();
 	if (!osb->osb_dlm_debug) {
@@ -2375,7 +2361,6 @@ static int ocfs2_initialize_super(struct super_block *sb,
 	}
 
 bail:
-	mlog_exit(status);
 	return status;
 }
 
@@ -2391,8 +2376,6 @@ static int ocfs2_verify_volume(struct ocfs2_dinode *di,
 {
 	int status = -EAGAIN;
 
-	mlog_entry_void();
-
 	if (memcmp(di->i_signature, OCFS2_SUPER_BLOCK_SIGNATURE,
 		   strlen(OCFS2_SUPER_BLOCK_SIGNATURE)) == 0) {
 		/* We have to do a raw check of the feature here */
@@ -2447,7 +2430,8 @@ static int ocfs2_verify_volume(struct ocfs2_dinode *di,
 	}
 
 out:
-	mlog_exit(status);
+	if (status && status != -EAGAIN)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2460,8 +2444,6 @@ static int ocfs2_check_volume(struct ocfs2_super *osb)
 						  * recover
 						  * ourselves. */
 
-	mlog_entry_void();
-
 	/* Init our journal object. */
 	status = ocfs2_journal_init(osb->journal, &dirty);
 	if (status < 0) {
@@ -2511,8 +2493,6 @@ static int ocfs2_check_volume(struct ocfs2_super *osb)
 		 * ourselves as mounted. */
 	}
 
-	mlog(0, "Journal loaded.\n");
-
 	status = ocfs2_load_local_alloc(osb);
 	if (status < 0) {
 		mlog_errno(status);
@@ -2544,7 +2524,8 @@ finally:
 	if (local_alloc)
 		kfree(local_alloc);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return status;
 }
 
@@ -2556,8 +2537,6 @@ finally:
  */
 static void ocfs2_delete_osb(struct ocfs2_super *osb)
 {
-	mlog_entry_void();
-
 	/* This function assumes that the caller has the main osb resource */
 
 	ocfs2_free_slot_info(osb);
@@ -2575,8 +2554,6 @@ static void ocfs2_delete_osb(struct ocfs2_super *osb)
 	kfree(osb->uuid_str);
 	ocfs2_put_dlm_debug(osb->osb_dlm_debug);
 	memset(osb, 0, sizeof(struct ocfs2_super));
-
-	mlog_exit_void();
 }
 
 /* Put OCFS2 into a readonly state, or (if the user specifies it),
diff --git a/fs/ocfs2/symlink.c b/fs/ocfs2/symlink.c
index 9975457c981..5d22872e2bb 100644
--- a/fs/ocfs2/symlink.c
+++ b/fs/ocfs2/symlink.c
@@ -40,7 +40,6 @@
 #include <linux/pagemap.h>
 #include <linux/namei.h>
 
-#define MLOG_MASK_PREFIX ML_NAMEI
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -62,8 +61,6 @@ static char *ocfs2_fast_symlink_getlink(struct inode *inode,
 	char *link = NULL;
 	struct ocfs2_dinode *fe;
 
-	mlog_entry_void();
-
 	status = ocfs2_read_inode_block(inode, bh);
 	if (status < 0) {
 		mlog_errno(status);
@@ -74,7 +71,6 @@ static char *ocfs2_fast_symlink_getlink(struct inode *inode,
 	fe = (struct ocfs2_dinode *) (*bh)->b_data;
 	link = (char *) fe->id2.i_symlink;
 bail:
-	mlog_exit(status);
 
 	return link;
 }
@@ -88,8 +84,6 @@ static int ocfs2_readlink(struct dentry *dentry,
 	struct buffer_head *bh = NULL;
 	struct inode *inode = dentry->d_inode;
 
-	mlog_entry_void();
-
 	link = ocfs2_fast_symlink_getlink(inode, &bh);
 	if (IS_ERR(link)) {
 		ret = PTR_ERR(link);
@@ -104,7 +98,8 @@ static int ocfs2_readlink(struct dentry *dentry,
 
 	brelse(bh);
 out:
-	mlog_exit(ret);
+	if (ret < 0)
+		mlog_errno(ret);
 	return ret;
 }
 
@@ -117,8 +112,6 @@ static void *ocfs2_fast_follow_link(struct dentry *dentry,
 	struct inode *inode = dentry->d_inode;
 	struct buffer_head *bh = NULL;
 
-	mlog_entry_void();
-
 	BUG_ON(!ocfs2_inode_is_fast_symlink(inode));
 	target = ocfs2_fast_symlink_getlink(inode, &bh);
 	if (IS_ERR(target)) {
@@ -142,7 +135,8 @@ bail:
 	nd_set_link(nd, status ? ERR_PTR(status) : link);
 	brelse(bh);
 
-	mlog_exit(status);
+	if (status)
+		mlog_errno(status);
 	return NULL;
 }
 
diff --git a/fs/ocfs2/sysfile.c b/fs/ocfs2/sysfile.c
index 902efb23b6a..3d635f4bbb2 100644
--- a/fs/ocfs2/sysfile.c
+++ b/fs/ocfs2/sysfile.c
@@ -27,7 +27,6 @@
 #include <linux/types.h>
 #include <linux/highmem.h>
 
-#define MLOG_MASK_PREFIX ML_INODE
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
diff --git a/fs/ocfs2/uptodate.c b/fs/ocfs2/uptodate.c
index a0a120e82b9..52eaf33d346 100644
--- a/fs/ocfs2/uptodate.c
+++ b/fs/ocfs2/uptodate.c
@@ -54,14 +54,13 @@
 #include <linux/buffer_head.h>
 #include <linux/rbtree.h>
 
-#define MLOG_MASK_PREFIX ML_UPTODATE
-
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
 
 #include "inode.h"
 #include "uptodate.h"
+#include "ocfs2_trace.h"
 
 struct ocfs2_meta_cache_item {
 	struct rb_node	c_node;
@@ -152,8 +151,8 @@ static unsigned int ocfs2_purge_copied_metadata_tree(struct rb_root *root)
 	while ((node = rb_last(root)) != NULL) {
 		item = rb_entry(node, struct ocfs2_meta_cache_item, c_node);
 
-		mlog(0, "Purge item %llu\n",
-		     (unsigned long long) item->c_block);
+		trace_ocfs2_purge_copied_metadata_tree(
+					(unsigned long long) item->c_block);
 
 		rb_erase(&item->c_node, root);
 		kmem_cache_free(ocfs2_uptodate_cachep, item);
@@ -180,9 +179,9 @@ void ocfs2_metadata_cache_purge(struct ocfs2_caching_info *ci)
 	tree = !(ci->ci_flags & OCFS2_CACHE_FL_INLINE);
 	to_purge = ci->ci_num_cached;
 
-	mlog(0, "Purge %u %s items from Owner %llu\n", to_purge,
-	     tree ? "array" : "tree",
-	     (unsigned long long)ocfs2_metadata_cache_owner(ci));
+	trace_ocfs2_metadata_cache_purge(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		to_purge, tree);
 
 	/* If we're a tree, save off the root so that we can safely
 	 * initialize the cache. We do the work to free tree members
@@ -249,10 +248,10 @@ static int ocfs2_buffer_cached(struct ocfs2_caching_info *ci,
 
 	ocfs2_metadata_cache_lock(ci);
 
-	mlog(0, "Owner %llu, query block %llu (inline = %u)\n",
-	     (unsigned long long)ocfs2_metadata_cache_owner(ci),
-	     (unsigned long long) bh->b_blocknr,
-	     !!(ci->ci_flags & OCFS2_CACHE_FL_INLINE));
+	trace_ocfs2_buffer_cached_begin(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long) bh->b_blocknr,
+		!!(ci->ci_flags & OCFS2_CACHE_FL_INLINE));
 
 	if (ci->ci_flags & OCFS2_CACHE_FL_INLINE)
 		index = ocfs2_search_cache_array(ci, bh->b_blocknr);
@@ -261,7 +260,7 @@ static int ocfs2_buffer_cached(struct ocfs2_caching_info *ci,
 
 	ocfs2_metadata_cache_unlock(ci);
 
-	mlog(0, "index = %d, item = %p\n", index, item);
+	trace_ocfs2_buffer_cached_end(index, item);
 
 	return (index != -1) || (item != NULL);
 }
@@ -306,8 +305,9 @@ static void ocfs2_append_cache_array(struct ocfs2_caching_info *ci,
 {
 	BUG_ON(ci->ci_num_cached >= OCFS2_CACHE_INFO_MAX_ARRAY);
 
-	mlog(0, "block %llu takes position %u\n", (unsigned long long) block,
-	     ci->ci_num_cached);
+	trace_ocfs2_append_cache_array(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)block, ci->ci_num_cached);
 
 	ci->ci_cache.ci_array[ci->ci_num_cached] = block;
 	ci->ci_num_cached++;
@@ -324,8 +324,9 @@ static void __ocfs2_insert_cache_tree(struct ocfs2_caching_info *ci,
 	struct rb_node **p = &ci->ci_cache.ci_tree.rb_node;
 	struct ocfs2_meta_cache_item *tmp;
 
-	mlog(0, "Insert block %llu num = %u\n", (unsigned long long) block,
-	     ci->ci_num_cached);
+	trace_ocfs2_insert_cache_tree(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)block, ci->ci_num_cached);
 
 	while(*p) {
 		parent = *p;
@@ -389,9 +390,9 @@ static void ocfs2_expand_cache(struct ocfs2_caching_info *ci,
 		tree[i] = NULL;
 	}
 
-	mlog(0, "Expanded %llu to a tree cache: flags 0x%x, num = %u\n",
-	     (unsigned long long)ocfs2_metadata_cache_owner(ci),
-	     ci->ci_flags, ci->ci_num_cached);
+	trace_ocfs2_expand_cache(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		ci->ci_flags, ci->ci_num_cached);
 }
 
 /* Slow path function - memory allocation is necessary. See the
@@ -405,9 +406,9 @@ static void __ocfs2_set_buffer_uptodate(struct ocfs2_caching_info *ci,
 	struct ocfs2_meta_cache_item *tree[OCFS2_CACHE_INFO_MAX_ARRAY] =
 		{ NULL, };
 
-	mlog(0, "Owner %llu, block %llu, expand = %d\n",
-	     (unsigned long long)ocfs2_metadata_cache_owner(ci),
-	     (unsigned long long)block, expand_tree);
+	trace_ocfs2_set_buffer_uptodate(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)block, expand_tree);
 
 	new = kmem_cache_alloc(ocfs2_uptodate_cachep, GFP_NOFS);
 	if (!new) {
@@ -433,7 +434,6 @@ static void __ocfs2_set_buffer_uptodate(struct ocfs2_caching_info *ci,
 
 	ocfs2_metadata_cache_lock(ci);
 	if (ocfs2_insert_can_use_array(ci)) {
-		mlog(0, "Someone cleared the tree underneath us\n");
 		/* Ok, items were removed from the cache in between
 		 * locks. Detect this and revert back to the fast path */
 		ocfs2_append_cache_array(ci, block);
@@ -490,9 +490,9 @@ void ocfs2_set_buffer_uptodate(struct ocfs2_caching_info *ci,
 	if (ocfs2_buffer_cached(ci, bh))
 		return;
 
-	mlog(0, "Owner %llu, inserting block %llu\n",
-	     (unsigned long long)ocfs2_metadata_cache_owner(ci),
-	     (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_set_buffer_uptodate_begin(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)bh->b_blocknr);
 
 	/* No need to recheck under spinlock - insertion is guarded by
 	 * co_io_lock() */
@@ -542,8 +542,9 @@ static void ocfs2_remove_metadata_array(struct ocfs2_caching_info *ci,
 	BUG_ON(index >= ci->ci_num_cached);
 	BUG_ON(!ci->ci_num_cached);
 
-	mlog(0, "remove index %d (num_cached = %u\n", index,
-	     ci->ci_num_cached);
+	trace_ocfs2_remove_metadata_array(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		index, ci->ci_num_cached);
 
 	ci->ci_num_cached--;
 
@@ -559,8 +560,9 @@ static void ocfs2_remove_metadata_array(struct ocfs2_caching_info *ci,
 static void ocfs2_remove_metadata_tree(struct ocfs2_caching_info *ci,
 				       struct ocfs2_meta_cache_item *item)
 {
-	mlog(0, "remove block %llu from tree\n",
-	     (unsigned long long) item->c_block);
+	trace_ocfs2_remove_metadata_tree(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long)item->c_block);
 
 	rb_erase(&item->c_node, &ci->ci_cache.ci_tree);
 	ci->ci_num_cached--;
@@ -573,10 +575,10 @@ static void ocfs2_remove_block_from_cache(struct ocfs2_caching_info *ci,
 	struct ocfs2_meta_cache_item *item = NULL;
 
 	ocfs2_metadata_cache_lock(ci);
-	mlog(0, "Owner %llu, remove %llu, items = %u, array = %u\n",
-	     (unsigned long long)ocfs2_metadata_cache_owner(ci),
-	     (unsigned long long) block, ci->ci_num_cached,
-	     ci->ci_flags & OCFS2_CACHE_FL_INLINE);
+	trace_ocfs2_remove_block_from_cache(
+		(unsigned long long)ocfs2_metadata_cache_owner(ci),
+		(unsigned long long) block, ci->ci_num_cached,
+		ci->ci_flags);
 
 	if (ci->ci_flags & OCFS2_CACHE_FL_INLINE) {
 		index = ocfs2_search_cache_array(ci, block);
@@ -626,9 +628,6 @@ int __init init_ocfs2_uptodate_cache(void)
 	if (!ocfs2_uptodate_cachep)
 		return -ENOMEM;
 
-	mlog(0, "%u inlined cache items per inode.\n",
-	     OCFS2_CACHE_INFO_MAX_ARRAY);
-
 	return 0;
 }
 
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index 67cd4391464..52e3dadd364 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -37,7 +37,6 @@
 #include <linux/string.h>
 #include <linux/security.h>
 
-#define MLOG_MASK_PREFIX ML_XATTR
 #include <cluster/masklog.h>
 
 #include "ocfs2.h"
@@ -57,6 +56,7 @@
 #include "xattr.h"
 #include "refcounttree.h"
 #include "acl.h"
+#include "ocfs2_trace.h"
 
 struct ocfs2_xattr_def_value_root {
 	struct ocfs2_xattr_value_root	xv;
@@ -474,8 +474,7 @@ static int ocfs2_validate_xattr_block(struct super_block *sb,
 	struct ocfs2_xattr_block *xb =
 		(struct ocfs2_xattr_block *)bh->b_data;
 
-	mlog(0, "Validating xattr block %llu\n",
-	     (unsigned long long)bh->b_blocknr);
+	trace_ocfs2_validate_xattr_block((unsigned long long)bh->b_blocknr);
 
 	BUG_ON(!buffer_uptodate(bh));
 
@@ -715,11 +714,11 @@ static int ocfs2_xattr_extend_allocation(struct inode *inode,
 	u32 prev_clusters, logical_start = le32_to_cpu(vb->vb_xv->xr_clusters);
 	struct ocfs2_extent_tree et;
 
-	mlog(0, "(clusters_to_add for xattr= %u)\n", clusters_to_add);
-
 	ocfs2_init_xattr_value_extent_tree(&et, INODE_CACHE(inode), vb);
 
 	while (clusters_to_add) {
+		trace_ocfs2_xattr_extend_allocation(clusters_to_add);
+
 		status = vb->vb_access(handle, INODE_CACHE(inode), vb->vb_bh,
 				       OCFS2_JOURNAL_ACCESS_WRITE);
 		if (status < 0) {
@@ -754,8 +753,6 @@ static int ocfs2_xattr_extend_allocation(struct inode *inode,
 			 */
 			BUG_ON(why == RESTART_META);
 
-			mlog(0, "restarting xattr value extension for %u"
-			     " clusters,.\n", clusters_to_add);
 			credits = ocfs2_calc_extend_credits(inode->i_sb,
 							    &vb->vb_xv->xr_list,
 							    clusters_to_add);
@@ -3246,8 +3243,8 @@ static int ocfs2_init_xattr_set_ctxt(struct inode *inode,
 	}
 
 	meta_add += extra_meta;
-	mlog(0, "Set xattr %s, reserve meta blocks = %d, clusters = %d, "
-	     "credits = %d\n", xi->xi_name, meta_add, clusters_add, *credits);
+	trace_ocfs2_init_xattr_set_ctxt(xi->xi_name, meta_add,
+					clusters_add, *credits);
 
 	if (meta_add) {
 		ret = ocfs2_reserve_new_metadata_blocks(osb, meta_add,
@@ -3887,8 +3884,10 @@ static int ocfs2_xattr_bucket_find(struct inode *inode,
 
 	if (found) {
 		xs->here = &xs->header->xh_entries[index];
-		mlog(0, "find xattr %s in bucket %llu, entry = %u\n", name,
-		     (unsigned long long)bucket_blkno(xs->bucket), index);
+		trace_ocfs2_xattr_bucket_find(OCFS2_I(inode)->ip_blkno,
+			name, name_index, name_hash,
+			(unsigned long long)bucket_blkno(xs->bucket),
+			index);
 	} else
 		ret = -ENODATA;
 
@@ -3915,8 +3914,10 @@ static int ocfs2_xattr_index_block_find(struct inode *inode,
 	if (le16_to_cpu(el->l_next_free_rec) == 0)
 		return -ENODATA;
 
-	mlog(0, "find xattr %s, hash = %u, index = %d in xattr tree\n",
-	     name, name_hash, name_index);
+	trace_ocfs2_xattr_index_block_find(OCFS2_I(inode)->ip_blkno,
+					name, name_index, name_hash,
+					(unsigned long long)root_bh->b_blocknr,
+					-1);
 
 	ret = ocfs2_xattr_get_rec(inode, name_hash, &p_blkno, &first_hash,
 				  &num_clusters, el);
@@ -3927,9 +3928,10 @@ static int ocfs2_xattr_index_block_find(struct inode *inode,
 
 	BUG_ON(p_blkno == 0 || num_clusters == 0 || first_hash > name_hash);
 
-	mlog(0, "find xattr extent rec %u clusters from %llu, the first hash "
-	     "in the rec is %u\n", num_clusters, (unsigned long long)p_blkno,
-	     first_hash);
+	trace_ocfs2_xattr_index_block_find_rec(OCFS2_I(inode)->ip_blkno,
+					name, name_index, first_hash,
+					(unsigned long long)p_blkno,
+					num_clusters);
 
 	ret = ocfs2_xattr_bucket_find(inode, name_index, name, name_hash,
 				      p_blkno, first_hash, num_clusters, xs);
@@ -3955,8 +3957,9 @@ static int ocfs2_iterate_xattr_buckets(struct inode *inode,
 		return -ENOMEM;
 	}
 
-	mlog(0, "iterating xattr buckets in %u clusters starting from %llu\n",
-	     clusters, (unsigned long long)blkno);
+	trace_ocfs2_iterate_xattr_buckets(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno,
+		(unsigned long long)blkno, clusters);
 
 	for (i = 0; i < num_buckets; i++, blkno += bucket->bu_blocks) {
 		ret = ocfs2_read_xattr_bucket(bucket, blkno);
@@ -3972,8 +3975,7 @@ static int ocfs2_iterate_xattr_buckets(struct inode *inode,
 		if (i == 0)
 			num_buckets = le16_to_cpu(bucket_xh(bucket)->xh_num_buckets);
 
-		mlog(0, "iterating xattr bucket %llu, first hash %u\n",
-		     (unsigned long long)blkno,
+		trace_ocfs2_iterate_xattr_bucket((unsigned long long)blkno,
 		     le32_to_cpu(bucket_xh(bucket)->xh_entries[0].xe_name_hash));
 		if (func) {
 			ret = func(inode, bucket, para);
@@ -4173,9 +4175,9 @@ static void ocfs2_cp_xattr_block_to_bucket(struct inode *inode,
 	char *src = xb_bh->b_data;
 	char *target = bucket_block(bucket, blks - 1);
 
-	mlog(0, "cp xattr from block %llu to bucket %llu\n",
-	     (unsigned long long)xb_bh->b_blocknr,
-	     (unsigned long long)bucket_blkno(bucket));
+	trace_ocfs2_cp_xattr_block_to_bucket_begin(
+				(unsigned long long)xb_bh->b_blocknr,
+				(unsigned long long)bucket_blkno(bucket));
 
 	for (i = 0; i < blks; i++)
 		memset(bucket_block(bucket, i), 0, blocksize);
@@ -4211,8 +4213,7 @@ static void ocfs2_cp_xattr_block_to_bucket(struct inode *inode,
 	for (i = 0; i < count; i++)
 		le16_add_cpu(&xh->xh_entries[i].xe_name_offset, off_change);
 
-	mlog(0, "copy entry: start = %u, size = %u, offset_change = %u\n",
-	     offset, size, off_change);
+	trace_ocfs2_cp_xattr_block_to_bucket_end(offset, size, off_change);
 
 	sort(target + offset, count, sizeof(struct ocfs2_xattr_entry),
 	     cmp_xe, swap_xe);
@@ -4261,8 +4262,8 @@ static int ocfs2_xattr_create_index_block(struct inode *inode,
 	struct ocfs2_xattr_tree_root *xr;
 	u16 xb_flags = le16_to_cpu(xb->xb_flags);
 
-	mlog(0, "create xattr index block for %llu\n",
-	     (unsigned long long)xb_bh->b_blocknr);
+	trace_ocfs2_xattr_create_index_block_begin(
+				(unsigned long long)xb_bh->b_blocknr);
 
 	BUG_ON(xb_flags & OCFS2_XATTR_INDEXED);
 	BUG_ON(!xs->bucket);
@@ -4295,8 +4296,7 @@ static int ocfs2_xattr_create_index_block(struct inode *inode,
 	 */
 	blkno = ocfs2_clusters_to_blocks(inode->i_sb, bit_off);
 
-	mlog(0, "allocate 1 cluster from %llu to xattr block\n",
-	     (unsigned long long)blkno);
+	trace_ocfs2_xattr_create_index_block((unsigned long long)blkno);
 
 	ret = ocfs2_init_xattr_bucket(xs->bucket, blkno);
 	if (ret) {
@@ -4400,8 +4400,7 @@ static int ocfs2_defrag_xattr_bucket(struct inode *inode,
 	entries = (char *)xh->xh_entries;
 	xh_free_start = le16_to_cpu(xh->xh_free_start);
 
-	mlog(0, "adjust xattr bucket in %llu, count = %u, "
-	     "xh_free_start = %u, xh_name_value_len = %u.\n",
+	trace_ocfs2_defrag_xattr_bucket(
 	     (unsigned long long)blkno, le16_to_cpu(xh->xh_count),
 	     xh_free_start, le16_to_cpu(xh->xh_name_value_len));
 
@@ -4503,8 +4502,9 @@ static int ocfs2_mv_xattr_bucket_cross_cluster(struct inode *inode,
 	BUG_ON(le16_to_cpu(bucket_xh(first)->xh_num_buckets) < num_buckets);
 	BUG_ON(OCFS2_XATTR_BUCKET_SIZE == OCFS2_SB(sb)->s_clustersize);
 
-	mlog(0, "move half of xattrs in cluster %llu to %llu\n",
-	     (unsigned long long)last_cluster_blkno, (unsigned long long)new_blkno);
+	trace_ocfs2_mv_xattr_bucket_cross_cluster(
+				(unsigned long long)last_cluster_blkno,
+				(unsigned long long)new_blkno);
 
 	ret = ocfs2_mv_xattr_buckets(inode, handle, bucket_blkno(first),
 				     last_cluster_blkno, new_blkno,
@@ -4614,8 +4614,8 @@ static int ocfs2_divide_xattr_bucket(struct inode *inode,
 	struct ocfs2_xattr_entry *xe;
 	int blocksize = inode->i_sb->s_blocksize;
 
-	mlog(0, "move some of xattrs from bucket %llu to %llu\n",
-	     (unsigned long long)blk, (unsigned long long)new_blk);
+	trace_ocfs2_divide_xattr_bucket_begin((unsigned long long)blk,
+					      (unsigned long long)new_blk);
 
 	s_bucket = ocfs2_xattr_bucket_new(inode);
 	t_bucket = ocfs2_xattr_bucket_new(inode);
@@ -4714,9 +4714,9 @@ static int ocfs2_divide_xattr_bucket(struct inode *inode,
 	 */
 	xe = &xh->xh_entries[start];
 	len = sizeof(struct ocfs2_xattr_entry) * (count - start);
-	mlog(0, "mv xattr entry len %d from %d to %d\n", len,
-	     (int)((char *)xe - (char *)xh),
-	     (int)((char *)xh->xh_entries - (char *)xh));
+	trace_ocfs2_divide_xattr_bucket_move(len,
+			(int)((char *)xe - (char *)xh),
+			(int)((char *)xh->xh_entries - (char *)xh));
 	memmove((char *)xh->xh_entries, (char *)xe, len);
 	xe = &xh->xh_entries[count - start];
 	len = sizeof(struct ocfs2_xattr_entry) * start;
@@ -4788,9 +4788,9 @@ static int ocfs2_cp_xattr_bucket(struct inode *inode,
 
 	BUG_ON(s_blkno == t_blkno);
 
-	mlog(0, "cp bucket %llu to %llu, target is %d\n",
-	     (unsigned long long)s_blkno, (unsigned long long)t_blkno,
-	     t_is_new);
+	trace_ocfs2_cp_xattr_bucket((unsigned long long)s_blkno,
+				    (unsigned long long)t_blkno,
+				    t_is_new);
 
 	s_bucket = ocfs2_xattr_bucket_new(inode);
 	t_bucket = ocfs2_xattr_bucket_new(inode);
@@ -4862,8 +4862,8 @@ static int ocfs2_mv_xattr_buckets(struct inode *inode, handle_t *handle,
 	int num_buckets = ocfs2_xattr_buckets_per_cluster(osb);
 	struct ocfs2_xattr_bucket *old_first, *new_first;
 
-	mlog(0, "mv xattrs from cluster %llu to %llu\n",
-	     (unsigned long long)last_blk, (unsigned long long)to_blk);
+	trace_ocfs2_mv_xattr_buckets((unsigned long long)last_blk,
+				     (unsigned long long)to_blk);
 
 	BUG_ON(start_bucket >= num_buckets);
 	if (start_bucket) {
@@ -5013,9 +5013,9 @@ static int ocfs2_adjust_xattr_cross_cluster(struct inode *inode,
 {
 	int ret;
 
-	mlog(0, "adjust xattrs from cluster %llu len %u to %llu\n",
-	     (unsigned long long)bucket_blkno(first), prev_clusters,
-	     (unsigned long long)new_blk);
+	trace_ocfs2_adjust_xattr_cross_cluster(
+			(unsigned long long)bucket_blkno(first),
+			(unsigned long long)new_blk, prev_clusters);
 
 	if (ocfs2_xattr_buckets_per_cluster(OCFS2_SB(inode->i_sb)) > 1) {
 		ret = ocfs2_mv_xattr_bucket_cross_cluster(inode,
@@ -5088,10 +5088,10 @@ static int ocfs2_add_new_xattr_cluster(struct inode *inode,
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	struct ocfs2_extent_tree et;
 
-	mlog(0, "Add new xattr cluster for %llu, previous xattr hash = %u, "
-	     "previous xattr blkno = %llu\n",
-	     (unsigned long long)OCFS2_I(inode)->ip_blkno,
-	     prev_cpos, (unsigned long long)bucket_blkno(first));
+	trace_ocfs2_add_new_xattr_cluster_begin(
+		(unsigned long long)OCFS2_I(inode)->ip_blkno,
+		(unsigned long long)bucket_blkno(first),
+		prev_cpos, prev_clusters);
 
 	ocfs2_init_xattr_tree_extent_tree(&et, INODE_CACHE(inode), root_bh);
 
@@ -5113,8 +5113,7 @@ static int ocfs2_add_new_xattr_cluster(struct inode *inode,
 	BUG_ON(num_bits > clusters_to_add);
 
 	block = ocfs2_clusters_to_blocks(osb->sb, bit_off);
-	mlog(0, "Allocating %u clusters at block %u for xattr in inode %llu\n",
-	     num_bits, bit_off, (unsigned long long)OCFS2_I(inode)->ip_blkno);
+	trace_ocfs2_add_new_xattr_cluster((unsigned long long)block, num_bits);
 
 	if (bucket_blkno(first) + (prev_clusters * bpc) == block &&
 	    (prev_clusters + num_bits) << osb->s_clustersize_bits <=
@@ -5130,8 +5129,6 @@ static int ocfs2_add_new_xattr_cluster(struct inode *inode,
 		 */
 		v_start = prev_cpos + prev_clusters;
 		*num_clusters = prev_clusters + num_bits;
-		mlog(0, "Add contiguous %u clusters to previous extent rec.\n",
-		     num_bits);
 	} else {
 		ret = ocfs2_adjust_xattr_cross_cluster(inode,
 						       handle,
@@ -5147,8 +5144,8 @@ static int ocfs2_add_new_xattr_cluster(struct inode *inode,
 		}
 	}
 
-	mlog(0, "Insert %u clusters at block %llu for xattr at %u\n",
-	     num_bits, (unsigned long long)block, v_start);
+	trace_ocfs2_add_new_xattr_cluster_insert((unsigned long long)block,
+						 v_start, num_bits);
 	ret = ocfs2_insert_extent(handle, &et, v_start, block,
 				  num_bits, 0, ctxt->meta_ac);
 	if (ret < 0) {
@@ -5183,9 +5180,9 @@ static int ocfs2_extend_xattr_bucket(struct inode *inode,
 	u64 end_blk;
 	u16 new_bucket = le16_to_cpu(bucket_xh(first)->xh_num_buckets);
 
-	mlog(0, "extend xattr bucket in %llu, xattr extend rec starting "
-	     "from %llu, len = %u\n", (unsigned long long)target_blk,
-	     (unsigned long long)bucket_blkno(first), num_clusters);
+	trace_ocfs2_extend_xattr_bucket((unsigned long long)target_blk,
+					(unsigned long long)bucket_blkno(first),
+					num_clusters, new_bucket);
 
 	/* The extent must have room for an additional bucket */
 	BUG_ON(new_bucket >=
@@ -5265,8 +5262,8 @@ static int ocfs2_add_new_xattr_bucket(struct inode *inode,
 	/* The bucket at the front of the extent */
 	struct ocfs2_xattr_bucket *first;
 
-	mlog(0, "Add new xattr bucket starting from %llu\n",
-	     (unsigned long long)bucket_blkno(target));
+	trace_ocfs2_add_new_xattr_bucket(
+				(unsigned long long)bucket_blkno(target));
 
 	/* The first bucket of the original extent */
 	first = ocfs2_xattr_bucket_new(inode);
@@ -5382,8 +5379,8 @@ static int ocfs2_xattr_bucket_value_truncate(struct inode *inode,
 	 * modified something.  We have to assume they did, and dirty
 	 * the whole bucket.  This leaves us in a consistent state.
 	 */
-	mlog(0, "truncate %u in xattr bucket %llu to %d bytes.\n",
-	     xe_off, (unsigned long long)bucket_blkno(bucket), len);
+	trace_ocfs2_xattr_bucket_value_truncate(
+			(unsigned long long)bucket_blkno(bucket), xe_off, len);
 	ret = ocfs2_xattr_value_truncate(inode, &vb, len, ctxt);
 	if (ret) {
 		mlog_errno(ret);
@@ -5433,8 +5430,9 @@ static int ocfs2_rm_xattr_cluster(struct inode *inode,
 
 	ocfs2_init_dealloc_ctxt(&dealloc);
 
-	mlog(0, "rm xattr extent rec at %u len = %u, start from %llu\n",
-	     cpos, len, (unsigned long long)blkno);
+	trace_ocfs2_rm_xattr_cluster(
+			(unsigned long long)OCFS2_I(inode)->ip_blkno,
+			(unsigned long long)blkno, cpos, len);
 
 	ocfs2_remove_xattr_clusters_from_cache(INODE_CACHE(inode), blkno,
 					       len);
@@ -5538,7 +5536,7 @@ static int ocfs2_xattr_set_entry_bucket(struct inode *inode,
 	int ret;
 	struct ocfs2_xa_loc loc;
 
-	mlog_entry("Set xattr %s in xattr bucket\n", xi->xi_name);
+	trace_ocfs2_xattr_set_entry_bucket(xi->xi_name);
 
 	ocfs2_init_xattr_bucket_xa_loc(&loc, xs->bucket,
 				       xs->not_found ? NULL : xs->here);
@@ -5570,7 +5568,6 @@ static int ocfs2_xattr_set_entry_bucket(struct inode *inode,
 
 
 out:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -5581,7 +5578,7 @@ static int ocfs2_xattr_set_entry_index_block(struct inode *inode,
 {
 	int ret;
 
-	mlog_entry("Set xattr %s in xattr index block\n", xi->xi_name);
+	trace_ocfs2_xattr_set_entry_index_block(xi->xi_name);
 
 	ret = ocfs2_xattr_set_entry_bucket(inode, xi, xs, ctxt);
 	if (!ret)
@@ -5637,7 +5634,6 @@ static int ocfs2_xattr_set_entry_index_block(struct inode *inode,
 		mlog_errno(ret);
 
 out:
-	mlog_exit(ret);
 	return ret;
 }
 
@@ -6041,9 +6037,9 @@ static int ocfs2_xattr_bucket_value_refcount(struct inode *inode,
 	if (ocfs2_meta_ecc(OCFS2_SB(inode->i_sb)))
 		p = &refcount;
 
-	mlog(0, "refcount bucket %llu, count = %u\n",
-	     (unsigned long long)bucket_blkno(bucket),
-	     le16_to_cpu(xh->xh_count));
+	trace_ocfs2_xattr_bucket_value_refcount(
+				(unsigned long long)bucket_blkno(bucket),
+				le16_to_cpu(xh->xh_count));
 	for (i = 0; i < le16_to_cpu(xh->xh_count); i++) {
 		xe = &xh->xh_entries[i];
 
@@ -6339,8 +6335,8 @@ static int ocfs2_reflink_xattr_header(handle_t *handle,
 	u32 clusters, cpos, p_cluster, num_clusters;
 	unsigned int ext_flags = 0;
 
-	mlog(0, "reflink xattr in container %llu, count = %u\n",
-	     (unsigned long long)old_bh->b_blocknr, le16_to_cpu(xh->xh_count));
+	trace_ocfs2_reflink_xattr_header((unsigned long long)old_bh->b_blocknr,
+					 le16_to_cpu(xh->xh_count));
 
 	last = &new_xh->xh_entries[le16_to_cpu(new_xh->xh_count)];
 	for (i = 0, j = 0; i < le16_to_cpu(xh->xh_count); i++, j++) {
@@ -6540,8 +6536,8 @@ static int ocfs2_create_empty_xattr_block(struct inode *inode,
 		goto out;
 	}
 
-	mlog(0, "create new xattr block for inode %llu, index = %d\n",
-	     (unsigned long long)fe_bh->b_blocknr, indexed);
+	trace_ocfs2_create_empty_xattr_block(
+				(unsigned long long)fe_bh->b_blocknr, indexed);
 	ret = ocfs2_create_xattr_block(inode, fe_bh, &ctxt, indexed,
 				       ret_bh);
 	if (ret)
@@ -6952,8 +6948,8 @@ static int ocfs2_reflink_xattr_buckets(handle_t *handle,
 		if (ret)
 			mlog_errno(ret);
 
-		mlog(0, "insert new xattr extent rec start %llu len %u to %u\n",
-		     (unsigned long long)new_blkno, num_clusters, reflink_cpos);
+		trace_ocfs2_reflink_xattr_buckets((unsigned long long)new_blkno,
+						  num_clusters, reflink_cpos);
 
 		len -= num_clusters;
 		blkno += ocfs2_clusters_to_blocks(inode->i_sb, num_clusters);
@@ -6982,8 +6978,7 @@ static int ocfs2_reflink_xattr_rec(struct inode *inode,
 	struct ocfs2_alloc_context *data_ac = NULL;
 	struct ocfs2_extent_tree et;
 
-	mlog(0, "reflink xattr buckets %llu len %u\n",
-	     (unsigned long long)blkno, len);
+	trace_ocfs2_reflink_xattr_rec((unsigned long long)blkno, len);
 
 	ocfs2_init_xattr_tree_extent_tree(&et,
 					  INODE_CACHE(args->reflink->new_inode),
diff --git a/fs/open.c b/fs/open.c
index 5a2c6ebc22b..3cac0bda46d 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -233,6 +233,14 @@ int do_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 
 	if (!(file->f_mode & FMODE_WRITE))
 		return -EBADF;
+
+	/* It's not possible punch hole on append only file */
+	if (mode & FALLOC_FL_PUNCH_HOLE && IS_APPEND(inode))
+		return -EPERM;
+
+	if (IS_IMMUTABLE(inode))
+		return -EPERM;
+
 	/*
 	 * Revalidate the write permissions, in case security policy has
 	 * changed since the files were opened.
@@ -565,13 +573,15 @@ SYSCALL_DEFINE5(fchownat, int, dfd, const char __user *, filename, uid_t, user,
 {
 	struct path path;
 	int error = -EINVAL;
-	int follow;
+	int lookup_flags;
 
-	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
+	if ((flag & ~(AT_SYMLINK_NOFOLLOW | AT_EMPTY_PATH)) != 0)
 		goto out;
 
-	follow = (flag & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW;
-	error = user_path_at(dfd, filename, follow, &path);
+	lookup_flags = (flag & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW;
+	if (flag & AT_EMPTY_PATH)
+		lookup_flags |= LOOKUP_EMPTY;
+	error = user_path_at(dfd, filename, lookup_flags, &path);
 	if (error)
 		goto out;
 	error = mnt_want_write(path.mnt);
@@ -661,11 +671,16 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
 					int (*open)(struct inode *, struct file *),
 					const struct cred *cred)
 {
+	static const struct file_operations empty_fops = {};
 	struct inode *inode;
 	int error;
 
 	f->f_mode = OPEN_FMODE(f->f_flags) | FMODE_LSEEK |
 				FMODE_PREAD | FMODE_PWRITE;
+
+	if (unlikely(f->f_flags & O_PATH))
+		f->f_mode = FMODE_PATH;
+
 	inode = dentry->d_inode;
 	if (f->f_mode & FMODE_WRITE) {
 		error = __get_file_write_access(inode, mnt);
@@ -679,9 +694,15 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
 	f->f_path.dentry = dentry;
 	f->f_path.mnt = mnt;
 	f->f_pos = 0;
-	f->f_op = fops_get(inode->i_fop);
 	file_sb_list_add(f, inode->i_sb);
 
+	if (unlikely(f->f_mode & FMODE_PATH)) {
+		f->f_op = &empty_fops;
+		return f;
+	}
+
+	f->f_op = fops_get(inode->i_fop);
+
 	error = security_dentry_open(f, cred);
 	if (error)
 		goto cleanup_all;
@@ -882,15 +903,110 @@ void fd_install(unsigned int fd, struct file *file)
 
 EXPORT_SYMBOL(fd_install);
 
+static inline int build_open_flags(int flags, int mode, struct open_flags *op)
+{
+	int lookup_flags = 0;
+	int acc_mode;
+
+	if (!(flags & O_CREAT))
+		mode = 0;
+	op->mode = mode;
+
+	/* Must never be set by userspace */
+	flags &= ~FMODE_NONOTIFY;
+
+	/*
+	 * O_SYNC is implemented as __O_SYNC|O_DSYNC.  As many places only
+	 * check for O_DSYNC if the need any syncing at all we enforce it's
+	 * always set instead of having to deal with possibly weird behaviour
+	 * for malicious applications setting only __O_SYNC.
+	 */
+	if (flags & __O_SYNC)
+		flags |= O_DSYNC;
+
+	/*
+	 * If we have O_PATH in the open flag. Then we
+	 * cannot have anything other than the below set of flags
+	 */
+	if (flags & O_PATH) {
+		flags &= O_DIRECTORY | O_NOFOLLOW | O_PATH;
+		acc_mode = 0;
+	} else {
+		acc_mode = MAY_OPEN | ACC_MODE(flags);
+	}
+
+	op->open_flag = flags;
+
+	/* O_TRUNC implies we need access checks for write permissions */
+	if (flags & O_TRUNC)
+		acc_mode |= MAY_WRITE;
+
+	/* Allow the LSM permission hook to distinguish append
+	   access from general write access. */
+	if (flags & O_APPEND)
+		acc_mode |= MAY_APPEND;
+
+	op->acc_mode = acc_mode;
+
+	op->intent = flags & O_PATH ? 0 : LOOKUP_OPEN;
+
+	if (flags & O_CREAT) {
+		op->intent |= LOOKUP_CREATE;
+		if (flags & O_EXCL)
+			op->intent |= LOOKUP_EXCL;
+	}
+
+	if (flags & O_DIRECTORY)
+		lookup_flags |= LOOKUP_DIRECTORY;
+	if (!(flags & O_NOFOLLOW))
+		lookup_flags |= LOOKUP_FOLLOW;
+	return lookup_flags;
+}
+
+/**
+ * filp_open - open file and return file pointer
+ *
+ * @filename:	path to open
+ * @flags:	open flags as per the open(2) second argument
+ * @mode:	mode for the new file if O_CREAT is set, else ignored
+ *
+ * This is the helper to open a file from kernelspace if you really
+ * have to.  But in generally you should not do this, so please move
+ * along, nothing to see here..
+ */
+struct file *filp_open(const char *filename, int flags, int mode)
+{
+	struct open_flags op;
+	int lookup = build_open_flags(flags, mode, &op);
+	return do_filp_open(AT_FDCWD, filename, &op, lookup);
+}
+EXPORT_SYMBOL(filp_open);
+
+struct file *file_open_root(struct dentry *dentry, struct vfsmount *mnt,
+			    const char *filename, int flags)
+{
+	struct open_flags op;
+	int lookup = build_open_flags(flags, 0, &op);
+	if (flags & O_CREAT)
+		return ERR_PTR(-EINVAL);
+	if (!filename && (flags & O_DIRECTORY))
+		if (!dentry->d_inode->i_op->lookup)
+			return ERR_PTR(-ENOTDIR);
+	return do_file_open_root(dentry, mnt, filename, &op, lookup);
+}
+EXPORT_SYMBOL(file_open_root);
+
 long do_sys_open(int dfd, const char __user *filename, int flags, int mode)
 {
+	struct open_flags op;
+	int lookup = build_open_flags(flags, mode, &op);
 	char *tmp = getname(filename);
 	int fd = PTR_ERR(tmp);
 
 	if (!IS_ERR(tmp)) {
 		fd = get_unused_fd_flags(flags);
 		if (fd >= 0) {
-			struct file *f = do_filp_open(dfd, tmp, flags, mode, 0);
+			struct file *f = do_filp_open(dfd, tmp, &op, lookup);
 			if (IS_ERR(f)) {
 				put_unused_fd(fd);
 				fd = PTR_ERR(f);
@@ -960,8 +1076,10 @@ int filp_close(struct file *filp, fl_owner_t id)
 	if (filp->f_op && filp->f_op->flush)
 		retval = filp->f_op->flush(filp, id);
 
-	dnotify_flush(filp, id);
-	locks_remove_posix(filp, id);
+	if (likely(!(filp->f_mode & FMODE_PATH))) {
+		dnotify_flush(filp, id);
+		locks_remove_posix(filp, id);
+	}
 	fput(filp);
 	return retval;
 }
diff --git a/fs/partitions/ldm.c b/fs/partitions/ldm.c
index 789c625c7aa..b10e3540d5b 100644
--- a/fs/partitions/ldm.c
+++ b/fs/partitions/ldm.c
@@ -251,6 +251,11 @@ static bool ldm_parse_vmdb (const u8 *data, struct vmdb *vm)
 	}
 
 	vm->vblk_size     = get_unaligned_be32(data + 0x08);
+	if (vm->vblk_size == 0) {
+		ldm_error ("Illegal VBLK size");
+		return false;
+	}
+
 	vm->vblk_offset   = get_unaligned_be32(data + 0x0C);
 	vm->last_vblk_seq = get_unaligned_be32(data + 0x04);
 
diff --git a/fs/partitions/osf.c b/fs/partitions/osf.c
index 48cec7cbca1..be03a0b08b4 100644
--- a/fs/partitions/osf.c
+++ b/fs/partitions/osf.c
@@ -10,10 +10,13 @@
 #include "check.h"
 #include "osf.h"
 
+#define MAX_OSF_PARTITIONS 8
+
 int osf_partition(struct parsed_partitions *state)
 {
 	int i;
 	int slot = 1;
+	unsigned int npartitions;
 	Sector sect;
 	unsigned char *data;
 	struct disklabel {
@@ -45,7 +48,7 @@ int osf_partition(struct parsed_partitions *state)
 			u8  p_fstype;
 			u8  p_frag;
 			__le16 p_cpg;
-		} d_partitions[8];
+		} d_partitions[MAX_OSF_PARTITIONS];
 	} * label;
 	struct d_partition * partition;
 
@@ -63,7 +66,12 @@ int osf_partition(struct parsed_partitions *state)
 		put_dev_sector(sect);
 		return 0;
 	}
-	for (i = 0 ; i < le16_to_cpu(label->d_npartitions); i++, partition++) {
+	npartitions = le16_to_cpu(label->d_npartitions);
+	if (npartitions > MAX_OSF_PARTITIONS) {
+		put_dev_sector(sect);
+		return 0;
+	}
+	for (i = 0 ; i < npartitions; i++, partition++) {
 		if (slot == state->limit)
 		        break;
 		if (le32_to_cpu(partition->p_size))
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 9d096e82b20..d49c4b5d2c3 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2620,35 +2620,6 @@ static const struct pid_entry proc_base_stuff[] = {
 		&proc_self_inode_operations, NULL, {}),
 };
 
-/*
- *	Exceptional case: normally we are not allowed to unhash a busy
- * directory. In this case, however, we can do it - no aliasing problems
- * due to the way we treat inodes.
- */
-static int proc_base_revalidate(struct dentry *dentry, struct nameidata *nd)
-{
-	struct inode *inode;
-	struct task_struct *task;
-
-	if (nd->flags & LOOKUP_RCU)
-		return -ECHILD;
-
-	inode = dentry->d_inode;
-	task = get_proc_task(inode);
-	if (task) {
-		put_task_struct(task);
-		return 1;
-	}
-	d_drop(dentry);
-	return 0;
-}
-
-static const struct dentry_operations proc_base_dentry_operations =
-{
-	.d_revalidate	= proc_base_revalidate,
-	.d_delete	= pid_delete_dentry,
-};
-
 static struct dentry *proc_base_instantiate(struct inode *dir,
 	struct dentry *dentry, struct task_struct *task, const void *ptr)
 {
@@ -2685,7 +2656,6 @@ static struct dentry *proc_base_instantiate(struct inode *dir,
 	if (p->fop)
 		inode->i_fop = p->fop;
 	ei->op = p->op;
-	d_set_d_op(dentry, &proc_base_dentry_operations);
 	d_add(dentry, inode);
 	error = NULL;
 out:
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 176ce4cda68..d6a7ca1fdac 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -27,6 +27,7 @@
 static void proc_evict_inode(struct inode *inode)
 {
 	struct proc_dir_entry *de;
+	struct ctl_table_header *head;
 
 	truncate_inode_pages(&inode->i_data, 0);
 	end_writeback(inode);
@@ -38,8 +39,11 @@ static void proc_evict_inode(struct inode *inode)
 	de = PROC_I(inode)->pde;
 	if (de)
 		pde_put(de);
-	if (PROC_I(inode)->sysctl)
-		sysctl_head_put(PROC_I(inode)->sysctl);
+	head = PROC_I(inode)->sysctl;
+	if (head) {
+		rcu_assign_pointer(PROC_I(inode)->sysctl, NULL);
+		sysctl_head_put(head);
+	}
 }
 
 struct vfsmount *proc_mnt;
diff --git a/fs/proc/proc_devtree.c b/fs/proc/proc_devtree.c
index d9396a4fc7f..927cbd115e5 100644
--- a/fs/proc/proc_devtree.c
+++ b/fs/proc/proc_devtree.c
@@ -233,7 +233,7 @@ void __init proc_device_tree_init(void)
 		return;
 	root = of_find_node_by_path("/");
 	if (root == NULL) {
-		printk(KERN_ERR "/proc/device-tree: can't find root\n");
+		pr_debug("/proc/device-tree: can't find root\n");
 		return;
 	}
 	proc_device_tree_add_node(root, proc_device_tree);
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index 09a1f92a34e..8eb2522111c 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -408,15 +408,18 @@ static int proc_sys_compare(const struct dentry *parent,
 		const struct dentry *dentry, const struct inode *inode,
 		unsigned int len, const char *str, const struct qstr *name)
 {
+	struct ctl_table_header *head;
 	/* Although proc doesn't have negative dentries, rcu-walk means
 	 * that inode here can be NULL */
+	/* AV: can it, indeed? */
 	if (!inode)
-		return 0;
+		return 1;
 	if (name->len != len)
 		return 1;
 	if (memcmp(name->name, str, len))
 		return 1;
-	return !sysctl_is_seen(PROC_I(inode)->sysctl);
+	head = rcu_dereference(PROC_I(inode)->sysctl);
+	return !head || !sysctl_is_seen(head);
 }
 
 static const struct dentry_operations proc_sys_dentry_operations = {
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 0bae036831e..1bba24bad82 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -1593,8 +1593,13 @@ int reiserfs_encode_fh(struct dentry *dentry, __u32 * data, int *lenp,
 	struct inode *inode = dentry->d_inode;
 	int maxlen = *lenp;
 
-	if (maxlen < 3)
+	if (need_parent && (maxlen < 5)) {
+		*lenp = 5;
 		return 255;
+	} else if (maxlen < 3) {
+		*lenp = 3;
+		return 255;
+	}
 
 	data[0] = inode->i_ino;
 	data[1] = le32_to_cpu(INODE_PKEY(inode)->k_dir_id);
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index ba5f51ec345..4b2eb564fda 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -771,7 +771,7 @@ static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 					EMPTY_DIR_SIZE_V1 : EMPTY_DIR_SIZE,
 					dentry, inode, &security);
 	if (retval) {
-		dir->i_nlink--;
+		DEC_DIR_INODE_NLINK(dir)
 		goto out_failed;
 	}
 
@@ -1122,10 +1122,6 @@ static int reiserfs_link(struct dentry *old_dentry, struct inode *dir,
 		reiserfs_write_unlock(dir->i_sb);
 		return -EMLINK;
 	}
-	if (inode->i_nlink == 0) {
-		reiserfs_write_unlock(dir->i_sb);
-		return -ENOENT;
-	}
 
 	/* inc before scheduling so reiserfs_unlink knows we are here */
 	inc_nlink(inode);
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 3cfb2e93364..5c11ca82b78 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -978,8 +978,6 @@ int reiserfs_permission(struct inode *inode, int mask, unsigned int flags)
 
 static int xattr_hide_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
-	if (nd->flags & LOOKUP_RCU)
-		return -ECHILD;
 	return -EPERM;
 }
 
diff --git a/fs/stat.c b/fs/stat.c
index d5c61cf2b70..961039121cb 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -75,13 +75,16 @@ int vfs_fstatat(int dfd, const char __user *filename, struct kstat *stat,
 	int error = -EINVAL;
 	int lookup_flags = 0;
 
-	if ((flag & ~(AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT)) != 0)
+	if ((flag & ~(AT_SYMLINK_NOFOLLOW | AT_NO_AUTOMOUNT |
+		      AT_EMPTY_PATH)) != 0)
 		goto out;
 
 	if (!(flag & AT_SYMLINK_NOFOLLOW))
 		lookup_flags |= LOOKUP_FOLLOW;
 	if (flag & AT_NO_AUTOMOUNT)
 		lookup_flags |= LOOKUP_NO_AUTOMOUNT;
+	if (flag & AT_EMPTY_PATH)
+		lookup_flags |= LOOKUP_EMPTY;
 
 	error = user_path_at(dfd, filename, lookup_flags, &path);
 	if (error)
@@ -297,7 +300,7 @@ SYSCALL_DEFINE4(readlinkat, int, dfd, const char __user *, pathname,
 	if (bufsiz <= 0)
 		return -EINVAL;
 
-	error = user_path_at(dfd, pathname, 0, &path);
+	error = user_path_at(dfd, pathname, LOOKUP_EMPTY, &path);
 	if (!error) {
 		struct inode *inode = path.dentry->d_inode;
 
diff --git a/fs/statfs.c b/fs/statfs.c
index 30ea8c8a996..8244924dec5 100644
--- a/fs/statfs.c
+++ b/fs/statfs.c
@@ -73,149 +73,135 @@ int vfs_statfs(struct path *path, struct kstatfs *buf)
 }
 EXPORT_SYMBOL(vfs_statfs);
 
-static int do_statfs_native(struct path *path, struct statfs *buf)
+int user_statfs(const char __user *pathname, struct kstatfs *st)
 {
-	struct kstatfs st;
-	int retval;
+	struct path path;
+	int error = user_path(pathname, &path);
+	if (!error) {
+		error = vfs_statfs(&path, st);
+		path_put(&path);
+	}
+	return error;
+}
 
-	retval = vfs_statfs(path, &st);
-	if (retval)
-		return retval;
+int fd_statfs(int fd, struct kstatfs *st)
+{
+	struct file *file = fget(fd);
+	int error = -EBADF;
+	if (file) {
+		error = vfs_statfs(&file->f_path, st);
+		fput(file);
+	}
+	return error;
+}
 
-	if (sizeof(*buf) == sizeof(st))
-		memcpy(buf, &st, sizeof(st));
+static int do_statfs_native(struct kstatfs *st, struct statfs __user *p)
+{
+	struct statfs buf;
+
+	if (sizeof(buf) == sizeof(*st))
+		memcpy(&buf, st, sizeof(*st));
 	else {
-		if (sizeof buf->f_blocks == 4) {
-			if ((st.f_blocks | st.f_bfree | st.f_bavail |
-			     st.f_bsize | st.f_frsize) &
+		if (sizeof buf.f_blocks == 4) {
+			if ((st->f_blocks | st->f_bfree | st->f_bavail |
+			     st->f_bsize | st->f_frsize) &
 			    0xffffffff00000000ULL)
 				return -EOVERFLOW;
 			/*
 			 * f_files and f_ffree may be -1; it's okay to stuff
 			 * that into 32 bits
 			 */
-			if (st.f_files != -1 &&
-			    (st.f_files & 0xffffffff00000000ULL))
+			if (st->f_files != -1 &&
+			    (st->f_files & 0xffffffff00000000ULL))
 				return -EOVERFLOW;
-			if (st.f_ffree != -1 &&
-			    (st.f_ffree & 0xffffffff00000000ULL))
+			if (st->f_ffree != -1 &&
+			    (st->f_ffree & 0xffffffff00000000ULL))
 				return -EOVERFLOW;
 		}
 
-		buf->f_type = st.f_type;
-		buf->f_bsize = st.f_bsize;
-		buf->f_blocks = st.f_blocks;
-		buf->f_bfree = st.f_bfree;
-		buf->f_bavail = st.f_bavail;
-		buf->f_files = st.f_files;
-		buf->f_ffree = st.f_ffree;
-		buf->f_fsid = st.f_fsid;
-		buf->f_namelen = st.f_namelen;
-		buf->f_frsize = st.f_frsize;
-		buf->f_flags = st.f_flags;
-		memset(buf->f_spare, 0, sizeof(buf->f_spare));
+		buf.f_type = st->f_type;
+		buf.f_bsize = st->f_bsize;
+		buf.f_blocks = st->f_blocks;
+		buf.f_bfree = st->f_bfree;
+		buf.f_bavail = st->f_bavail;
+		buf.f_files = st->f_files;
+		buf.f_ffree = st->f_ffree;
+		buf.f_fsid = st->f_fsid;
+		buf.f_namelen = st->f_namelen;
+		buf.f_frsize = st->f_frsize;
+		buf.f_flags = st->f_flags;
+		memset(buf.f_spare, 0, sizeof(buf.f_spare));
 	}
+	if (copy_to_user(p, &buf, sizeof(buf)))
+		return -EFAULT;
 	return 0;
 }
 
-static int do_statfs64(struct path *path, struct statfs64 *buf)
+static int do_statfs64(struct kstatfs *st, struct statfs64 __user *p)
 {
-	struct kstatfs st;
-	int retval;
-
-	retval = vfs_statfs(path, &st);
-	if (retval)
-		return retval;
-
-	if (sizeof(*buf) == sizeof(st))
-		memcpy(buf, &st, sizeof(st));
+	struct statfs64 buf;
+	if (sizeof(buf) == sizeof(*st))
+		memcpy(&buf, st, sizeof(*st));
 	else {
-		buf->f_type = st.f_type;
-		buf->f_bsize = st.f_bsize;
-		buf->f_blocks = st.f_blocks;
-		buf->f_bfree = st.f_bfree;
-		buf->f_bavail = st.f_bavail;
-		buf->f_files = st.f_files;
-		buf->f_ffree = st.f_ffree;
-		buf->f_fsid = st.f_fsid;
-		buf->f_namelen = st.f_namelen;
-		buf->f_frsize = st.f_frsize;
-		buf->f_flags = st.f_flags;
-		memset(buf->f_spare, 0, sizeof(buf->f_spare));
+		buf.f_type = st->f_type;
+		buf.f_bsize = st->f_bsize;
+		buf.f_blocks = st->f_blocks;
+		buf.f_bfree = st->f_bfree;
+		buf.f_bavail = st->f_bavail;
+		buf.f_files = st->f_files;
+		buf.f_ffree = st->f_ffree;
+		buf.f_fsid = st->f_fsid;
+		buf.f_namelen = st->f_namelen;
+		buf.f_frsize = st->f_frsize;
+		buf.f_flags = st->f_flags;
+		memset(buf.f_spare, 0, sizeof(buf.f_spare));
 	}
+	if (copy_to_user(p, &buf, sizeof(buf)))
+		return -EFAULT;
 	return 0;
 }
 
 SYSCALL_DEFINE2(statfs, const char __user *, pathname, struct statfs __user *, buf)
 {
-	struct path path;
-	int error;
-
-	error = user_path(pathname, &path);
-	if (!error) {
-		struct statfs tmp;
-		error = do_statfs_native(&path, &tmp);
-		if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
-			error = -EFAULT;
-		path_put(&path);
-	}
+	struct kstatfs st;
+	int error = user_statfs(pathname, &st);
+	if (!error)
+		error = do_statfs_native(&st, buf);
 	return error;
 }
 
 SYSCALL_DEFINE3(statfs64, const char __user *, pathname, size_t, sz, struct statfs64 __user *, buf)
 {
-	struct path path;
-	long error;
-
+	struct kstatfs st;
+	int error;
 	if (sz != sizeof(*buf))
 		return -EINVAL;
-	error = user_path(pathname, &path);
-	if (!error) {
-		struct statfs64 tmp;
-		error = do_statfs64(&path, &tmp);
-		if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
-			error = -EFAULT;
-		path_put(&path);
-	}
+	error = user_statfs(pathname, &st);
+	if (!error)
+		error = do_statfs64(&st, buf);
 	return error;
 }
 
 SYSCALL_DEFINE2(fstatfs, unsigned int, fd, struct statfs __user *, buf)
 {
-	struct file *file;
-	struct statfs tmp;
-	int error;
-
-	error = -EBADF;
-	file = fget(fd);
-	if (!file)
-		goto out;
-	error = do_statfs_native(&file->f_path, &tmp);
-	if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
-		error = -EFAULT;
-	fput(file);
-out:
+	struct kstatfs st;
+	int error = fd_statfs(fd, &st);
+	if (!error)
+		error = do_statfs_native(&st, buf);
 	return error;
 }
 
 SYSCALL_DEFINE3(fstatfs64, unsigned int, fd, size_t, sz, struct statfs64 __user *, buf)
 {
-	struct file *file;
-	struct statfs64 tmp;
+	struct kstatfs st;
 	int error;
 
 	if (sz != sizeof(*buf))
 		return -EINVAL;
 
-	error = -EBADF;
-	file = fget(fd);
-	if (!file)
-		goto out;
-	error = do_statfs64(&file->f_path, &tmp);
-	if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
-		error = -EFAULT;
-	fput(file);
-out:
+	error = fd_statfs(fd, &st);
+	if (!error)
+		error = do_statfs64(&st, buf);
 	return error;
 }
 
diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c
index b427b1208c2..e474fbcf8bd 100644
--- a/fs/sysv/namei.c
+++ b/fs/sysv/namei.c
@@ -245,7 +245,6 @@ static int sysv_rename(struct inode * old_dir, struct dentry * old_dentry,
 		new_de = sysv_find_entry(new_dentry, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inode_inc_link_count(old_inode);
 		sysv_set_link(new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
@@ -257,18 +256,15 @@ static int sysv_rename(struct inode * old_dir, struct dentry * old_dentry,
 			if (new_dir->i_nlink >= SYSV_SB(new_dir->i_sb)->s_link_max)
 				goto out_dir;
 		}
-		inode_inc_link_count(old_inode);
 		err = sysv_add_link(new_dentry, old_inode);
-		if (err) {
-			inode_dec_link_count(old_inode);
+		if (err)
 			goto out_dir;
-		}
 		if (dir_de)
 			inode_inc_link_count(new_dir);
 	}
 
 	sysv_delete_entry(old_de, old_page);
-	inode_dec_link_count(old_inode);
+	mark_inode_dirty(old_inode);
 
 	if (dir_de) {
 		sysv_set_link(dir_de, dir_page, new_dir);
diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c
index 14f64b689d7..7217d67a80a 100644
--- a/fs/ubifs/dir.c
+++ b/fs/ubifs/dir.c
@@ -522,24 +522,6 @@ static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
 	ubifs_assert(mutex_is_locked(&dir->i_mutex));
 	ubifs_assert(mutex_is_locked(&inode->i_mutex));
 
-	/*
-	 * Return -ENOENT if we've raced with unlink and i_nlink is 0.  Doing
-	 * otherwise has the potential to corrupt the orphan inode list.
-	 *
-	 * Indeed, consider a scenario when 'vfs_link(dirA/fileA)' and
-	 * 'vfs_unlink(dirA/fileA, dirB/fileB)' race. 'vfs_link()' does not
-	 * lock 'dirA->i_mutex', so this is possible. Both of the functions
-	 * lock 'fileA->i_mutex' though. Suppose 'vfs_unlink()' wins, and takes
-	 * 'fileA->i_mutex' mutex first. Suppose 'fileA->i_nlink' is 1. In this
-	 * case 'ubifs_unlink()' will drop the last reference, and put 'inodeA'
-	 * to the list of orphans. After this, 'vfs_link()' will link
-	 * 'dirB/fileB' to 'inodeA'. This is a problem because, for example,
-	 * the subsequent 'vfs_unlink(dirB/fileB)' will add the same inode
-	 * to the list of orphans.
-	 */
-	 if (inode->i_nlink == 0)
-		 return -ENOENT;
-
 	err = dbg_check_synced_i_size(inode);
 	if (err)
 		return err;
diff --git a/fs/udf/namei.c b/fs/udf/namei.c
index 2be0f9eb86d..f1dce848ef9 100644
--- a/fs/udf/namei.c
+++ b/fs/udf/namei.c
@@ -32,6 +32,8 @@
 #include <linux/crc-itu-t.h>
 #include <linux/exportfs.h>
 
+enum { UDF_MAX_LINKS = 0xffff };
+
 static inline int udf_match(int len1, const unsigned char *name1, int len2,
 			    const unsigned char *name2)
 {
@@ -650,7 +652,7 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	struct udf_inode_info *iinfo;
 
 	err = -EMLINK;
-	if (dir->i_nlink >= (256 << sizeof(dir->i_nlink)) - 1)
+	if (dir->i_nlink >= UDF_MAX_LINKS)
 		goto out;
 
 	err = -EIO;
@@ -1034,9 +1036,8 @@ static int udf_link(struct dentry *old_dentry, struct inode *dir,
 	struct fileIdentDesc cfi, *fi;
 	int err;
 
-	if (inode->i_nlink >= (256 << sizeof(inode->i_nlink)) - 1) {
+	if (inode->i_nlink >= UDF_MAX_LINKS)
 		return -EMLINK;
-	}
 
 	fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err);
 	if (!fi) {
@@ -1131,9 +1132,7 @@ static int udf_rename(struct inode *old_dir, struct dentry *old_dentry,
 			goto end_rename;
 
 		retval = -EMLINK;
-		if (!new_inode &&
-			new_dir->i_nlink >=
-				(256 << sizeof(new_dir->i_nlink)) - 1)
+		if (!new_inode && new_dir->i_nlink >= UDF_MAX_LINKS)
 			goto end_rename;
 	}
 	if (!nfi) {
@@ -1287,8 +1286,13 @@ static int udf_encode_fh(struct dentry *de, __u32 *fh, int *lenp,
 	struct fid *fid = (struct fid *)fh;
 	int type = FILEID_UDF_WITHOUT_PARENT;
 
-	if (len < 3 || (connectable && len < 5))
+	if (connectable && (len < 5)) {
+		*lenp = 5;
 		return 255;
+	} else if (len < 3) {
+		*lenp = 3;
+		return 255;
+	}
 
 	*lenp = 3;
 	fid->udf.block = location.logicalBlockNum;
diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
index 12f39b9e443..d6f681535eb 100644
--- a/fs/ufs/namei.c
+++ b/fs/ufs/namei.c
@@ -306,7 +306,6 @@ static int ufs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		new_de = ufs_find_entry(new_dir, &new_dentry->d_name, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inode_inc_link_count(old_inode);
 		ufs_set_link(new_dir, new_de, new_page, old_inode);
 		new_inode->i_ctime = CURRENT_TIME_SEC;
 		if (dir_de)
@@ -318,12 +317,9 @@ static int ufs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			if (new_dir->i_nlink >= UFS_LINK_MAX)
 				goto out_dir;
 		}
-		inode_inc_link_count(old_inode);
 		err = ufs_add_link(new_dentry, old_inode);
-		if (err) {
-			inode_dec_link_count(old_inode);
+		if (err)
 			goto out_dir;
-		}
 		if (dir_de)
 			inode_inc_link_count(new_dir);
 	}
@@ -331,12 +327,11 @@ static int ufs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	/*
 	 * Like most other Unix systems, set the ctime for inodes on a
  	 * rename.
-	 * inode_dec_link_count() will mark the inode dirty.
 	 */
 	old_inode->i_ctime = CURRENT_TIME_SEC;
 
 	ufs_delete_entry(old_dir, old_de, old_page);
-	inode_dec_link_count(old_inode);
+	mark_inode_dirty(old_inode);
 
 	if (dir_de) {
 		ufs_set_link(old_inode, dir_de, dir_page, new_dir);
diff --git a/fs/xfs/linux-2.6/xfs_discard.c b/fs/xfs/linux-2.6/xfs_discard.c
index 05201ae719e..d61611c8801 100644
--- a/fs/xfs/linux-2.6/xfs_discard.c
+++ b/fs/xfs/linux-2.6/xfs_discard.c
@@ -152,6 +152,8 @@ xfs_ioc_trim(
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -XFS_ERROR(EPERM);
+	if (!blk_queue_discard(q))
+		return -XFS_ERROR(EOPNOTSUPP);
 	if (copy_from_user(&range, urange, sizeof(range)))
 		return -XFS_ERROR(EFAULT);
 
diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c
index fc0114da7fd..f4f878fc008 100644
--- a/fs/xfs/linux-2.6/xfs_export.c
+++ b/fs/xfs/linux-2.6/xfs_export.c
@@ -89,8 +89,10 @@ xfs_fs_encode_fh(
 	 * seven combinations work.  The real answer is "don't use v2".
 	 */
 	len = xfs_fileid_length(fileid_type);
-	if (*max_len < len)
+	if (*max_len < len) {
+		*max_len = len;
 		return 255;
+	}
 	*max_len = len;
 
 	switch (fileid_type) {
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index f5e2a19e0f8..0ca0e3c024d 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -695,14 +695,19 @@ xfs_ioc_fsgeometry_v1(
 	xfs_mount_t		*mp,
 	void			__user *arg)
 {
-	xfs_fsop_geom_v1_t	fsgeo;
+	xfs_fsop_geom_t         fsgeo;
 	int			error;
 
-	error = xfs_fs_geometry(mp, (xfs_fsop_geom_t *)&fsgeo, 3);
+	error = xfs_fs_geometry(mp, &fsgeo, 3);
 	if (error)
 		return -error;
 
-	if (copy_to_user(arg, &fsgeo, sizeof(fsgeo)))
+	/*
+	 * Caller should have passed an argument of type
+	 * xfs_fsop_geom_v1_t.  This is a proper subset of the
+	 * xfs_fsop_geom_t that xfs_fs_geometry() fills in.
+	 */
+	if (copy_to_user(arg, &fsgeo, sizeof(xfs_fsop_geom_v1_t)))
 		return -XFS_ERROR(EFAULT);
 	return 0;
 }
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index cec89dd5d7d..85668efb3e3 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -53,6 +53,9 @@ xfs_fs_geometry(
 	xfs_fsop_geom_t		*geo,
 	int			new_version)
 {
+
+	memset(geo, 0, sizeof(*geo));
+
 	geo->blocksize = mp->m_sb.sb_blocksize;
 	geo->rtextsize = mp->m_sb.sb_rextsize;
 	geo->agblocks = mp->m_sb.sb_agblocks;