Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6 into sh-latest

148 files changed, 5274 insertions, 4789 deletions

diff --git a/fs/block_dev.c b/fs/block_dev.c
index 5147bdd3b8e..257b00e9842 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1102,6 +1102,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 			if (!bdev->bd_part)
 				goto out_clear;
 
+			ret = 0;
 			if (disk->fops->open) {
 				ret = disk->fops->open(bdev, mode);
 				if (ret == -ERESTARTSYS) {
@@ -1118,9 +1119,18 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 					put_disk(disk);
 					goto restart;
 				}
-				if (ret)
-					goto out_clear;
 			}
+			/*
+			 * If the device is invalidated, rescan partition
+			 * if open succeeded or failed with -ENOMEDIUM.
+			 * The latter is necessary to prevent ghost
+			 * partitions on a removed medium.
+			 */
+			if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM))
+				rescan_partitions(disk, bdev);
+			if (ret)
+				goto out_clear;
+
 			if (!bdev->bd_openers) {
 				bd_set_size(bdev,(loff_t)get_capacity(disk)<<9);
 				bdi = blk_get_backing_dev_info(bdev);
@@ -1128,8 +1138,6 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 					bdi = &default_backing_dev_info;
 				bdev_inode_switch_bdi(bdev->bd_inode, bdi);
 			}
-			if (bdev->bd_invalidated)
-				rescan_partitions(disk, bdev);
 		} else {
 			struct block_device *whole;
 			whole = bdget_disk(disk, 0);
@@ -1153,13 +1161,14 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 		}
 	} else {
 		if (bdev->bd_contains == bdev) {
-			if (bdev->bd_disk->fops->open) {
+			ret = 0;
+			if (bdev->bd_disk->fops->open)
 				ret = bdev->bd_disk->fops->open(bdev, mode);
-				if (ret)
-					goto out_unlock_bdev;
-			}
-			if (bdev->bd_invalidated)
+			/* the same as first opener case, read comment there */
+			if (bdev->bd_invalidated && (!ret || ret == -ENOMEDIUM))
 				rescan_partitions(bdev->bd_disk, bdev);
+			if (ret)
+				goto out_unlock_bdev;
 		}
 		/* only one opener holds refs to the module and disk */
 		module_put(disk->fops->owner);
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index de34bfad9ec..44ea5b92e1b 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -178,16 +178,18 @@ static int btrfs_xattr_acl_set(struct dentry *dentry, const char *name,
 
 	if (value) {
 		acl = posix_acl_from_xattr(value, size);
-		if (acl == NULL) {
-			value = NULL;
-			size = 0;
-		} else if (IS_ERR(acl)) {
+		if (IS_ERR(acl))
 			return PTR_ERR(acl);
+
+		if (acl) {
+			ret = posix_acl_valid(acl);
+			if (ret)
+				goto out;
 		}
 	}
 
 	ret = btrfs_set_acl(NULL, dentry->d_inode, acl, type);
-
+out:
 	posix_acl_release(acl);
 
 	return ret;
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 3458b572554..8f4b81de3ae 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -718,7 +718,7 @@ struct btrfs_space_info {
 	u64 total_bytes;	/* total bytes in the space,
 				   this doesn't take mirrors into account */
 	u64 bytes_used;		/* total bytes used,
-				   this does't take mirrors into account */
+				   this doesn't take mirrors into account */
 	u64 bytes_pinned;	/* total bytes pinned, will be freed when the
 				   transaction finishes */
 	u64 bytes_reserved;	/* total bytes the allocator has reserved for
@@ -740,8 +740,10 @@ struct btrfs_space_info {
 	 */
 	unsigned long reservation_progress;
 
-	int full;		/* indicates that we cannot allocate any more
+	int full:1;		/* indicates that we cannot allocate any more
 				   chunks for this space */
+	int chunk_alloc:1;	/* set if we are allocating a chunk */
+
 	int force_alloc;	/* set if we need to force a chunk alloc for
 				   this space */
 
@@ -2576,6 +2578,11 @@ int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode,
 int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
 			      struct inode *inode, u64 start, u64 end);
 int btrfs_release_file(struct inode *inode, struct file *file);
+void btrfs_drop_pages(struct page **pages, size_t num_pages);
+int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode,
+		      struct page **pages, size_t num_pages,
+		      loff_t pos, size_t write_bytes,
+		      struct extent_state **cached);
 
 /* tree-defrag.c */
 int btrfs_defrag_leaves(struct btrfs_trans_handle *trans,
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 8f1d44ba332..228cf36ece8 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -2824,6 +2824,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
 
 	spin_lock(&delayed_refs->lock);
 	if (delayed_refs->num_entries == 0) {
+		spin_unlock(&delayed_refs->lock);
 		printk(KERN_INFO "delayed_refs has NO entry\n");
 		return ret;
 	}
@@ -3057,7 +3058,7 @@ static int btrfs_cleanup_transaction(struct btrfs_root *root)
 		btrfs_destroy_pinned_extent(root,
 					    root->fs_info->pinned_extents);
 
-		t->use_count = 0;
+		atomic_set(&t->use_count, 0);
 		list_del_init(&t->list);
 		memset(t, 0, sizeof(*t));
 		kmem_cache_free(btrfs_transaction_cachep, t);
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index f619c3cb13b..9ee6bd55e16 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -33,6 +33,25 @@
 #include "locking.h"
 #include "free-space-cache.h"
 
+/* control flags for do_chunk_alloc's force field
+ * CHUNK_ALLOC_NO_FORCE means to only allocate a chunk
+ * if we really need one.
+ *
+ * CHUNK_ALLOC_FORCE means it must try to allocate one
+ *
+ * CHUNK_ALLOC_LIMITED means to only try and allocate one
+ * if we have very few chunks already allocated.  This is
+ * used as part of the clustering code to help make sure
+ * we have a good pool of storage to cluster in, without
+ * filling the FS with empty chunks
+ *
+ */
+enum {
+	CHUNK_ALLOC_NO_FORCE = 0,
+	CHUNK_ALLOC_FORCE = 1,
+	CHUNK_ALLOC_LIMITED = 2,
+};
+
 static int update_block_group(struct btrfs_trans_handle *trans,
 			      struct btrfs_root *root,
 			      u64 bytenr, u64 num_bytes, int alloc);
@@ -3019,7 +3038,8 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags,
 	found->bytes_readonly = 0;
 	found->bytes_may_use = 0;
 	found->full = 0;
-	found->force_alloc = 0;
+	found->force_alloc = CHUNK_ALLOC_NO_FORCE;
+	found->chunk_alloc = 0;
 	*space_info = found;
 	list_add_rcu(&found->list, &info->space_info);
 	atomic_set(&found->caching_threads, 0);
@@ -3150,7 +3170,7 @@ again:
 		if (!data_sinfo->full && alloc_chunk) {
 			u64 alloc_target;
 
-			data_sinfo->force_alloc = 1;
+			data_sinfo->force_alloc = CHUNK_ALLOC_FORCE;
 			spin_unlock(&data_sinfo->lock);
 alloc:
 			alloc_target = btrfs_get_alloc_profile(root, 1);
@@ -3160,7 +3180,8 @@ alloc:
 
 			ret = do_chunk_alloc(trans, root->fs_info->extent_root,
 					     bytes + 2 * 1024 * 1024,
-					     alloc_target, 0);
+					     alloc_target,
+					     CHUNK_ALLOC_NO_FORCE);
 			btrfs_end_transaction(trans, root);
 			if (ret < 0) {
 				if (ret != -ENOSPC)
@@ -3239,31 +3260,56 @@ static void force_metadata_allocation(struct btrfs_fs_info *info)
 	rcu_read_lock();
 	list_for_each_entry_rcu(found, head, list) {
 		if (found->flags & BTRFS_BLOCK_GROUP_METADATA)
-			found->force_alloc = 1;
+			found->force_alloc = CHUNK_ALLOC_FORCE;
 	}
 	rcu_read_unlock();
 }
 
 static int should_alloc_chunk(struct btrfs_root *root,
-			      struct btrfs_space_info *sinfo, u64 alloc_bytes)
+			      struct btrfs_space_info *sinfo, u64 alloc_bytes,
+			      int force)
 {
 	u64 num_bytes = sinfo->total_bytes - sinfo->bytes_readonly;
+	u64 num_allocated = sinfo->bytes_used + sinfo->bytes_reserved;
 	u64 thresh;
 
-	if (sinfo->bytes_used + sinfo->bytes_reserved +
-	    alloc_bytes + 256 * 1024 * 1024 < num_bytes)
+	if (force == CHUNK_ALLOC_FORCE)
+		return 1;
+
+	/*
+	 * in limited mode, we want to have some free space up to
+	 * about 1% of the FS size.
+	 */
+	if (force == CHUNK_ALLOC_LIMITED) {
+		thresh = btrfs_super_total_bytes(&root->fs_info->super_copy);
+		thresh = max_t(u64, 64 * 1024 * 1024,
+			       div_factor_fine(thresh, 1));
+
+		if (num_bytes - num_allocated < thresh)
+			return 1;
+	}
+
+	/*
+	 * we have two similar checks here, one based on percentage
+	 * and once based on a hard number of 256MB.  The idea
+	 * is that if we have a good amount of free
+	 * room, don't allocate a chunk.  A good mount is
+	 * less than 80% utilized of the chunks we have allocated,
+	 * or more than 256MB free
+	 */
+	if (num_allocated + alloc_bytes + 256 * 1024 * 1024 < num_bytes)
 		return 0;
 
-	if (sinfo->bytes_used + sinfo->bytes_reserved +
-	    alloc_bytes < div_factor(num_bytes, 8))
+	if (num_allocated + alloc_bytes < div_factor(num_bytes, 8))
 		return 0;
 
 	thresh = btrfs_super_total_bytes(&root->fs_info->super_copy);
+
+	/* 256MB or 5% of the FS */
 	thresh = max_t(u64, 256 * 1024 * 1024, div_factor_fine(thresh, 5));
 
 	if (num_bytes > thresh && sinfo->bytes_used < div_factor(num_bytes, 3))
 		return 0;
-
 	return 1;
 }
 
@@ -3273,10 +3319,9 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 {
 	struct btrfs_space_info *space_info;
 	struct btrfs_fs_info *fs_info = extent_root->fs_info;
+	int wait_for_alloc = 0;
 	int ret = 0;
 
-	mutex_lock(&fs_info->chunk_mutex);
-
 	flags = btrfs_reduce_alloc_profile(extent_root, flags);
 
 	space_info = __find_space_info(extent_root->fs_info, flags);
@@ -3287,21 +3332,40 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 	}
 	BUG_ON(!space_info);
 
+again:
 	spin_lock(&space_info->lock);
 	if (space_info->force_alloc)
-		force = 1;
+		force = space_info->force_alloc;
 	if (space_info->full) {
 		spin_unlock(&space_info->lock);
-		goto out;
+		return 0;
 	}
 
-	if (!force && !should_alloc_chunk(extent_root, space_info,
-					  alloc_bytes)) {
+	if (!should_alloc_chunk(extent_root, space_info, alloc_bytes, force)) {
 		spin_unlock(&space_info->lock);
-		goto out;
+		return 0;
+	} else if (space_info->chunk_alloc) {
+		wait_for_alloc = 1;
+	} else {
+		space_info->chunk_alloc = 1;
 	}
+
 	spin_unlock(&space_info->lock);
 
+	mutex_lock(&fs_info->chunk_mutex);
+
+	/*
+	 * The chunk_mutex is held throughout the entirety of a chunk
+	 * allocation, so once we've acquired the chunk_mutex we know that the
+	 * other guy is done and we need to recheck and see if we should
+	 * allocate.
+	 */
+	if (wait_for_alloc) {
+		mutex_unlock(&fs_info->chunk_mutex);
+		wait_for_alloc = 0;
+		goto again;
+	}
+
 	/*
 	 * If we have mixed data/metadata chunks we want to make sure we keep
 	 * allocating mixed chunks instead of individual chunks.
@@ -3327,9 +3391,10 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans,
 		space_info->full = 1;
 	else
 		ret = 1;
-	space_info->force_alloc = 0;
+
+	space_info->force_alloc = CHUNK_ALLOC_NO_FORCE;
+	space_info->chunk_alloc = 0;
 	spin_unlock(&space_info->lock);
-out:
 	mutex_unlock(&extent_root->fs_info->chunk_mutex);
 	return ret;
 }
@@ -5303,11 +5368,13 @@ loop:
 
 		if (allowed_chunk_alloc) {
 			ret = do_chunk_alloc(trans, root, num_bytes +
-					     2 * 1024 * 1024, data, 1);
+					     2 * 1024 * 1024, data,
+					     CHUNK_ALLOC_LIMITED);
 			allowed_chunk_alloc = 0;
 			done_chunk_alloc = 1;
-		} else if (!done_chunk_alloc) {
-			space_info->force_alloc = 1;
+		} else if (!done_chunk_alloc &&
+			   space_info->force_alloc == CHUNK_ALLOC_NO_FORCE) {
+			space_info->force_alloc = CHUNK_ALLOC_LIMITED;
 		}
 
 		if (loop < LOOP_NO_EMPTY_SIZE) {
@@ -5393,7 +5460,8 @@ again:
 	 */
 	if (empty_size || root->ref_cows)
 		ret = do_chunk_alloc(trans, root->fs_info->extent_root,
-				     num_bytes + 2 * 1024 * 1024, data, 0);
+				     num_bytes + 2 * 1024 * 1024, data,
+				     CHUNK_ALLOC_NO_FORCE);
 
 	WARN_ON(num_bytes < root->sectorsize);
 	ret = find_free_extent(trans, root, num_bytes, empty_size,
@@ -5405,7 +5473,7 @@ again:
 		num_bytes = num_bytes & ~(root->sectorsize - 1);
 		num_bytes = max(num_bytes, min_alloc_size);
 		do_chunk_alloc(trans, root->fs_info->extent_root,
-			       num_bytes, data, 1);
+			       num_bytes, data, CHUNK_ALLOC_FORCE);
 		goto again;
 	}
 	if (ret == -ENOSPC && btrfs_test_opt(root, ENOSPC_DEBUG)) {
@@ -7991,6 +8059,10 @@ static noinline int relocate_one_extent(struct btrfs_root *extent_root,
 				u64 group_start = group->key.objectid;
 				new_extents = kmalloc(sizeof(*new_extents),
 						      GFP_NOFS);
+				if (!new_extents) {
+					ret = -ENOMEM;
+					goto out;
+				}
 				nr_extents = 1;
 				ret = get_new_locations(reloc_inode,
 							extent_key,
@@ -8109,13 +8181,15 @@ int btrfs_set_block_group_ro(struct btrfs_root *root,
 
 	alloc_flags = update_block_group_flags(root, cache->flags);
 	if (alloc_flags != cache->flags)
-		do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, 1);
+		do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags,
+			       CHUNK_ALLOC_FORCE);
 
 	ret = set_block_group_ro(cache);
 	if (!ret)
 		goto out;
 	alloc_flags = get_alloc_profile(root, cache->space_info->flags);
-	ret = do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, 1);
+	ret = do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags,
+			     CHUNK_ALLOC_FORCE);
 	if (ret < 0)
 		goto out;
 	ret = set_block_group_ro(cache);
@@ -8128,7 +8202,8 @@ 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);
+	return do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags,
+			      CHUNK_ALLOC_FORCE);
 }
 
 /*
@@ -8781,23 +8856,38 @@ out:
 int btrfs_init_space_info(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_space_info *space_info;
+	struct btrfs_super_block *disk_super;
+	u64 features;
+	u64 flags;
+	int mixed = 0;
 	int ret;
 
-	ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_SYSTEM, 0, 0,
-								 &space_info);
-	if (ret)
-		return ret;
+	disk_super = &fs_info->super_copy;
+	if (!btrfs_super_root(disk_super))
+		return 1;
 
-	ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA, 0, 0,
-								 &space_info);
-	if (ret)
-		return ret;
+	features = btrfs_super_incompat_flags(disk_super);
+	if (features & BTRFS_FEATURE_INCOMPAT_MIXED_GROUPS)
+		mixed = 1;
 
-	ret = update_space_info(fs_info, BTRFS_BLOCK_GROUP_DATA, 0, 0,
-								 &space_info);
+	flags = BTRFS_BLOCK_GROUP_SYSTEM;
+	ret = update_space_info(fs_info, flags, 0, 0, &space_info);
 	if (ret)
-		return ret;
+		goto out;
 
+	if (mixed) {
+		flags = BTRFS_BLOCK_GROUP_METADATA | BTRFS_BLOCK_GROUP_DATA;
+		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+	} else {
+		flags = BTRFS_BLOCK_GROUP_METADATA;
+		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+		if (ret)
+			goto out;
+
+		flags = BTRFS_BLOCK_GROUP_DATA;
+		ret = update_space_info(fs_info, flags, 0, 0, &space_info);
+	}
+out:
 	return ret;
 }
 
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 20ddb28602a..96fcfa522da 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -10,6 +10,7 @@
 #include <linux/swap.h>
 #include <linux/writeback.h>
 #include <linux/pagevec.h>
+#include <linux/prefetch.h>
 #include "extent_io.h"
 #include "extent_map.h"
 #include "compat.h"
@@ -690,6 +691,15 @@ static void cache_state(struct extent_state *state,
 	}
 }
 
+static void uncache_state(struct extent_state **cached_ptr)
+{
+	if (cached_ptr && (*cached_ptr)) {
+		struct extent_state *state = *cached_ptr;
+		*cached_ptr = NULL;
+		free_extent_state(state);
+	}
+}
+
 /*
  * set some bits on a range in the tree.  This may require allocations or
  * sleeping, so the gfp mask is used to indicate what is allowed.
@@ -940,10 +950,10 @@ static int clear_extent_new(struct extent_io_tree *tree, u64 start, u64 end,
 }
 
 int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
-			gfp_t mask)
+			struct extent_state **cached_state, gfp_t mask)
 {
-	return set_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, NULL,
-			      NULL, mask);
+	return set_extent_bit(tree, start, end, EXTENT_UPTODATE, 0,
+			      NULL, cached_state, mask);
 }
 
 static int clear_extent_uptodate(struct extent_io_tree *tree, u64 start,
@@ -1012,8 +1022,7 @@ int unlock_extent_cached(struct extent_io_tree *tree, u64 start, u64 end,
 				mask);
 }
 
-int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end,
-		  gfp_t mask)
+int unlock_extent(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask)
 {
 	return clear_extent_bit(tree, start, end, EXTENT_LOCKED, 1, 0, NULL,
 				mask);
@@ -1735,6 +1744,9 @@ static void end_bio_extent_readpage(struct bio *bio, int err)
 
 	do {
 		struct page *page = bvec->bv_page;
+		struct extent_state *cached = NULL;
+		struct extent_state *state;
+
 		tree = &BTRFS_I(page->mapping->host)->io_tree;
 
 		start = ((u64)page->index << PAGE_CACHE_SHIFT) +
@@ -1749,9 +1761,20 @@ static void end_bio_extent_readpage(struct bio *bio, int err)
 		if (++bvec <= bvec_end)
 			prefetchw(&bvec->bv_page->flags);
 
+		spin_lock(&tree->lock);
+		state = find_first_extent_bit_state(tree, start, EXTENT_LOCKED);
+		if (state && state->start == start) {
+			/*
+			 * take a reference on the state, unlock will drop
+			 * the ref
+			 */
+			cache_state(state, &cached);
+		}
+		spin_unlock(&tree->lock);
+
 		if (uptodate && tree->ops && tree->ops->readpage_end_io_hook) {
 			ret = tree->ops->readpage_end_io_hook(page, start, end,
-							      NULL);
+							      state);
 			if (ret)
 				uptodate = 0;
 		}
@@ -1764,15 +1787,16 @@ static void end_bio_extent_readpage(struct bio *bio, int err)
 					test_bit(BIO_UPTODATE, &bio->bi_flags);
 				if (err)
 					uptodate = 0;
+				uncache_state(&cached);
 				continue;
 			}
 		}
 
 		if (uptodate) {
-			set_extent_uptodate(tree, start, end,
+			set_extent_uptodate(tree, start, end, &cached,
 					    GFP_ATOMIC);
 		}
-		unlock_extent(tree, start, end, GFP_ATOMIC);
+		unlock_extent_cached(tree, start, end, &cached, GFP_ATOMIC);
 
 		if (whole_page) {
 			if (uptodate) {
@@ -1811,6 +1835,7 @@ static void end_bio_extent_preparewrite(struct bio *bio, int err)
 
 	do {
 		struct page *page = bvec->bv_page;
+		struct extent_state *cached = NULL;
 		tree = &BTRFS_I(page->mapping->host)->io_tree;
 
 		start = ((u64)page->index << PAGE_CACHE_SHIFT) +
@@ -1821,13 +1846,14 @@ static void end_bio_extent_preparewrite(struct bio *bio, int err)
 			prefetchw(&bvec->bv_page->flags);
 
 		if (uptodate) {
-			set_extent_uptodate(tree, start, end, GFP_ATOMIC);
+			set_extent_uptodate(tree, start, end, &cached,
+					    GFP_ATOMIC);
 		} else {
 			ClearPageUptodate(page);
 			SetPageError(page);
 		}
 
-		unlock_extent(tree, start, end, GFP_ATOMIC);
+		unlock_extent_cached(tree, start, end, &cached, GFP_ATOMIC);
 
 	} while (bvec >= bio->bi_io_vec);
 
@@ -2016,14 +2042,17 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
 	while (cur <= end) {
 		if (cur >= last_byte) {
 			char *userpage;
+			struct extent_state *cached = NULL;
+
 			iosize = PAGE_CACHE_SIZE - page_offset;
 			userpage = kmap_atomic(page, KM_USER0);
 			memset(userpage + page_offset, 0, iosize);
 			flush_dcache_page(page);
 			kunmap_atomic(userpage, KM_USER0);
 			set_extent_uptodate(tree, cur, cur + iosize - 1,
-					    GFP_NOFS);
-			unlock_extent(tree, cur, cur + iosize - 1, GFP_NOFS);
+					    &cached, GFP_NOFS);
+			unlock_extent_cached(tree, cur, cur + iosize - 1,
+					     &cached, GFP_NOFS);
 			break;
 		}
 		em = get_extent(inode, page, page_offset, cur,
@@ -2063,14 +2092,17 @@ static int __extent_read_full_page(struct extent_io_tree *tree,
 		/* we've found a hole, just zero and go on */
 		if (block_start == EXTENT_MAP_HOLE) {
 			char *userpage;
+			struct extent_state *cached = NULL;
+
 			userpage = kmap_atomic(page, KM_USER0);
 			memset(userpage + page_offset, 0, iosize);
 			flush_dcache_page(page);
 			kunmap_atomic(userpage, KM_USER0);
 
 			set_extent_uptodate(tree, cur, cur + iosize - 1,
-					    GFP_NOFS);
-			unlock_extent(tree, cur, cur + iosize - 1, GFP_NOFS);
+					    &cached, GFP_NOFS);
+			unlock_extent_cached(tree, cur, cur + iosize - 1,
+			                     &cached, GFP_NOFS);
 			cur = cur + iosize;
 			page_offset += iosize;
 			continue;
@@ -2650,7 +2682,7 @@ int extent_readpages(struct extent_io_tree *tree,
 		prefetchw(&page->flags);
 		list_del(&page->lru);
 		if (!add_to_page_cache_lru(page, mapping,
-					page->index, GFP_KERNEL)) {
+					page->index, GFP_NOFS)) {
 			__extent_read_full_page(tree, page, get_extent,
 						&bio, 0, &bio_flags);
 		}
@@ -2789,9 +2821,12 @@ int extent_prepare_write(struct extent_io_tree *tree,
 			iocount++;
 			block_start = block_start + iosize;
 		} else {
-			set_extent_uptodate(tree, block_start, cur_end,
+			struct extent_state *cached = NULL;
+
+			set_extent_uptodate(tree, block_start, cur_end, &cached,
 					    GFP_NOFS);
-			unlock_extent(tree, block_start, cur_end, GFP_NOFS);
+			unlock_extent_cached(tree, block_start, cur_end,
+					     &cached, GFP_NOFS);
 			block_start = cur_end + 1;
 		}
 		page_offset = block_start & (PAGE_CACHE_SIZE - 1);
@@ -3457,7 +3492,7 @@ int set_extent_buffer_uptodate(struct extent_io_tree *tree,
 	num_pages = num_extent_pages(eb->start, eb->len);
 
 	set_extent_uptodate(tree, eb->start, eb->start + eb->len - 1,
-			    GFP_NOFS);
+			    NULL, GFP_NOFS);
 	for (i = 0; i < num_pages; i++) {
 		page = extent_buffer_page(eb, i);
 		if ((i == 0 && (eb->start & (PAGE_CACHE_SIZE - 1))) ||
@@ -3885,6 +3920,12 @@ static void move_pages(struct page *dst_page, struct page *src_page,
 	kunmap_atomic(dst_kaddr, KM_USER0);
 }
 
+static inline bool areas_overlap(unsigned long src, unsigned long dst, unsigned long len)
+{
+	unsigned long distance = (src > dst) ? src - dst : dst - src;
+	return distance < len;
+}
+
 static void copy_pages(struct page *dst_page, struct page *src_page,
 		       unsigned long dst_off, unsigned long src_off,
 		       unsigned long len)
@@ -3892,10 +3933,12 @@ static void copy_pages(struct page *dst_page, struct page *src_page,
 	char *dst_kaddr = kmap_atomic(dst_page, KM_USER0);
 	char *src_kaddr;
 
-	if (dst_page != src_page)
+	if (dst_page != src_page) {
 		src_kaddr = kmap_atomic(src_page, KM_USER1);
-	else
+	} else {
 		src_kaddr = dst_kaddr;
+		BUG_ON(areas_overlap(src_off, dst_off, len));
+	}
 
 	memcpy(dst_kaddr + dst_off, src_kaddr + src_off, len);
 	kunmap_atomic(dst_kaddr, KM_USER0);
@@ -3970,7 +4013,7 @@ void memmove_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset,
 		       "len %lu len %lu\n", dst_offset, len, dst->len);
 		BUG_ON(1);
 	}
-	if (dst_offset < src_offset) {
+	if (!areas_overlap(src_offset, dst_offset, len)) {
 		memcpy_extent_buffer(dst, dst_offset, src_offset, len);
 		return;
 	}
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index f62c5442835..af2d7179c37 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -208,7 +208,7 @@ int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
 		   int bits, int exclusive_bits, u64 *failed_start,
 		   struct extent_state **cached_state, gfp_t mask);
 int set_extent_uptodate(struct extent_io_tree *tree, u64 start, u64 end,
-			gfp_t mask);
+			struct extent_state **cached_state, gfp_t mask);
 int set_extent_new(struct extent_io_tree *tree, u64 start, u64 end,
 		   gfp_t mask);
 int set_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index e621ea54a3f..75899a01dde 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -104,7 +104,7 @@ static noinline int btrfs_copy_from_user(loff_t pos, int num_pages,
 /*
  * unlocks pages after btrfs_file_write is done with them
  */
-static noinline void btrfs_drop_pages(struct page **pages, size_t num_pages)
+void btrfs_drop_pages(struct page **pages, size_t num_pages)
 {
 	size_t i;
 	for (i = 0; i < num_pages; i++) {
@@ -127,16 +127,13 @@ static noinline void btrfs_drop_pages(struct page **pages, size_t num_pages)
  * this also makes the decision about creating an inline extent vs
  * doing real data extents, marking pages dirty and delalloc as required.
  */
-static noinline int dirty_and_release_pages(struct btrfs_root *root,
-					    struct file *file,
-					    struct page **pages,
-					    size_t num_pages,
-					    loff_t pos,
-					    size_t write_bytes)
+int btrfs_dirty_pages(struct btrfs_root *root, struct inode *inode,
+		      struct page **pages, size_t num_pages,
+		      loff_t pos, size_t write_bytes,
+		      struct extent_state **cached)
 {
 	int err = 0;
 	int i;
-	struct inode *inode = fdentry(file)->d_inode;
 	u64 num_bytes;
 	u64 start_pos;
 	u64 end_of_last_block;
@@ -149,7 +146,7 @@ static noinline int dirty_and_release_pages(struct btrfs_root *root,
 
 	end_of_last_block = start_pos + num_bytes - 1;
 	err = btrfs_set_extent_delalloc(inode, start_pos, end_of_last_block,
-					NULL);
+					cached);
 	if (err)
 		return err;
 
@@ -992,9 +989,9 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file,
 		}
 
 		if (copied > 0) {
-			ret = dirty_and_release_pages(root, file, pages,
-						      dirty_pages, pos,
-						      copied);
+			ret = btrfs_dirty_pages(root, inode, pages,
+						dirty_pages, pos, copied,
+						NULL);
 			if (ret) {
 				btrfs_delalloc_release_space(inode,
 					dirty_pages << PAGE_CACHE_SHIFT);
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index f561c953205..63731a1fb0a 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -508,6 +508,7 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 	struct inode *inode;
 	struct rb_node *node;
 	struct list_head *pos, *n;
+	struct page **pages;
 	struct page *page;
 	struct extent_state *cached_state = NULL;
 	struct btrfs_free_cluster *cluster = NULL;
@@ -517,13 +518,13 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 	u64 start, end, len;
 	u64 bytes = 0;
 	u32 *crc, *checksums;
-	pgoff_t index = 0, last_index = 0;
 	unsigned long first_page_offset;
-	int num_checksums;
+	int index = 0, num_pages = 0;
 	int entries = 0;
 	int bitmaps = 0;
 	int ret = 0;
 	bool next_page = false;
+	bool out_of_space = false;
 
 	root = root->fs_info->tree_root;
 
@@ -551,24 +552,31 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 		return 0;
 	}
 
-	last_index = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT;
+	num_pages = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >>
+		PAGE_CACHE_SHIFT;
 	filemap_write_and_wait(inode->i_mapping);
 	btrfs_wait_ordered_range(inode, inode->i_size &
 				 ~(root->sectorsize - 1), (u64)-1);
 
 	/* We need a checksum per page. */
-	num_checksums = i_size_read(inode) / PAGE_CACHE_SIZE;
-	crc = checksums  = kzalloc(sizeof(u32) * num_checksums, GFP_NOFS);
+	crc = checksums = kzalloc(sizeof(u32) * num_pages, GFP_NOFS);
 	if (!crc) {
 		iput(inode);
 		return 0;
 	}
 
+	pages = kzalloc(sizeof(struct page *) * num_pages, GFP_NOFS);
+	if (!pages) {
+		kfree(crc);
+		iput(inode);
+		return 0;
+	}
+
 	/* Since the first page has all of our checksums and our generation we
 	 * need to calculate the offset into the page that we can start writing
 	 * our entries.
 	 */
-	first_page_offset = (sizeof(u32) * num_checksums) + sizeof(u64);
+	first_page_offset = (sizeof(u32) * num_pages) + sizeof(u64);
 
 	/* Get the cluster for this block_group if it exists */
 	if (!list_empty(&block_group->cluster_list))
@@ -590,20 +598,18 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 	 * after find_get_page at this point.  Just putting this here so people
 	 * know and don't freak out.
 	 */
-	while (index <= last_index) {
+	while (index < num_pages) {
 		page = grab_cache_page(inode->i_mapping, index);
 		if (!page) {
-			pgoff_t i = 0;
+			int i;
 
-			while (i < index) {
-				page = find_get_page(inode->i_mapping, i);
-				unlock_page(page);
-				page_cache_release(page);
-				page_cache_release(page);
-				i++;
+			for (i = 0; i < num_pages; i++) {
+				unlock_page(pages[i]);
+				page_cache_release(pages[i]);
 			}
 			goto out_free;
 		}
+		pages[index] = page;
 		index++;
 	}
 
@@ -631,7 +637,12 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 			offset = start_offset;
 		}
 
-		page = find_get_page(inode->i_mapping, index);
+		if (index >= num_pages) {
+			out_of_space = true;
+			break;
+		}
+
+		page = pages[index];
 
 		addr = kmap(page);
 		entry = addr + start_offset;
@@ -708,23 +719,6 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 
 		bytes += PAGE_CACHE_SIZE;
 
-		ClearPageChecked(page);
-		set_page_extent_mapped(page);
-		SetPageUptodate(page);
-		set_page_dirty(page);
-
-		/*
-		 * We need to release our reference we got for grab_cache_page,
-		 * except for the first page which will hold our checksums, we
-		 * do that below.
-		 */
-		if (index != 0) {
-			unlock_page(page);
-			page_cache_release(page);
-		}
-
-		page_cache_release(page);
-
 		index++;
 	} while (node || next_page);
 
@@ -734,7 +728,11 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 		struct btrfs_free_space *entry =
 			list_entry(pos, struct btrfs_free_space, list);
 
-		page = find_get_page(inode->i_mapping, index);
+		if (index >= num_pages) {
+			out_of_space = true;
+			break;
+		}
+		page = pages[index];
 
 		addr = kmap(page);
 		memcpy(addr, entry->bitmap, PAGE_CACHE_SIZE);
@@ -745,64 +743,58 @@ int btrfs_write_out_cache(struct btrfs_root *root,
 		crc++;
 		bytes += PAGE_CACHE_SIZE;
 
-		ClearPageChecked(page);
-		set_page_extent_mapped(page);
-		SetPageUptodate(page);
-		set_page_dirty(page);
-		unlock_page(page);
-		page_cache_release(page);
-		page_cache_release(page);
 		list_del_init(&entry->list);
 		index++;
 	}
 
+	if (out_of_space) {
+		btrfs_drop_pages(pages, num_pages);
+		unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0,
+				     i_size_read(inode) - 1, &cached_state,
+				     GFP_NOFS);
+		ret = 0;
+		goto out_free;
+	}
+
 	/* Zero out the rest of the pages just to make sure */
-	while (index <= last_index) {
+	while (index < num_pages) {
 		void *addr;
 
-		page = find_get_page(inode->i_mapping, index);
-
+		page = pages[index];
 		addr = kmap(page);
 		memset(addr, 0, PAGE_CACHE_SIZE);
 		kunmap(page);
-		ClearPageChecked(page);
-		set_page_extent_mapped(page);
-		SetPageUptodate(page);
-		set_page_dirty(page);
-		unlock_page(page);
-		page_cache_release(page);
-		page_cache_release(page);
 		bytes += PAGE_CACHE_SIZE;
 		index++;
 	}
 
-	btrfs_set_extent_delalloc(inode, 0, bytes - 1, &cached_state);
-
 	/* Write the checksums and trans id to the first page */
 	{
 		void *addr;
 		u64 *gen;
 
-		page = find_get_page(inode->i_mapping, 0);
+		page = pages[0];
 
 		addr = kmap(page);
-		memcpy(addr, checksums, sizeof(u32) * num_checksums);
-		gen = addr + (sizeof(u32) * num_checksums);
+		memcpy(addr, checksums, sizeof(u32) * num_pages);
+		gen = addr + (sizeof(u32) * num_pages);
 		*gen = trans->transid;
 		kunmap(page);
-		ClearPageChecked(page);
-		set_page_extent_mapped(page);
-		SetPageUptodate(page);
-		set_page_dirty(page);
-		unlock_page(page);
-		page_cache_release(page);
-		page_cache_release(page);
 	}
-	BTRFS_I(inode)->generation = trans->transid;
 
+	ret = btrfs_dirty_pages(root, inode, pages, num_pages, 0,
+					    bytes, &cached_state);
+	btrfs_drop_pages(pages, num_pages);
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, 0,
 			     i_size_read(inode) - 1, &cached_state, GFP_NOFS);
 
+	if (ret) {
+		ret = 0;
+		goto out_free;
+	}
+
+	BTRFS_I(inode)->generation = trans->transid;
+
 	filemap_write_and_wait(inode->i_mapping);
 
 	key.objectid = BTRFS_FREE_SPACE_OBJECTID;
@@ -853,6 +845,7 @@ out_free:
 		BTRFS_I(inode)->generation = 0;
 	}
 	kfree(checksums);
+	kfree(pages);
 	btrfs_update_inode(trans, root, inode);
 	iput(inode);
 	return ret;
@@ -1775,10 +1768,13 @@ void btrfs_remove_free_space_cache(struct btrfs_block_group_cache *block_group)
 
 	while ((node = rb_last(&block_group->free_space_offset)) != NULL) {
 		info = rb_entry(node, struct btrfs_free_space, offset_index);
-		unlink_free_space(block_group, info);
-		if (info->bitmap)
-			kfree(info->bitmap);
-		kmem_cache_free(btrfs_free_space_cachep, info);
+		if (!info->bitmap) {
+			unlink_free_space(block_group, info);
+			kmem_cache_free(btrfs_free_space_cachep, info);
+		} else {
+			free_bitmap(block_group, info);
+		}
+
 		if (need_resched()) {
 			spin_unlock(&block_group->tree_lock);
 			cond_resched();
@@ -2308,7 +2304,7 @@ int btrfs_trim_block_group(struct btrfs_block_group_cache *block_group,
 			start = entry->offset;
 			bytes = min(entry->bytes, end - start);
 			unlink_free_space(block_group, entry);
-			kfree(entry);
+			kmem_cache_free(btrfs_free_space_cachep, entry);
 		}
 
 		spin_unlock(&block_group->tree_lock);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 5cc64ab9c48..7cd8ab0ef04 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -954,6 +954,7 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
 			 1, 0, NULL, GFP_NOFS);
 	while (start < end) {
 		async_cow = kmalloc(sizeof(*async_cow), GFP_NOFS);
+		BUG_ON(!async_cow);
 		async_cow->inode = inode;
 		async_cow->root = root;
 		async_cow->locked_page = locked_page;
@@ -1770,9 +1771,12 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 	add_pending_csums(trans, inode, ordered_extent->file_offset,
 			  &ordered_extent->list);
 
-	btrfs_ordered_update_i_size(inode, 0, ordered_extent);
-	ret = btrfs_update_inode(trans, root, inode);
-	BUG_ON(ret);
+	ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent);
+	if (!ret) {
+		ret = btrfs_update_inode(trans, root, inode);
+		BUG_ON(ret);
+	}
+	ret = 0;
 out:
 	if (nolock) {
 		if (trans)
@@ -2590,6 +2594,13 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
 			    struct btrfs_inode_item *item,
 			    struct inode *inode)
 {
+	if (!leaf->map_token)
+		map_private_extent_buffer(leaf, (unsigned long)item,
+					  sizeof(struct btrfs_inode_item),
+					  &leaf->map_token, &leaf->kaddr,
+					  &leaf->map_start, &leaf->map_len,
+					  KM_USER1);
+
 	btrfs_set_inode_uid(leaf, item, inode->i_uid);
 	btrfs_set_inode_gid(leaf, item, inode->i_gid);
 	btrfs_set_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size);
@@ -2618,6 +2629,11 @@ static void fill_inode_item(struct btrfs_trans_handle *trans,
 	btrfs_set_inode_rdev(leaf, item, inode->i_rdev);
 	btrfs_set_inode_flags(leaf, item, BTRFS_I(inode)->flags);
 	btrfs_set_inode_block_group(leaf, item, BTRFS_I(inode)->block_group);
+
+	if (leaf->map_token) {
+		unmap_extent_buffer(leaf, leaf->map_token, KM_USER1);
+		leaf->map_token = NULL;
+	}
 }
 
 /*
@@ -4207,10 +4223,8 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 	struct btrfs_key found_key;
 	struct btrfs_path *path;
 	int ret;
-	u32 nritems;
 	struct extent_buffer *leaf;
 	int slot;
-	int advance;
 	unsigned char d_type;
 	int over = 0;
 	u32 di_cur;
@@ -4253,27 +4267,19 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 	if (ret < 0)
 		goto err;
-	advance = 0;
 
 	while (1) {
 		leaf = path->nodes[0];
-		nritems = btrfs_header_nritems(leaf);
 		slot = path->slots[0];
-		if (advance || slot >= nritems) {
-			if (slot >= nritems - 1) {
-				ret = btrfs_next_leaf(root, path);
-				if (ret)
-					break;
-				leaf = path->nodes[0];
-				nritems = btrfs_header_nritems(leaf);
-				slot = path->slots[0];
-			} else {
-				slot++;
-				path->slots[0]++;
-			}
+		if (slot >= btrfs_header_nritems(leaf)) {
+			ret = btrfs_next_leaf(root, path);
+			if (ret < 0)
+				goto err;
+			else if (ret > 0)
+				break;
+			continue;
 		}
 
-		advance = 1;
 		item = btrfs_item_nr(leaf, slot);
 		btrfs_item_key_to_cpu(leaf, &found_key, slot);
 
@@ -4282,7 +4288,7 @@ static int btrfs_real_readdir(struct file *filp, void *dirent,
 		if (btrfs_key_type(&found_key) != key_type)
 			break;
 		if (found_key.offset < filp->f_pos)
-			continue;
+			goto next;
 
 		filp->f_pos = found_key.offset;
 
@@ -4335,6 +4341,8 @@ skip:
 			di_cur += di_len;
 			di = (struct btrfs_dir_item *)((char *)di + di_len);
 		}
+next:
+		path->slots[0]++;
 	}
 
 	/* Reached end of directory/root. Bump pos past the last item. */
@@ -4527,14 +4535,17 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans,
 	BUG_ON(!path);
 
 	inode = new_inode(root->fs_info->sb);
-	if (!inode)
+	if (!inode) {
+		btrfs_free_path(path);
 		return ERR_PTR(-ENOMEM);
+	}
 
 	if (dir) {
 		trace_btrfs_inode_request(dir);
 
 		ret = btrfs_set_inode_index(dir, index);
 		if (ret) {
+			btrfs_free_path(path);
 			iput(inode);
 			return ERR_PTR(ret);
 		}
@@ -4721,9 +4732,10 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
 	inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
 				dentry->d_name.len, dir->i_ino, objectid,
 				BTRFS_I(dir)->block_group, mode, &index);
-	err = PTR_ERR(inode);
-	if (IS_ERR(inode))
+	if (IS_ERR(inode)) {
+		err = PTR_ERR(inode);
 		goto out_unlock;
+	}
 
 	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
 	if (err) {
@@ -4782,9 +4794,10 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
 	inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name,
 				dentry->d_name.len, dir->i_ino, objectid,
 				BTRFS_I(dir)->block_group, mode, &index);
-	err = PTR_ERR(inode);
-	if (IS_ERR(inode))
+	if (IS_ERR(inode)) {
+		err = PTR_ERR(inode);
 		goto out_unlock;
+	}
 
 	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
 	if (err) {
@@ -4834,9 +4847,6 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	if (inode->i_nlink == ~0U)
 		return -EMLINK;
 
-	btrfs_inc_nlink(inode);
-	inode->i_ctime = CURRENT_TIME;
-
 	err = btrfs_set_inode_index(dir, &index);
 	if (err)
 		goto fail;
@@ -4852,6 +4862,9 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 		goto fail;
 	}
 
+	btrfs_inc_nlink(inode);
+	inode->i_ctime = CURRENT_TIME;
+
 	btrfs_set_trans_block_group(trans, dir);
 	ihold(inode);
 
@@ -4989,6 +5002,8 @@ static noinline int uncompress_inline(struct btrfs_path *path,
 	inline_size = btrfs_file_extent_inline_item_len(leaf,
 					btrfs_item_nr(leaf, path->slots[0]));
 	tmp = kmalloc(inline_size, GFP_NOFS);
+	if (!tmp)
+		return -ENOMEM;
 	ptr = btrfs_file_extent_inline_start(item);
 
 	read_extent_buffer(leaf, tmp, ptr, inline_size);
@@ -5221,7 +5236,7 @@ again:
 			btrfs_mark_buffer_dirty(leaf);
 		}
 		set_extent_uptodate(io_tree, em->start,
-				    extent_map_end(em) - 1, GFP_NOFS);
+				    extent_map_end(em) - 1, NULL, GFP_NOFS);
 		goto insert;
 	} else {
 		printk(KERN_ERR "btrfs unknown found_type %d\n", found_type);
@@ -5428,17 +5443,30 @@ out:
 }
 
 static struct extent_map *btrfs_new_extent_direct(struct inode *inode,
+						  struct extent_map *em,
 						  u64 start, u64 len)
 {
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_trans_handle *trans;
-	struct extent_map *em;
 	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
 	struct btrfs_key ins;
 	u64 alloc_hint;
 	int ret;
+	bool insert = false;
 
-	btrfs_drop_extent_cache(inode, start, start + len - 1, 0);
+	/*
+	 * Ok if the extent map we looked up is a hole and is for the exact
+	 * range we want, there is no reason to allocate a new one, however if
+	 * it is not right then we need to free this one and drop the cache for
+	 * our range.
+	 */
+	if (em->block_start != EXTENT_MAP_HOLE || em->start != start ||
+	    em->len != len) {
+		free_extent_map(em);
+		em = NULL;
+		insert = true;
+		btrfs_drop_extent_cache(inode, start, start + len - 1, 0);
+	}
 
 	trans = btrfs_join_transaction(root, 0);
 	if (IS_ERR(trans))
@@ -5454,10 +5482,12 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode,
 		goto out;
 	}
 
-	em = alloc_extent_map(GFP_NOFS);
 	if (!em) {
-		em = ERR_PTR(-ENOMEM);
-		goto out;
+		em = alloc_extent_map(GFP_NOFS);
+		if (!em) {
+			em = ERR_PTR(-ENOMEM);
+			goto out;
+		}
 	}
 
 	em->start = start;
@@ -5467,9 +5497,15 @@ static struct extent_map *btrfs_new_extent_direct(struct inode *inode,
 	em->block_start = ins.objectid;
 	em->block_len = ins.offset;
 	em->bdev = root->fs_info->fs_devices->latest_bdev;
+
+	/*
+	 * We need to do this because if we're using the original em we searched
+	 * for, we could have EXTENT_FLAG_VACANCY set, and we don't want that.
+	 */
+	em->flags = 0;
 	set_bit(EXTENT_FLAG_PINNED, &em->flags);
 
-	while (1) {
+	while (insert) {
 		write_lock(&em_tree->lock);
 		ret = add_extent_mapping(em_tree, em);
 		write_unlock(&em_tree->lock);
@@ -5687,8 +5723,7 @@ must_cow:
 	 * it above
 	 */
 	len = bh_result->b_size;
-	free_extent_map(em);
-	em = btrfs_new_extent_direct(inode, start, len);
+	em = btrfs_new_extent_direct(inode, em, start, len);
 	if (IS_ERR(em))
 		return PTR_ERR(em);
 	len = min(len, em->len - (start - em->start));
@@ -5851,8 +5886,10 @@ again:
 	}
 
 	add_pending_csums(trans, inode, ordered->file_offset, &ordered->list);
-	btrfs_ordered_update_i_size(inode, 0, ordered);
-	btrfs_update_inode(trans, root, inode);
+	ret = btrfs_ordered_update_i_size(inode, 0, ordered);
+	if (!ret)
+		btrfs_update_inode(trans, root, inode);
+	ret = 0;
 out_unlock:
 	unlock_extent_cached(&BTRFS_I(inode)->io_tree, ordered->file_offset,
 			     ordered->file_offset + ordered->len - 1,
@@ -5938,7 +5975,7 @@ static struct bio *btrfs_dio_bio_alloc(struct block_device *bdev,
 
 static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
 					 int rw, u64 file_offset, int skip_sum,
-					 u32 *csums)
+					 u32 *csums, int async_submit)
 {
 	int write = rw & REQ_WRITE;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -5949,13 +5986,24 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
 	if (ret)
 		goto err;
 
-	if (write && !skip_sum) {
+	if (skip_sum)
+		goto map;
+
+	if (write && async_submit) {
 		ret = btrfs_wq_submit_bio(root->fs_info,
 				   inode, rw, bio, 0, 0,
 				   file_offset,
 				   __btrfs_submit_bio_start_direct_io,
 				   __btrfs_submit_bio_done);
 		goto err;
+	} else if (write) {
+		/*
+		 * If we aren't doing async submit, calculate the csum of the
+		 * bio now.
+		 */
+		ret = btrfs_csum_one_bio(root, inode, bio, file_offset, 1);
+		if (ret)
+			goto err;
 	} else if (!skip_sum) {
 		ret = btrfs_lookup_bio_sums_dio(root, inode, bio,
 					  file_offset, csums);
@@ -5963,7 +6011,8 @@ static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode,
 			goto err;
 	}
 
-	ret = btrfs_map_bio(root, rw, bio, 0, 1);
+map:
+	ret = btrfs_map_bio(root, rw, bio, 0, async_submit);
 err:
 	bio_put(bio);
 	return ret;
@@ -5985,23 +6034,30 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 	int nr_pages = 0;
 	u32 *csums = dip->csums;
 	int ret = 0;
+	int async_submit = 0;
 	int write = rw & REQ_WRITE;
 
-	bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS);
-	if (!bio)
-		return -ENOMEM;
-	bio->bi_private = dip;
-	bio->bi_end_io = btrfs_end_dio_bio;
-	atomic_inc(&dip->pending_bios);
-
 	map_length = orig_bio->bi_size;
 	ret = btrfs_map_block(map_tree, READ, start_sector << 9,
 			      &map_length, NULL, 0);
 	if (ret) {
-		bio_put(bio);
+		bio_put(orig_bio);
 		return -EIO;
 	}
 
+	if (map_length >= orig_bio->bi_size) {
+		bio = orig_bio;
+		goto submit;
+	}
+
+	async_submit = 1;
+	bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS);
+	if (!bio)
+		return -ENOMEM;
+	bio->bi_private = dip;
+	bio->bi_end_io = btrfs_end_dio_bio;
+	atomic_inc(&dip->pending_bios);
+
 	while (bvec <= (orig_bio->bi_io_vec + orig_bio->bi_vcnt - 1)) {
 		if (unlikely(map_length < submit_len + bvec->bv_len ||
 		    bio_add_page(bio, bvec->bv_page, bvec->bv_len,
@@ -6015,7 +6071,7 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 			atomic_inc(&dip->pending_bios);
 			ret = __btrfs_submit_dio_bio(bio, inode, rw,
 						     file_offset, skip_sum,
-						     csums);
+						     csums, async_submit);
 			if (ret) {
 				bio_put(bio);
 				atomic_dec(&dip->pending_bios);
@@ -6052,8 +6108,9 @@ static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip,
 		}
 	}
 
+submit:
 	ret = __btrfs_submit_dio_bio(bio, inode, rw, file_offset, skip_sum,
-				     csums);
+				     csums, async_submit);
 	if (!ret)
 		return 0;
 
@@ -6148,6 +6205,7 @@ static ssize_t check_direct_IO(struct btrfs_root *root, int rw, struct kiocb *io
 			unsigned long nr_segs)
 {
 	int seg;
+	int i;
 	size_t size;
 	unsigned long addr;
 	unsigned blocksize_mask = root->sectorsize - 1;
@@ -6162,8 +6220,22 @@ static ssize_t check_direct_IO(struct btrfs_root *root, int rw, struct kiocb *io
 		addr = (unsigned long)iov[seg].iov_base;
 		size = iov[seg].iov_len;
 		end += size;
-		if ((addr & blocksize_mask) || (size & blocksize_mask)) 
+		if ((addr & blocksize_mask) || (size & blocksize_mask))
 			goto out;
+
+		/* If this is a write we don't need to check anymore */
+		if (rw & WRITE)
+			continue;
+
+		/*
+		 * Check to make sure we don't have duplicate iov_base's in this
+		 * iovec, if so return EINVAL, otherwise we'll get csum errors
+		 * when reading back.
+		 */
+		for (i = seg + 1; i < nr_segs; i++) {
+			if (iov[seg].iov_base == iov[i].iov_base)
+				goto out;
+		}
 	}
 	retval = 0;
 out:
@@ -7206,9 +7278,10 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 				dentry->d_name.len, dir->i_ino, objectid,
 				BTRFS_I(dir)->block_group, S_IFLNK|S_IRWXUGO,
 				&index);
-	err = PTR_ERR(inode);
-	if (IS_ERR(inode))
+	if (IS_ERR(inode)) {
+		err = PTR_ERR(inode);
 		goto out_unlock;
+	}
 
 	err = btrfs_init_inode_security(trans, inode, dir, &dentry->d_name);
 	if (err) {
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index cfc264fefdb..2616f7ed479 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -81,6 +81,13 @@ static unsigned int btrfs_flags_to_ioctl(unsigned int flags)
 		iflags |= FS_NOATIME_FL;
 	if (flags & BTRFS_INODE_DIRSYNC)
 		iflags |= FS_DIRSYNC_FL;
+	if (flags & BTRFS_INODE_NODATACOW)
+		iflags |= FS_NOCOW_FL;
+
+	if ((flags & BTRFS_INODE_COMPRESS) && !(flags & BTRFS_INODE_NOCOMPRESS))
+		iflags |= FS_COMPR_FL;
+	else if (flags & BTRFS_INODE_NOCOMPRESS)
+		iflags |= FS_NOCOMP_FL;
 
 	return iflags;
 }
@@ -144,16 +151,13 @@ static int check_flags(unsigned int flags)
 	if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \
 		      FS_NOATIME_FL | FS_NODUMP_FL | \
 		      FS_SYNC_FL | FS_DIRSYNC_FL | \
-		      FS_NOCOMP_FL | FS_COMPR_FL | \
-		      FS_NOCOW_FL | FS_COW_FL))
+		      FS_NOCOMP_FL | FS_COMPR_FL |
+		      FS_NOCOW_FL))
 		return -EOPNOTSUPP;
 
 	if ((flags & FS_NOCOMP_FL) && (flags & FS_COMPR_FL))
 		return -EINVAL;
 
-	if ((flags & FS_NOCOW_FL) && (flags & FS_COW_FL))
-		return -EINVAL;
-
 	return 0;
 }
 
@@ -218,6 +222,10 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 		ip->flags |= BTRFS_INODE_DIRSYNC;
 	else
 		ip->flags &= ~BTRFS_INODE_DIRSYNC;
+	if (flags & FS_NOCOW_FL)
+		ip->flags |= BTRFS_INODE_NODATACOW;
+	else
+		ip->flags &= ~BTRFS_INODE_NODATACOW;
 
 	/*
 	 * The COMPRESS flag can only be changed by users, while the NOCOMPRESS
@@ -230,11 +238,9 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg)
 	} else if (flags & FS_COMPR_FL) {
 		ip->flags |= BTRFS_INODE_COMPRESS;
 		ip->flags &= ~BTRFS_INODE_NOCOMPRESS;
+	} else {
+		ip->flags &= ~(BTRFS_INODE_COMPRESS | BTRFS_INODE_NOCOMPRESS);
 	}
-	if (flags & FS_NOCOW_FL)
-		ip->flags |= BTRFS_INODE_NODATACOW;
-	else if (flags & FS_COW_FL)
-		ip->flags &= ~BTRFS_INODE_NODATACOW;
 
 	trans = btrfs_join_transaction(root, 1);
 	BUG_ON(IS_ERR(trans));
@@ -2287,7 +2293,7 @@ long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg)
 	struct btrfs_ioctl_space_info space;
 	struct btrfs_ioctl_space_info *dest;
 	struct btrfs_ioctl_space_info *dest_orig;
-	struct btrfs_ioctl_space_info *user_dest;
+	struct btrfs_ioctl_space_info __user *user_dest;
 	struct btrfs_space_info *info;
 	u64 types[] = {BTRFS_BLOCK_GROUP_DATA,
 		       BTRFS_BLOCK_GROUP_SYSTEM,
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 58e7de9cc90..0ac712efcdf 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -159,7 +159,7 @@ enum {
 	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_enospc_debug, Opt_err,
+	Opt_enospc_debug, Opt_subvolrootid, Opt_err,
 };
 
 static match_table_t tokens = {
@@ -189,6 +189,7 @@ static match_table_t tokens = {
 	{Opt_clear_cache, "clear_cache"},
 	{Opt_user_subvol_rm_allowed, "user_subvol_rm_allowed"},
 	{Opt_enospc_debug, "enospc_debug"},
+	{Opt_subvolrootid, "subvolrootid=%d"},
 	{Opt_err, NULL},
 };
 
@@ -232,6 +233,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 			break;
 		case Opt_subvol:
 		case Opt_subvolid:
+		case Opt_subvolrootid:
 		case Opt_device:
 			/*
 			 * These are parsed by btrfs_parse_early_options
@@ -388,7 +390,7 @@ out:
  */
 static int btrfs_parse_early_options(const char *options, fmode_t flags,
 		void *holder, char **subvol_name, u64 *subvol_objectid,
-		struct btrfs_fs_devices **fs_devices)
+		u64 *subvol_rootid, struct btrfs_fs_devices **fs_devices)
 {
 	substring_t args[MAX_OPT_ARGS];
 	char *opts, *orig, *p;
@@ -429,6 +431,18 @@ static int btrfs_parse_early_options(const char *options, fmode_t flags,
 					*subvol_objectid = intarg;
 			}
 			break;
+		case Opt_subvolrootid:
+			intarg = 0;
+			error = match_int(&args[0], &intarg);
+			if (!error) {
+				/* we want the original fs_tree */
+				if (!intarg)
+					*subvol_rootid =
+						BTRFS_FS_TREE_OBJECTID;
+				else
+					*subvol_rootid = intarg;
+			}
+			break;
 		case Opt_device:
 			error = btrfs_scan_one_device(match_strdup(&args[0]),
 					flags, holder, fs_devices);
@@ -736,6 +750,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
 	fmode_t mode = FMODE_READ;
 	char *subvol_name = NULL;
 	u64 subvol_objectid = 0;
+	u64 subvol_rootid = 0;
 	int error = 0;
 
 	if (!(flags & MS_RDONLY))
@@ -743,7 +758,7 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
 
 	error = btrfs_parse_early_options(data, mode, fs_type,
 					  &subvol_name, &subvol_objectid,
-					  &fs_devices);
+					  &subvol_rootid, &fs_devices);
 	if (error)
 		return ERR_PTR(error);
 
@@ -807,15 +822,17 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
 		s->s_flags |= MS_ACTIVE;
 	}
 
-	root = get_default_root(s, subvol_objectid);
-	if (IS_ERR(root)) {
-		error = PTR_ERR(root);
-		deactivate_locked_super(s);
-		goto error_free_subvol_name;
-	}
 	/* if they gave us a subvolume name bind mount into that */
 	if (strcmp(subvol_name, ".")) {
 		struct dentry *new_root;
+
+		root = get_default_root(s, subvol_rootid);
+		if (IS_ERR(root)) {
+			error = PTR_ERR(root);
+			deactivate_locked_super(s);
+			goto error_free_subvol_name;
+		}
+
 		mutex_lock(&root->d_inode->i_mutex);
 		new_root = lookup_one_len(subvol_name, root,
 				      strlen(subvol_name));
@@ -836,6 +853,13 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags,
 		}
 		dput(root);
 		root = new_root;
+	} else {
+		root = get_default_root(s, subvol_objectid);
+		if (IS_ERR(root)) {
+			error = PTR_ERR(root);
+			deactivate_locked_super(s);
+			goto error_free_subvol_name;
+		}
 	}
 
 	kfree(subvol_name);
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 5b158da7e0b..c571734d5e5 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -32,10 +32,8 @@
 
 static noinline void put_transaction(struct btrfs_transaction *transaction)
 {
-	WARN_ON(transaction->use_count == 0);
-	transaction->use_count--;
-	if (transaction->use_count == 0) {
-		list_del_init(&transaction->list);
+	WARN_ON(atomic_read(&transaction->use_count) == 0);
+	if (atomic_dec_and_test(&transaction->use_count)) {
 		memset(transaction, 0, sizeof(*transaction));
 		kmem_cache_free(btrfs_transaction_cachep, transaction);
 	}
@@ -60,14 +58,14 @@ static noinline int join_transaction(struct btrfs_root *root)
 		if (!cur_trans)
 			return -ENOMEM;
 		root->fs_info->generation++;
-		cur_trans->num_writers = 1;
+		atomic_set(&cur_trans->num_writers, 1);
 		cur_trans->num_joined = 0;
 		cur_trans->transid = root->fs_info->generation;
 		init_waitqueue_head(&cur_trans->writer_wait);
 		init_waitqueue_head(&cur_trans->commit_wait);
 		cur_trans->in_commit = 0;
 		cur_trans->blocked = 0;
-		cur_trans->use_count = 1;
+		atomic_set(&cur_trans->use_count, 1);
 		cur_trans->commit_done = 0;
 		cur_trans->start_time = get_seconds();
 
@@ -88,7 +86,7 @@ static noinline int join_transaction(struct btrfs_root *root)
 		root->fs_info->running_transaction = cur_trans;
 		spin_unlock(&root->fs_info->new_trans_lock);
 	} else {
-		cur_trans->num_writers++;
+		atomic_inc(&cur_trans->num_writers);
 		cur_trans->num_joined++;
 	}
 
@@ -145,7 +143,7 @@ static void wait_current_trans(struct btrfs_root *root)
 	cur_trans = root->fs_info->running_transaction;
 	if (cur_trans && cur_trans->blocked) {
 		DEFINE_WAIT(wait);
-		cur_trans->use_count++;
+		atomic_inc(&cur_trans->use_count);
 		while (1) {
 			prepare_to_wait(&root->fs_info->transaction_wait, &wait,
 					TASK_UNINTERRUPTIBLE);
@@ -181,6 +179,7 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root,
 {
 	struct btrfs_trans_handle *h;
 	struct btrfs_transaction *cur_trans;
+	int retries = 0;
 	int ret;
 
 	if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR)
@@ -204,7 +203,7 @@ again:
 	}
 
 	cur_trans = root->fs_info->running_transaction;
-	cur_trans->use_count++;
+	atomic_inc(&cur_trans->use_count);
 	if (type != TRANS_JOIN_NOLOCK)
 		mutex_unlock(&root->fs_info->trans_mutex);
 
@@ -224,10 +223,18 @@ again:
 
 	if (num_items > 0) {
 		ret = btrfs_trans_reserve_metadata(h, root, num_items);
-		if (ret == -EAGAIN) {
+		if (ret == -EAGAIN && !retries) {
+			retries++;
 			btrfs_commit_transaction(h, root);
 			goto again;
+		} else if (ret == -EAGAIN) {
+			/*
+			 * We have already retried and got EAGAIN, so really we
+			 * don't have space, so set ret to -ENOSPC.
+			 */
+			ret = -ENOSPC;
 		}
+
 		if (ret < 0) {
 			btrfs_end_transaction(h, root);
 			return ERR_PTR(ret);
@@ -327,7 +334,7 @@ int btrfs_wait_for_commit(struct btrfs_root *root, u64 transid)
 			goto out_unlock;  /* nothing committing|committed */
 	}
 
-	cur_trans->use_count++;
+	atomic_inc(&cur_trans->use_count);
 	mutex_unlock(&root->fs_info->trans_mutex);
 
 	wait_for_commit(root, cur_trans);
@@ -457,18 +464,14 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 			wake_up_process(info->transaction_kthread);
 	}
 
-	if (lock)
-		mutex_lock(&info->trans_mutex);
 	WARN_ON(cur_trans != info->running_transaction);
-	WARN_ON(cur_trans->num_writers < 1);
-	cur_trans->num_writers--;
+	WARN_ON(atomic_read(&cur_trans->num_writers) < 1);
+	atomic_dec(&cur_trans->num_writers);
 
 	smp_mb();
 	if (waitqueue_active(&cur_trans->writer_wait))
 		wake_up(&cur_trans->writer_wait);
 	put_transaction(cur_trans);
-	if (lock)
-		mutex_unlock(&info->trans_mutex);
 
 	if (current->journal_info == trans)
 		current->journal_info = NULL;
@@ -1178,7 +1181,7 @@ int btrfs_commit_transaction_async(struct btrfs_trans_handle *trans,
 	/* take transaction reference */
 	mutex_lock(&root->fs_info->trans_mutex);
 	cur_trans = trans->transaction;
-	cur_trans->use_count++;
+	atomic_inc(&cur_trans->use_count);
 	mutex_unlock(&root->fs_info->trans_mutex);
 
 	btrfs_end_transaction(trans, root);
@@ -1237,7 +1240,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 	mutex_lock(&root->fs_info->trans_mutex);
 	if (cur_trans->in_commit) {
-		cur_trans->use_count++;
+		atomic_inc(&cur_trans->use_count);
 		mutex_unlock(&root->fs_info->trans_mutex);
 		btrfs_end_transaction(trans, root);
 
@@ -1259,7 +1262,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 		prev_trans = list_entry(cur_trans->list.prev,
 					struct btrfs_transaction, list);
 		if (!prev_trans->commit_done) {
-			prev_trans->use_count++;
+			atomic_inc(&prev_trans->use_count);
 			mutex_unlock(&root->fs_info->trans_mutex);
 
 			wait_for_commit(root, prev_trans);
@@ -1300,14 +1303,14 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 				TASK_UNINTERRUPTIBLE);
 
 		smp_mb();
-		if (cur_trans->num_writers > 1)
+		if (atomic_read(&cur_trans->num_writers) > 1)
 			schedule_timeout(MAX_SCHEDULE_TIMEOUT);
 		else if (should_grow)
 			schedule_timeout(1);
 
 		mutex_lock(&root->fs_info->trans_mutex);
 		finish_wait(&cur_trans->writer_wait, &wait);
-	} while (cur_trans->num_writers > 1 ||
+	} while (atomic_read(&cur_trans->num_writers) > 1 ||
 		 (should_grow && cur_trans->num_joined != joined));
 
 	ret = create_pending_snapshots(trans, root->fs_info);
@@ -1394,6 +1397,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 
 	wake_up(&cur_trans->commit_wait);
 
+	list_del_init(&cur_trans->list);
 	put_transaction(cur_trans);
 	put_transaction(cur_trans);
 
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 229a594cacd..e441acc6c58 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -27,11 +27,11 @@ struct btrfs_transaction {
 	 * total writers in this transaction, it must be zero before the
 	 * transaction can end
 	 */
-	unsigned long num_writers;
+	atomic_t num_writers;
 
 	unsigned long num_joined;
 	int in_commit;
-	int use_count;
+	atomic_t use_count;
 	int commit_done;
 	int blocked;
 	struct list_head list;
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index c50271ad315..f997ec0c1ba 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -2209,8 +2209,10 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
 
 	log = root->log_root;
 	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
+	if (!path) {
+		err = -ENOMEM;
+		goto out_unlock;
+	}
 
 	di = btrfs_lookup_dir_item(trans, log, path, dir->i_ino,
 				   name, name_len, -1);
@@ -2271,6 +2273,7 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans,
 	}
 fail:
 	btrfs_free_path(path);
+out_unlock:
 	mutex_unlock(&BTRFS_I(dir)->log_mutex);
 	if (ret == -ENOSPC) {
 		root->fs_info->last_trans_log_full_commit = trans->transid;
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 309a57b9fc8..c7367ae5a3e 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -155,6 +155,15 @@ static noinline int run_scheduled_bios(struct btrfs_device *device)
 	unsigned long limit;
 	unsigned long last_waited = 0;
 	int force_reg = 0;
+	struct blk_plug plug;
+
+	/*
+	 * this function runs all the bios we've collected for
+	 * a particular device.  We don't want to wander off to
+	 * another device without first sending all of these down.
+	 * So, setup a plug here and finish it off before we return
+	 */
+	blk_start_plug(&plug);
 
 	bdi = blk_get_backing_dev_info(device->bdev);
 	fs_info = device->dev_root->fs_info;
@@ -294,6 +303,7 @@ loop_lock:
 	spin_unlock(&device->io_lock);
 
 done:
+	blk_finish_plug(&plug);
 	return 0;
 }
 
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index a5303b871b1..cfd660550de 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -180,11 +180,10 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 	struct btrfs_path *path;
 	struct extent_buffer *leaf;
 	struct btrfs_dir_item *di;
-	int ret = 0, slot, advance;
+	int ret = 0, slot;
 	size_t total_size = 0, size_left = size;
 	unsigned long name_ptr;
 	size_t name_len;
-	u32 nritems;
 
 	/*
 	 * ok we want all objects associated with this id.
@@ -204,34 +203,24 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 	ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 	if (ret < 0)
 		goto err;
-	advance = 0;
+
 	while (1) {
 		leaf = path->nodes[0];
-		nritems = btrfs_header_nritems(leaf);
 		slot = path->slots[0];
 
 		/* this is where we start walking through the path */
-		if (advance || slot >= nritems) {
+		if (slot >= btrfs_header_nritems(leaf)) {
 			/*
 			 * if we've reached the last slot in this leaf we need
 			 * to go to the next leaf and reset everything
 			 */
-			if (slot >= nritems-1) {
-				ret = btrfs_next_leaf(root, path);
-				if (ret)
-					break;
-				leaf = path->nodes[0];
-				nritems = btrfs_header_nritems(leaf);
-				slot = path->slots[0];
-			} else {
-				/*
-				 * just walking through the slots on this leaf
-				 */
-				slot++;
-				path->slots[0]++;
-			}
+			ret = btrfs_next_leaf(root, path);
+			if (ret < 0)
+				goto err;
+			else if (ret > 0)
+				break;
+			continue;
 		}
-		advance = 1;
 
 		btrfs_item_key_to_cpu(leaf, &found_key, slot);
 
@@ -250,7 +239,7 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 
 		/* we are just looking for how big our buffer needs to be */
 		if (!size)
-			continue;
+			goto next;
 
 		if (!buffer || (name_len + 1) > size_left) {
 			ret = -ERANGE;
@@ -263,6 +252,8 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size)
 
 		size_left -= name_len + 1;
 		buffer += name_len + 1;
+next:
+		path->slots[0]++;
 	}
 	ret = total_size;
 
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index e159c529fd2..38b8ab55492 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -775,6 +775,13 @@ get_more_pages:
 					    ci->i_truncate_seq,
 					    ci->i_truncate_size,
 					    &inode->i_mtime, true, 1, 0);
+
+				if (!req) {
+					rc = -ENOMEM;
+					unlock_page(page);
+					break;
+				}
+
 				max_pages = req->r_num_pages;
 
 				alloc_page_vec(fsc, req);
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 5323c330bbf..2a5404c1c42 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -819,7 +819,7 @@ int __ceph_caps_used(struct ceph_inode_info *ci)
 		used |= CEPH_CAP_FILE_CACHE;
 	if (ci->i_wr_ref)
 		used |= CEPH_CAP_FILE_WR;
-	if (ci->i_wrbuffer_ref)
+	if (ci->i_wb_ref || ci->i_wrbuffer_ref)
 		used |= CEPH_CAP_FILE_BUFFER;
 	return used;
 }
@@ -1331,10 +1331,11 @@ static void ceph_flush_snaps(struct ceph_inode_info *ci)
 }
 
 /*
- * Mark caps dirty.  If inode is newly dirty, add to the global dirty
- * list.
+ * Mark caps dirty.  If inode is newly dirty, return the dirty flags.
+ * Caller is then responsible for calling __mark_inode_dirty with the
+ * returned flags value.
  */
-void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
+int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
 {
 	struct ceph_mds_client *mdsc =
 		ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
@@ -1357,7 +1358,7 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
 		list_add(&ci->i_dirty_item, &mdsc->cap_dirty);
 		spin_unlock(&mdsc->cap_dirty_lock);
 		if (ci->i_flushing_caps == 0) {
-			igrab(inode);
+			ihold(inode);
 			dirty |= I_DIRTY_SYNC;
 		}
 	}
@@ -1365,9 +1366,8 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
 	if (((was | ci->i_flushing_caps) & CEPH_CAP_FILE_BUFFER) &&
 	    (mask & CEPH_CAP_FILE_BUFFER))
 		dirty |= I_DIRTY_DATASYNC;
-	if (dirty)
-		__mark_inode_dirty(inode, dirty);
 	__cap_delay_requeue(mdsc, ci);
+	return dirty;
 }
 
 /*
@@ -1990,11 +1990,11 @@ static void __take_cap_refs(struct ceph_inode_info *ci, int got)
 	if (got & CEPH_CAP_FILE_WR)
 		ci->i_wr_ref++;
 	if (got & CEPH_CAP_FILE_BUFFER) {
-		if (ci->i_wrbuffer_ref == 0)
-			igrab(&ci->vfs_inode);
-		ci->i_wrbuffer_ref++;
-		dout("__take_cap_refs %p wrbuffer %d -> %d (?)\n",
-		     &ci->vfs_inode, ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref);
+		if (ci->i_wb_ref == 0)
+			ihold(&ci->vfs_inode);
+		ci->i_wb_ref++;
+		dout("__take_cap_refs %p wb %d -> %d (?)\n",
+		     &ci->vfs_inode, ci->i_wb_ref-1, ci->i_wb_ref);
 	}
 }
 
@@ -2169,12 +2169,12 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
 		if (--ci->i_rdcache_ref == 0)
 			last++;
 	if (had & CEPH_CAP_FILE_BUFFER) {
-		if (--ci->i_wrbuffer_ref == 0) {
+		if (--ci->i_wb_ref == 0) {
 			last++;
 			put++;
 		}
-		dout("put_cap_refs %p wrbuffer %d -> %d (?)\n",
-		     inode, ci->i_wrbuffer_ref+1, ci->i_wrbuffer_ref);
+		dout("put_cap_refs %p wb %d -> %d (?)\n",
+		     inode, ci->i_wb_ref+1, ci->i_wb_ref);
 	}
 	if (had & CEPH_CAP_FILE_WR)
 		if (--ci->i_wr_ref == 0) {
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 159b512d5a2..203252d88d9 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -734,9 +734,12 @@ retry_snap:
 		}
 	}
 	if (ret >= 0) {
+		int dirty;
 		spin_lock(&inode->i_lock);
-		__ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
+		dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
 		spin_unlock(&inode->i_lock);
+		if (dirty)
+			__mark_inode_dirty(inode, dirty);
 	}
 
 out:
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index b54c97da1c4..70b6a4839c3 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -355,6 +355,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 	ci->i_rd_ref = 0;
 	ci->i_rdcache_ref = 0;
 	ci->i_wr_ref = 0;
+	ci->i_wb_ref = 0;
 	ci->i_wrbuffer_ref = 0;
 	ci->i_wrbuffer_ref_head = 0;
 	ci->i_shared_gen = 0;
@@ -1567,6 +1568,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
 	int release = 0, dirtied = 0;
 	int mask = 0;
 	int err = 0;
+	int inode_dirty_flags = 0;
 
 	if (ceph_snap(inode) != CEPH_NOSNAP)
 		return -EROFS;
@@ -1725,13 +1727,16 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
 		dout("setattr %p ATTR_FILE ... hrm!\n", inode);
 
 	if (dirtied) {
-		__ceph_mark_dirty_caps(ci, dirtied);
+		inode_dirty_flags = __ceph_mark_dirty_caps(ci, dirtied);
 		inode->i_ctime = CURRENT_TIME;
 	}
 
 	release &= issued;
 	spin_unlock(&inode->i_lock);
 
+	if (inode_dirty_flags)
+		__mark_inode_dirty(inode, inode_dirty_flags);
+
 	if (mask) {
 		req->r_inode = igrab(inode);
 		req->r_inode_drop = release;
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index f60b07b0feb..d0fae4ce9ba 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -3304,8 +3304,8 @@ static void con_put(struct ceph_connection *con)
 {
 	struct ceph_mds_session *s = con->private;
 
+	dout("mdsc con_put %p (%d)\n", s, atomic_read(&s->s_ref) - 1);
 	ceph_put_mds_session(s);
-	dout("mdsc con_put %p (%d)\n", s, atomic_read(&s->s_ref));
 }
 
 /*
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index e86ec1155f8..24067d68a55 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -206,7 +206,7 @@ void ceph_put_snap_realm(struct ceph_mds_client *mdsc,
 		up_write(&mdsc->snap_rwsem);
 	} else {
 		spin_lock(&mdsc->snap_empty_lock);
-		list_add(&mdsc->snap_empty, &realm->empty_item);
+		list_add(&realm->empty_item, &mdsc->snap_empty);
 		spin_unlock(&mdsc->snap_empty_lock);
 	}
 }
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 619fe719968..f5cabefa98d 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -293,7 +293,7 @@ struct ceph_inode_info {
 
 	/* held references to caps */
 	int i_pin_ref;
-	int i_rd_ref, i_rdcache_ref, i_wr_ref;
+	int i_rd_ref, i_rdcache_ref, i_wr_ref, i_wb_ref;
 	int i_wrbuffer_ref, i_wrbuffer_ref_head;
 	u32 i_shared_gen;       /* increment each time we get FILE_SHARED */
 	u32 i_rdcache_gen;      /* incremented each time we get FILE_CACHE. */
@@ -506,7 +506,7 @@ static inline int __ceph_caps_dirty(struct ceph_inode_info *ci)
 {
 	return ci->i_dirty_caps | ci->i_flushing_caps;
 }
-extern void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask);
+extern int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask);
 
 extern int ceph_caps_revoking(struct ceph_inode_info *ci, int mask);
 extern int __ceph_caps_used(struct ceph_inode_info *ci);
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index 8c9eba6ef9d..f2b62869618 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -703,6 +703,7 @@ int ceph_setxattr(struct dentry *dentry, const char *name,
 	struct ceph_inode_xattr *xattr = NULL;
 	int issued;
 	int required_blob_size;
+	int dirty;
 
 	if (ceph_snap(inode) != CEPH_NOSNAP)
 		return -EROFS;
@@ -763,11 +764,12 @@ retry:
 	dout("setxattr %p issued %s\n", inode, ceph_cap_string(issued));
 	err = __set_xattr(ci, newname, name_len, newval,
 			  val_len, 1, 1, 1, &xattr);
-	__ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
+	dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
 	ci->i_xattrs.dirty = true;
 	inode->i_ctime = CURRENT_TIME;
 	spin_unlock(&inode->i_lock);
-
+	if (dirty)
+		__mark_inode_dirty(inode, dirty);
 	return err;
 
 do_sync:
@@ -810,6 +812,7 @@ int ceph_removexattr(struct dentry *dentry, const char *name)
 	struct ceph_vxattr_cb *vxattrs = ceph_inode_vxattrs(inode);
 	int issued;
 	int err;
+	int dirty;
 
 	if (ceph_snap(inode) != CEPH_NOSNAP)
 		return -EROFS;
@@ -833,12 +836,13 @@ int ceph_removexattr(struct dentry *dentry, const char *name)
 		goto do_sync;
 
 	err = __remove_xattr_by_name(ceph_inode(inode), name);
-	__ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
+	dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
 	ci->i_xattrs.dirty = true;
 	inode->i_ctime = CURRENT_TIME;
 
 	spin_unlock(&inode->i_lock);
-
+	if (dirty)
+		__mark_inode_dirty(inode, dirty);
 	return err;
 do_sync:
 	spin_unlock(&inode->i_lock);
diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig
index 7cb0f7f847e..75c47cd8d08 100644
--- a/fs/cifs/Kconfig
+++ b/fs/cifs/Kconfig
@@ -7,6 +7,7 @@ config CIFS
 	select CRYPTO_MD5
 	select CRYPTO_HMAC
 	select CRYPTO_ARC4
+	select CRYPTO_DES
 	help
 	  This is the client VFS module for the Common Internet File System
 	  (CIFS) protocol which is the successor to the Server Message Block
@@ -152,16 +153,28 @@ config CIFS_ACL
 	    Allows to fetch CIFS/NTFS ACL from the server.  The DACL blob
 	    is handed over to the application/caller.
 
-config CIFS_EXPERIMENTAL
-	  bool "CIFS Experimental Features (EXPERIMENTAL)"
+config CIFS_SMB2
+	bool "SMB2 network file system support (EXPERIMENTAL)"
+	depends on EXPERIMENTAL && INET && BROKEN
+	select NLS
+	select KEYS
+	select FSCACHE
+	select DNS_RESOLVER
+
+	help
+	  This enables experimental support for the SMB2 (Server Message Block
+	  version 2) protocol. The SMB2 protocol is the successor to the
+	  popular CIFS and SMB network file sharing protocols. SMB2 is the
+	  native file sharing mechanism for recent versions of Windows
+	  operating systems (since Vista).  SMB2 enablement will eventually
+	  allow users better performance, security and features, than would be
+	  possible with cifs. Note that smb2 mount options also are simpler
+	  (compared to cifs) due to protocol improvements.
+
+	  Unless you are a developer or tester, say N.
+
+config CIFS_NFSD_EXPORT
+	  bool "Allow nfsd to export CIFS file system (EXPERIMENTAL)"
 	  depends on CIFS && EXPERIMENTAL
 	  help
-	    Enables cifs features under testing. These features are
-	    experimental and currently include DFS support and directory
-	    change notification ie fcntl(F_DNOTIFY), as well as the upcall
-	    mechanism which will be used for Kerberos session negotiation
-	    and uid remapping.  Some of these features also may depend on
-	    setting a value of 1 to the pseudo-file /proc/fs/cifs/Experimental
-	    (which is disabled by default). See the file fs/cifs/README
-	    for more details.  If unsure, say N.
-
+	   Allows NFS server to export a CIFS mounted share (nfsd over cifs)
diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile
index d87558448e3..005d524c3a4 100644
--- a/fs/cifs/Makefile
+++ b/fs/cifs/Makefile
@@ -4,7 +4,7 @@
 obj-$(CONFIG_CIFS) += cifs.o
 
 cifs-y := cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o inode.o \
-	  link.o misc.o netmisc.o smbdes.o smbencrypt.o transport.o asn1.o \
+	  link.o misc.o netmisc.o smbencrypt.o transport.o asn1.o \
 	  cifs_unicode.o nterr.o xattr.o cifsencrypt.o \
 	  readdir.o ioctl.o sess.o export.o
 
diff --git a/fs/cifs/README b/fs/cifs/README
index 74ab165fc64..4a3ca0e5ca2 100644
--- a/fs/cifs/README
+++ b/fs/cifs/README
@@ -704,18 +704,6 @@ the start of smb requests and responses can be enabled via:
 
 	echo 1 > /proc/fs/cifs/traceSMB
 
-Two other experimental features are under development. To test these
-requires enabling CONFIG_CIFS_EXPERIMENTAL
-
-	cifsacl support needed to retrieve approximated mode bits based on
-		the contents on the CIFS ACL.
-
-	lease support: cifs will check the oplock state before calling into
-	the vfs to see if we can grant a lease on a file.
-
-	DNOTIFY fcntl: needed for support of directory change 
-			    notification and perhaps later for file leases)
-
 Per share (per client mount) statistics are available in /proc/fs/cifs/Stats
 if the kernel was configured with cifs statistics enabled.  The statistics
 represent the number of successful (ie non-zero return code from the server) 
diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c
index 30d01bc9085..18f4272d904 100644
--- a/fs/cifs/cifs_debug.c
+++ b/fs/cifs/cifs_debug.c
@@ -63,7 +63,7 @@ void cifs_dump_detail(struct smb_hdr *smb)
 	cERROR(1, "Cmd: %d Err: 0x%x Flags: 0x%x Flgs2: 0x%x Mid: %d Pid: %d",
 		  smb->Command, smb->Status.CifsError,
 		  smb->Flags, smb->Flags2, smb->Mid, smb->Pid);
-	cERROR(1, "smb buf %p len %d", smb, smbCalcSize_LE(smb));
+	cERROR(1, "smb buf %p len %d", smb, smbCalcSize(smb));
 }
 
 
diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h
index ac51cd2d33a..a9d5692e0c2 100644
--- a/fs/cifs/cifs_fs_sb.h
+++ b/fs/cifs/cifs_fs_sb.h
@@ -58,9 +58,7 @@ struct cifs_sb_info {
 	unsigned int mnt_cifs_flags;
 	int	prepathlen;
 	char   *prepath; /* relative path under the share to mount to */
-#ifdef CONFIG_CIFS_DFS_UPCALL
-	char   *mountdata; /* mount options received at mount time */
-#endif
+	char   *mountdata; /* options received at mount time or via DFS refs */
 	struct backing_dev_info bdi;
 	struct delayed_work prune_tlinks;
 };
diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c
index 23d43cde430..1b2e180b018 100644
--- a/fs/cifs/cifs_unicode.c
+++ b/fs/cifs/cifs_unicode.c
@@ -277,6 +277,7 @@ cifsConvertToUCS(__le16 *target, const char *source, int srclen,
 
 	for (i = 0, j = 0; i < srclen; j++) {
 		src_char = source[i];
+		charlen = 1;
 		switch (src_char) {
 		case 0:
 			put_unaligned(0, &target[j]);
@@ -316,16 +317,13 @@ cifsConvertToUCS(__le16 *target, const char *source, int srclen,
 				dst_char = cpu_to_le16(0x003f);
 				charlen = 1;
 			}
-			/*
-			 * character may take more than one byte in the source
-			 * string, but will take exactly two bytes in the
-			 * target string
-			 */
-			i += charlen;
-			continue;
 		}
+		/*
+		 * character may take more than one byte in the source string,
+		 * but will take exactly two bytes in the target string
+		 */
+		i += charlen;
 		put_unaligned(dst_char, &target[j]);
-		i++; /* move to next char in source string */
 	}
 
 ctoUCS_out:
diff --git a/fs/cifs/cifs_unicode.h b/fs/cifs/cifs_unicode.h
index 644dd882a56..6d02fd56056 100644
--- a/fs/cifs/cifs_unicode.h
+++ b/fs/cifs/cifs_unicode.h
@@ -82,6 +82,9 @@ int cifs_strtoUCS(__le16 *, const char *, int, const struct nls_table *);
 char *cifs_strndup_from_ucs(const char *src, const int maxlen,
 			    const bool is_unicode,
 			    const struct nls_table *codepage);
+extern int cifsConvertToUCS(__le16 *target, const char *source, int maxlen,
+			const struct nls_table *cp, int mapChars);
+
 #endif
 
 /*
diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c
index beeebf19423..f3c6fb9942a 100644
--- a/fs/cifs/cifsacl.c
+++ b/fs/cifs/cifsacl.c
@@ -23,24 +23,16 @@
 
 #include <linux/fs.h>
 #include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/keyctl.h>
+#include <linux/key-type.h>
+#include <keys/user-type.h>
 #include "cifspdu.h"
 #include "cifsglob.h"
 #include "cifsacl.h"
 #include "cifsproto.h"
 #include "cifs_debug.h"
 
-
-static struct cifs_wksid wksidarr[NUM_WK_SIDS] = {
-	{{1, 0, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0, 0} }, "null user"},
-	{{1, 1, {0, 0, 0, 0, 0, 1}, {0, 0, 0, 0, 0} }, "nobody"},
-	{{1, 1, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(11), 0, 0, 0, 0} }, "net-users"},
-	{{1, 1, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(18), 0, 0, 0, 0} }, "sys"},
-	{{1, 2, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(32), __constant_cpu_to_le32(544), 0, 0, 0} }, "root"},
-	{{1, 2, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(32), __constant_cpu_to_le32(545), 0, 0, 0} }, "users"},
-	{{1, 2, {0, 0, 0, 0, 0, 5}, {__constant_cpu_to_le32(32), __constant_cpu_to_le32(546), 0, 0, 0} }, "guest"} }
-;
-
-
 /* security id for everyone/world system group */
 static const struct cifs_sid sid_everyone = {
 	1, 1, {0, 0, 0, 0, 0, 1}, {0} };
@@ -50,50 +42,385 @@ static const struct cifs_sid sid_authusers = {
 /* group users */
 static const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {} };
 
+const struct cred *root_cred;
 
-int match_sid(struct cifs_sid *ctsid)
+static void
+shrink_idmap_tree(struct rb_root *root, int nr_to_scan, int *nr_rem,
+			int *nr_del)
 {
-	int i, j;
-	int num_subauth, num_sat, num_saw;
-	struct cifs_sid *cwsid;
+	struct rb_node *node;
+	struct rb_node *tmp;
+	struct cifs_sid_id *psidid;
+
+	node = rb_first(root);
+	while (node) {
+		tmp = node;
+		node = rb_next(tmp);
+		psidid = rb_entry(tmp, struct cifs_sid_id, rbnode);
+		if (nr_to_scan == 0 || *nr_del == nr_to_scan)
+			++(*nr_rem);
+		else {
+			if (time_after(jiffies, psidid->time + SID_MAP_EXPIRE)
+						&& psidid->refcount == 0) {
+				rb_erase(tmp, root);
+				++(*nr_del);
+			} else
+				++(*nr_rem);
+		}
+	}
+}
+
+/*
+ * Run idmap cache shrinker.
+ */
+static int
+cifs_idmap_shrinker(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask)
+{
+	int nr_del = 0;
+	int nr_rem = 0;
+	struct rb_root *root;
+
+	root = &uidtree;
+	spin_lock(&siduidlock);
+	shrink_idmap_tree(root, nr_to_scan, &nr_rem, &nr_del);
+	spin_unlock(&siduidlock);
+
+	root = &gidtree;
+	spin_lock(&sidgidlock);
+	shrink_idmap_tree(root, nr_to_scan, &nr_rem, &nr_del);
+	spin_unlock(&sidgidlock);
+
+	return nr_rem;
+}
+
+static struct shrinker cifs_shrinker = {
+	.shrink = cifs_idmap_shrinker,
+	.seeks = DEFAULT_SEEKS,
+};
+
+static int
+cifs_idmap_key_instantiate(struct key *key, const void *data, size_t datalen)
+{
+	char *payload;
+
+	payload = kmalloc(datalen, GFP_KERNEL);
+	if (!payload)
+		return -ENOMEM;
+
+	memcpy(payload, data, datalen);
+	key->payload.data = payload;
+	return 0;
+}
+
+static inline void
+cifs_idmap_key_destroy(struct key *key)
+{
+	kfree(key->payload.data);
+}
 
-	if (!ctsid)
-		return -1;
+struct key_type cifs_idmap_key_type = {
+	.name        = "cifs.idmap",
+	.instantiate = cifs_idmap_key_instantiate,
+	.destroy     = cifs_idmap_key_destroy,
+	.describe    = user_describe,
+	.match       = user_match,
+};
+
+static void
+sid_to_str(struct cifs_sid *sidptr, char *sidstr)
+{
+	int i;
+	unsigned long saval;
+	char *strptr;
 
-	for (i = 0; i < NUM_WK_SIDS; ++i) {
-		cwsid = &(wksidarr[i].cifssid);
+	strptr = sidstr;
 
-		/* compare the revision */
-		if (ctsid->revision != cwsid->revision)
-			continue;
+	sprintf(strptr, "%s", "S");
+	strptr = sidstr + strlen(sidstr);
 
-		/* compare all of the six auth values */
-		for (j = 0; j < 6; ++j) {
-			if (ctsid->authority[j] != cwsid->authority[j])
-				break;
+	sprintf(strptr, "-%d", sidptr->revision);
+	strptr = sidstr + strlen(sidstr);
+
+	for (i = 0; i < 6; ++i) {
+		if (sidptr->authority[i]) {
+			sprintf(strptr, "-%d", sidptr->authority[i]);
+			strptr = sidstr + strlen(sidstr);
 		}
-		if (j < 6)
-			continue; /* all of the auth values did not match */
-
-		/* compare all of the subauth values if any */
-		num_sat = ctsid->num_subauth;
-		num_saw = cwsid->num_subauth;
-		num_subauth = num_sat < num_saw ? num_sat : num_saw;
-		if (num_subauth) {
-			for (j = 0; j < num_subauth; ++j) {
-				if (ctsid->sub_auth[j] != cwsid->sub_auth[j])
-					break;
-			}
-			if (j < num_subauth)
-				continue; /* all sub_auth values do not match */
+	}
+
+	for (i = 0; i < sidptr->num_subauth; ++i) {
+		saval = le32_to_cpu(sidptr->sub_auth[i]);
+		sprintf(strptr, "-%ld", saval);
+		strptr = sidstr + strlen(sidstr);
+	}
+}
+
+static void
+id_rb_insert(struct rb_root *root, struct cifs_sid *sidptr,
+		struct cifs_sid_id **psidid, char *typestr)
+{
+	int rc;
+	char *strptr;
+	struct rb_node *node = root->rb_node;
+	struct rb_node *parent = NULL;
+	struct rb_node **linkto = &(root->rb_node);
+	struct cifs_sid_id *lsidid;
+
+	while (node) {
+		lsidid = rb_entry(node, struct cifs_sid_id, rbnode);
+		parent = node;
+		rc = compare_sids(sidptr, &((lsidid)->sid));
+		if (rc > 0) {
+			linkto = &(node->rb_left);
+			node = node->rb_left;
+		} else if (rc < 0) {
+			linkto = &(node->rb_right);
+			node = node->rb_right;
+		}
+	}
+
+	memcpy(&(*psidid)->sid, sidptr, sizeof(struct cifs_sid));
+	(*psidid)->time = jiffies - (SID_MAP_RETRY + 1);
+	(*psidid)->refcount = 0;
+
+	sprintf((*psidid)->sidstr, "%s", typestr);
+	strptr = (*psidid)->sidstr + strlen((*psidid)->sidstr);
+	sid_to_str(&(*psidid)->sid, strptr);
+
+	clear_bit(SID_ID_PENDING, &(*psidid)->state);
+	clear_bit(SID_ID_MAPPED, &(*psidid)->state);
+
+	rb_link_node(&(*psidid)->rbnode, parent, linkto);
+	rb_insert_color(&(*psidid)->rbnode, root);
+}
+
+static struct cifs_sid_id *
+id_rb_search(struct rb_root *root, struct cifs_sid *sidptr)
+{
+	int rc;
+	struct rb_node *node = root->rb_node;
+	struct cifs_sid_id *lsidid;
+
+	while (node) {
+		lsidid = rb_entry(node, struct cifs_sid_id, rbnode);
+		rc = compare_sids(sidptr, &((lsidid)->sid));
+		if (rc > 0) {
+			node = node->rb_left;
+		} else if (rc < 0) {
+			node = node->rb_right;
+		} else /* node found */
+			return lsidid;
+	}
+
+	return NULL;
+}
+
+static int
+sidid_pending_wait(void *unused)
+{
+	schedule();
+	return signal_pending(current) ? -ERESTARTSYS : 0;
+}
+
+static int
+sid_to_id(struct cifs_sb_info *cifs_sb, struct cifs_sid *psid,
+		struct cifs_fattr *fattr, uint sidtype)
+{
+	int rc;
+	unsigned long cid;
+	struct key *idkey;
+	const struct cred *saved_cred;
+	struct cifs_sid_id *psidid, *npsidid;
+	struct rb_root *cidtree;
+	spinlock_t *cidlock;
+
+	if (sidtype == SIDOWNER) {
+		cid = cifs_sb->mnt_uid; /* default uid, in case upcall fails */
+		cidlock = &siduidlock;
+		cidtree = &uidtree;
+	} else if (sidtype == SIDGROUP) {
+		cid = cifs_sb->mnt_gid; /* default gid, in case upcall fails */
+		cidlock = &sidgidlock;
+		cidtree = &gidtree;
+	} else
+		return -ENOENT;
+
+	spin_lock(cidlock);
+	psidid = id_rb_search(cidtree, psid);
+
+	if (!psidid) { /* node does not exist, allocate one & attempt adding */
+		spin_unlock(cidlock);
+		npsidid = kzalloc(sizeof(struct cifs_sid_id), GFP_KERNEL);
+		if (!npsidid)
+			return -ENOMEM;
+
+		npsidid->sidstr = kmalloc(SIDLEN, GFP_KERNEL);
+		if (!npsidid->sidstr) {
+			kfree(npsidid);
+			return -ENOMEM;
+		}
+
+		spin_lock(cidlock);
+		psidid = id_rb_search(cidtree, psid);
+		if (psidid) { /* node happened to get inserted meanwhile */
+			++psidid->refcount;
+			spin_unlock(cidlock);
+			kfree(npsidid->sidstr);
+			kfree(npsidid);
+		} else {
+			psidid = npsidid;
+			id_rb_insert(cidtree, psid, &psidid,
+					sidtype == SIDOWNER ? "os:" : "gs:");
+			++psidid->refcount;
+			spin_unlock(cidlock);
 		}
+	} else {
+		++psidid->refcount;
+		spin_unlock(cidlock);
+	}
+
+	/*
+	 * If we are here, it is safe to access psidid and its fields
+	 * since a reference was taken earlier while holding the spinlock.
+	 * A reference on the node is put without holding the spinlock
+	 * and it is OK to do so in this case, shrinker will not erase
+	 * this node until all references are put and we do not access
+	 * any fields of the node after a reference is put .
+	 */
+	if (test_bit(SID_ID_MAPPED, &psidid->state)) {
+		cid = psidid->id;
+		psidid->time = jiffies; /* update ts for accessing */
+		goto sid_to_id_out;
+	}
 
-		cFYI(1, "matching sid: %s\n", wksidarr[i].sidname);
-		return 0; /* sids compare/match */
+	if (time_after(psidid->time + SID_MAP_RETRY, jiffies))
+		goto sid_to_id_out;
+
+	if (!test_and_set_bit(SID_ID_PENDING, &psidid->state)) {
+		saved_cred = override_creds(root_cred);
+		idkey = request_key(&cifs_idmap_key_type, psidid->sidstr, "");
+		if (IS_ERR(idkey))
+			cFYI(1, "%s: Can't map SID to an id", __func__);
+		else {
+			cid = *(unsigned long *)idkey->payload.value;
+			psidid->id = cid;
+			set_bit(SID_ID_MAPPED, &psidid->state);
+			key_put(idkey);
+			kfree(psidid->sidstr);
+		}
+		revert_creds(saved_cred);
+		psidid->time = jiffies; /* update ts for accessing */
+		clear_bit(SID_ID_PENDING, &psidid->state);
+		wake_up_bit(&psidid->state, SID_ID_PENDING);
+	} else {
+		rc = wait_on_bit(&psidid->state, SID_ID_PENDING,
+				sidid_pending_wait, TASK_INTERRUPTIBLE);
+		if (rc) {
+			cFYI(1, "%s: sidid_pending_wait interrupted %d",
+					__func__, rc);
+			--psidid->refcount; /* decremented without spinlock */
+			return rc;
+		}
+		if (test_bit(SID_ID_MAPPED, &psidid->state))
+			cid = psidid->id;
 	}
 
-	cFYI(1, "No matching sid");
-	return -1;
+sid_to_id_out:
+	--psidid->refcount; /* decremented without spinlock */
+	if (sidtype == SIDOWNER)
+		fattr->cf_uid = cid;
+	else
+		fattr->cf_gid = cid;
+
+	return 0;
+}
+
+int
+init_cifs_idmap(void)
+{
+	struct cred *cred;
+	struct key *keyring;
+	int ret;
+
+	cFYI(1, "Registering the %s key type\n", cifs_idmap_key_type.name);
+
+	/* create an override credential set with a special thread keyring in
+	 * which requests are cached
+	 *
+	 * this is used to prevent malicious redirections from being installed
+	 * with add_key().
+	 */
+	cred = prepare_kernel_cred(NULL);
+	if (!cred)
+		return -ENOMEM;
+
+	keyring = key_alloc(&key_type_keyring, ".cifs_idmap", 0, 0, cred,
+			    (KEY_POS_ALL & ~KEY_POS_SETATTR) |
+			    KEY_USR_VIEW | KEY_USR_READ,
+			    KEY_ALLOC_NOT_IN_QUOTA);
+	if (IS_ERR(keyring)) {
+		ret = PTR_ERR(keyring);
+		goto failed_put_cred;
+	}
+
+	ret = key_instantiate_and_link(keyring, NULL, 0, NULL, NULL);
+	if (ret < 0)
+		goto failed_put_key;
+
+	ret = register_key_type(&cifs_idmap_key_type);
+	if (ret < 0)
+		goto failed_put_key;
+
+	/* instruct request_key() to use this special keyring as a cache for
+	 * the results it looks up */
+	cred->thread_keyring = keyring;
+	cred->jit_keyring = KEY_REQKEY_DEFL_THREAD_KEYRING;
+	root_cred = cred;
+
+	spin_lock_init(&siduidlock);
+	uidtree = RB_ROOT;
+	spin_lock_init(&sidgidlock);
+	gidtree = RB_ROOT;
+
+	register_shrinker(&cifs_shrinker);
+
+	cFYI(1, "cifs idmap keyring: %d\n", key_serial(keyring));
+	return 0;
+
+failed_put_key:
+	key_put(keyring);
+failed_put_cred:
+	put_cred(cred);
+	return ret;
+}
+
+void
+exit_cifs_idmap(void)
+{
+	key_revoke(root_cred->thread_keyring);
+	unregister_key_type(&cifs_idmap_key_type);
+	put_cred(root_cred);
+	unregister_shrinker(&cifs_shrinker);
+	cFYI(1, "Unregistered %s key type\n", cifs_idmap_key_type.name);
+}
+
+void
+cifs_destroy_idmaptrees(void)
+{
+	struct rb_root *root;
+	struct rb_node *node;
+
+	root = &uidtree;
+	spin_lock(&siduidlock);
+	while ((node = rb_first(root)))
+		rb_erase(node, root);
+	spin_unlock(&siduidlock);
+
+	root = &gidtree;
+	spin_lock(&sidgidlock);
+	while ((node = rb_first(root)))
+		rb_erase(node, root);
+	spin_unlock(&sidgidlock);
 }
 
 /* if the two SIDs (roughly equivalent to a UUID for a user or group) are
@@ -104,16 +431,24 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid)
 	int num_subauth, num_sat, num_saw;
 
 	if ((!ctsid) || (!cwsid))
-		return 0;
+		return 1;
 
 	/* compare the revision */
-	if (ctsid->revision != cwsid->revision)
-		return 0;
+	if (ctsid->revision != cwsid->revision) {
+		if (ctsid->revision > cwsid->revision)
+			return 1;
+		else
+			return -1;
+	}
 
 	/* compare all of the six auth values */
 	for (i = 0; i < 6; ++i) {
-		if (ctsid->authority[i] != cwsid->authority[i])
-			return 0;
+		if (ctsid->authority[i] != cwsid->authority[i]) {
+			if (ctsid->authority[i] > cwsid->authority[i])
+				return 1;
+			else
+				return -1;
+		}
 	}
 
 	/* compare all of the subauth values if any */
@@ -122,12 +457,16 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid)
 	num_subauth = num_sat < num_saw ? num_sat : num_saw;
 	if (num_subauth) {
 		for (i = 0; i < num_subauth; ++i) {
-			if (ctsid->sub_auth[i] != cwsid->sub_auth[i])
-				return 0;
+			if (ctsid->sub_auth[i] != cwsid->sub_auth[i]) {
+				if (ctsid->sub_auth[i] > cwsid->sub_auth[i])
+					return 1;
+				else
+					return -1;
+			}
 		}
 	}
 
-	return 1; /* sids compare/match */
+	return 0; /* sids compare/match */
 }
 
 
@@ -382,22 +721,22 @@ static void parse_dacl(struct cifs_acl *pdacl, char *end_of_acl,
 #ifdef CONFIG_CIFS_DEBUG2
 			dump_ace(ppace[i], end_of_acl);
 #endif
-			if (compare_sids(&(ppace[i]->sid), pownersid))
+			if (compare_sids(&(ppace[i]->sid), pownersid) == 0)
 				access_flags_to_mode(ppace[i]->access_req,
 						     ppace[i]->type,
 						     &fattr->cf_mode,
 						     &user_mask);
-			if (compare_sids(&(ppace[i]->sid), pgrpsid))
+			if (compare_sids(&(ppace[i]->sid), pgrpsid) == 0)
 				access_flags_to_mode(ppace[i]->access_req,
 						     ppace[i]->type,
 						     &fattr->cf_mode,
 						     &group_mask);
-			if (compare_sids(&(ppace[i]->sid), &sid_everyone))
+			if (compare_sids(&(ppace[i]->sid), &sid_everyone) == 0)
 				access_flags_to_mode(ppace[i]->access_req,
 						     ppace[i]->type,
 						     &fattr->cf_mode,
 						     &other_mask);
-			if (compare_sids(&(ppace[i]->sid), &sid_authusers))
+			if (compare_sids(&(ppace[i]->sid), &sid_authusers) == 0)
 				access_flags_to_mode(ppace[i]->access_req,
 						     ppace[i]->type,
 						     &fattr->cf_mode,
@@ -475,10 +814,10 @@ static int parse_sid(struct cifs_sid *psid, char *end_of_acl)
 
 
 /* Convert CIFS ACL to POSIX form */
-static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len,
-			  struct cifs_fattr *fattr)
+static int parse_sec_desc(struct cifs_sb_info *cifs_sb,
+		struct cifs_ntsd *pntsd, int acl_len, struct cifs_fattr *fattr)
 {
-	int rc;
+	int rc = 0;
 	struct cifs_sid *owner_sid_ptr, *group_sid_ptr;
 	struct cifs_acl *dacl_ptr; /* no need for SACL ptr */
 	char *end_of_acl = ((char *)pntsd) + acl_len;
@@ -500,12 +839,26 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len,
 		 le32_to_cpu(pntsd->sacloffset), dacloffset);
 /*	cifs_dump_mem("owner_sid: ", owner_sid_ptr, 64); */
 	rc = parse_sid(owner_sid_ptr, end_of_acl);
-	if (rc)
+	if (rc) {
+		cFYI(1, "%s: Error %d parsing Owner SID", __func__, rc);
+		return rc;
+	}
+	rc = sid_to_id(cifs_sb, owner_sid_ptr, fattr, SIDOWNER);
+	if (rc) {
+		cFYI(1, "%s: Error %d mapping Owner SID to uid", __func__, rc);
 		return rc;
+	}
 
 	rc = parse_sid(group_sid_ptr, end_of_acl);
-	if (rc)
+	if (rc) {
+		cFYI(1, "%s: Error %d mapping Owner SID to gid", __func__, rc);
 		return rc;
+	}
+	rc = sid_to_id(cifs_sb, group_sid_ptr, fattr, SIDGROUP);
+	if (rc) {
+		cFYI(1, "%s: Error %d mapping Group SID to gid", __func__, rc);
+		return rc;
+	}
 
 	if (dacloffset)
 		parse_dacl(dacl_ptr, end_of_acl, owner_sid_ptr,
@@ -520,7 +873,7 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len,
 	memcpy((void *)(&(cifscred->gsid)), (void *)group_sid_ptr,
 			sizeof(struct cifs_sid)); */
 
-	return 0;
+	return rc;
 }
 
 
@@ -688,7 +1041,7 @@ out:
 }
 
 /* Set an ACL on the server */
-static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
+int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 				struct inode *inode, const char *path)
 {
 	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
@@ -727,7 +1080,7 @@ cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, struct cifs_fattr *fattr,
 		rc = PTR_ERR(pntsd);
 		cERROR(1, "%s: error %d getting sec desc", __func__, rc);
 	} else {
-		rc = parse_sec_desc(pntsd, acllen, fattr);
+		rc = parse_sec_desc(cifs_sb, pntsd, acllen, fattr);
 		kfree(pntsd);
 		if (rc)
 			cERROR(1, "parse sec desc failed rc = %d", rc);
diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h
index c4ae7d03656..5c902c7ce52 100644
--- a/fs/cifs/cifsacl.h
+++ b/fs/cifs/cifsacl.h
@@ -39,6 +39,15 @@
 #define ACCESS_ALLOWED	0
 #define ACCESS_DENIED	1
 
+#define SIDOWNER 1
+#define SIDGROUP 2
+#define SIDLEN 150 /* S- 1 revision- 6 authorities- max 5 sub authorities */
+
+#define SID_ID_MAPPED 0
+#define SID_ID_PENDING 1
+#define SID_MAP_EXPIRE (3600 * HZ) /* map entry expires after one hour */
+#define SID_MAP_RETRY (300 * HZ)   /* wait 5 minutes for next attempt to map */
+
 struct cifs_ntsd {
 	__le16 revision; /* revision level */
 	__le16 type;
@@ -74,7 +83,21 @@ struct cifs_wksid {
 	char sidname[SIDNAMELENGTH];
 } __attribute__((packed));
 
-extern int match_sid(struct cifs_sid *);
+struct cifs_sid_id {
+	unsigned int refcount; /* increment with spinlock, decrement without */
+	unsigned long id;
+	unsigned long time;
+	unsigned long state;
+	char *sidstr;
+	struct rb_node rbnode;
+	struct cifs_sid sid;
+};
+
+#ifdef __KERNEL__
+extern struct key_type cifs_idmap_key_type;
+extern const struct cred *root_cred;
+#endif /* KERNEL */
+
 extern int compare_sids(const struct cifs_sid *, const struct cifs_sid *);
 
 #endif /* _CIFSACL_H */
diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c
index d1a016be73b..45c3f78c8f8 100644
--- a/fs/cifs/cifsencrypt.c
+++ b/fs/cifs/cifsencrypt.c
@@ -60,7 +60,7 @@ static int cifs_calculate_signature(const struct smb_hdr *cifs_pdu,
 		server->session_key.response, server->session_key.len);
 
 	crypto_shash_update(&server->secmech.sdescmd5->shash,
-		cifs_pdu->Protocol, cifs_pdu->smb_buf_length);
+		cifs_pdu->Protocol, be32_to_cpu(cifs_pdu->smb_buf_length));
 
 	rc = crypto_shash_final(&server->secmech.sdescmd5->shash, signature);
 
@@ -268,10 +268,11 @@ int setup_ntlm_response(struct cifsSesInfo *ses)
 }
 
 #ifdef CONFIG_CIFS_WEAK_PW_HASH
-void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt,
+int calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt,
 			char *lnm_session_key)
 {
 	int i;
+	int rc;
 	char password_with_pad[CIFS_ENCPWD_SIZE];
 
 	memset(password_with_pad, 0, CIFS_ENCPWD_SIZE);
@@ -282,7 +283,7 @@ void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt,
 		memset(lnm_session_key, 0, CIFS_SESS_KEY_SIZE);
 		memcpy(lnm_session_key, password_with_pad,
 			CIFS_ENCPWD_SIZE);
-		return;
+		return 0;
 	}
 
 	/* calculate old style session key */
@@ -299,10 +300,9 @@ void calc_lanman_hash(const char *password, const char *cryptkey, bool encrypt,
 	for (i = 0; i < CIFS_ENCPWD_SIZE; i++)
 		password_with_pad[i] = toupper(password_with_pad[i]);
 
-	SMBencrypt(password_with_pad, cryptkey, lnm_session_key);
+	rc = SMBencrypt(password_with_pad, cryptkey, lnm_session_key);
 
-	/* clear password before we return/free memory */
-	memset(password_with_pad, 0, CIFS_ENCPWD_SIZE);
+	return rc;
 }
 #endif /* CIFS_WEAK_PW_HASH */
 
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 5c412b33cd7..493b74ca564 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -128,29 +128,22 @@ cifs_read_super(struct super_block *sb, void *data,
 	}
 	cifs_sb->bdi.ra_pages = default_backing_dev_info.ra_pages;
 
-#ifdef CONFIG_CIFS_DFS_UPCALL
-	/* copy mount params to sb for use in submounts */
-	/* BB: should we move this after the mount so we
-	 * do not have to do the copy on failed mounts?
-	 * BB: May be it is better to do simple copy before
-	 * complex operation (mount), and in case of fail
-	 * just exit instead of doing mount and attempting
-	 * undo it if this copy fails?*/
+	/*
+	 * Copy mount params to sb for use in submounts. Better to do
+	 * the copy here and deal with the error before cleanup gets
+	 * complicated post-mount.
+	 */
 	if (data) {
-		int len = strlen(data);
-		cifs_sb->mountdata = kzalloc(len + 1, GFP_KERNEL);
+		cifs_sb->mountdata = kstrndup(data, PAGE_SIZE, GFP_KERNEL);
 		if (cifs_sb->mountdata == NULL) {
 			bdi_destroy(&cifs_sb->bdi);
 			kfree(sb->s_fs_info);
 			sb->s_fs_info = NULL;
 			return -ENOMEM;
 		}
-		strncpy(cifs_sb->mountdata, data, len + 1);
-		cifs_sb->mountdata[len] = '\0';
 	}
-#endif
 
-	rc = cifs_mount(sb, cifs_sb, data, devname);
+	rc = cifs_mount(sb, cifs_sb, devname);
 
 	if (rc) {
 		if (!silent)
@@ -163,7 +156,7 @@ cifs_read_super(struct super_block *sb, void *data,
 	sb->s_bdi = &cifs_sb->bdi;
 	sb->s_blocksize = CIFS_MAX_MSGSIZE;
 	sb->s_blocksize_bits = 14;	/* default 2**14 = CIFS_MAX_MSGSIZE */
-	inode = cifs_root_iget(sb, ROOT_I);
+	inode = cifs_root_iget(sb);
 
 	if (IS_ERR(inode)) {
 		rc = PTR_ERR(inode);
@@ -184,12 +177,12 @@ cifs_read_super(struct super_block *sb, void *data,
 	else
 		sb->s_d_op = &cifs_dentry_ops;
 
-#ifdef CONFIG_CIFS_EXPERIMENTAL
+#ifdef CIFS_NFSD_EXPORT
 	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) {
 		cFYI(1, "export ops supported");
 		sb->s_export_op = &cifs_export_ops;
 	}
-#endif /* EXPERIMENTAL */
+#endif /* CIFS_NFSD_EXPORT */
 
 	return 0;
 
@@ -202,12 +195,10 @@ out_no_root:
 
 out_mount_failed:
 	if (cifs_sb) {
-#ifdef CONFIG_CIFS_DFS_UPCALL
 		if (cifs_sb->mountdata) {
 			kfree(cifs_sb->mountdata);
 			cifs_sb->mountdata = NULL;
 		}
-#endif
 		unload_nls(cifs_sb->local_nls);
 		bdi_destroy(&cifs_sb->bdi);
 		kfree(cifs_sb);
@@ -231,12 +222,10 @@ cifs_put_super(struct super_block *sb)
 	rc = cifs_umount(sb, cifs_sb);
 	if (rc)
 		cERROR(1, "cifs_umount failed with return code %d", rc);
-#ifdef CONFIG_CIFS_DFS_UPCALL
 	if (cifs_sb->mountdata) {
 		kfree(cifs_sb->mountdata);
 		cifs_sb->mountdata = NULL;
 	}
-#endif
 
 	unload_nls(cifs_sb->local_nls);
 	bdi_destroy(&cifs_sb->bdi);
@@ -618,16 +607,31 @@ static loff_t cifs_llseek(struct file *file, loff_t offset, int origin)
 {
 	/* origin == SEEK_END => we must revalidate the cached file length */
 	if (origin == SEEK_END) {
-		int retval;
-
-		/* some applications poll for the file length in this strange
-		   way so we must seek to end on non-oplocked files by
-		   setting the revalidate time to zero */
-		CIFS_I(file->f_path.dentry->d_inode)->time = 0;
-
-		retval = cifs_revalidate_file(file);
-		if (retval < 0)
-			return (loff_t)retval;
+		int rc;
+		struct inode *inode = file->f_path.dentry->d_inode;
+
+		/*
+		 * We need to be sure that all dirty pages are written and the
+		 * server has the newest file length.
+		 */
+		if (!CIFS_I(inode)->clientCanCacheRead && inode->i_mapping &&
+		    inode->i_mapping->nrpages != 0) {
+			rc = filemap_fdatawait(inode->i_mapping);
+			if (rc) {
+				mapping_set_error(inode->i_mapping, rc);
+				return rc;
+			}
+		}
+		/*
+		 * Some applications poll for the file length in this strange
+		 * way so we must seek to end on non-oplocked files by
+		 * setting the revalidate time to zero.
+		 */
+		CIFS_I(inode)->time = 0;
+
+		rc = cifs_revalidate_file_attr(file);
+		if (rc < 0)
+			return (loff_t)rc;
 	}
 	return generic_file_llseek_unlocked(file, offset, origin);
 }
@@ -760,10 +764,11 @@ const struct file_operations cifs_file_strict_ops = {
 };
 
 const struct file_operations cifs_file_direct_ops = {
-	/* no aio, no readv -
-	   BB reevaluate whether they can be done with directio, no cache */
-	.read = cifs_user_read,
-	.write = cifs_user_write,
+	/* BB reevaluate whether they can be done with directio, no cache */
+	.read = do_sync_read,
+	.write = do_sync_write,
+	.aio_read = cifs_user_readv,
+	.aio_write = cifs_user_writev,
 	.open = cifs_open,
 	.release = cifs_close,
 	.lock = cifs_lock,
@@ -815,10 +820,11 @@ const struct file_operations cifs_file_strict_nobrl_ops = {
 };
 
 const struct file_operations cifs_file_direct_nobrl_ops = {
-	/* no mmap, no aio, no readv -
-	   BB reevaluate whether they can be done with directio, no cache */
-	.read = cifs_user_read,
-	.write = cifs_user_write,
+	/* BB reevaluate whether they can be done with directio, no cache */
+	.read = do_sync_read,
+	.write = do_sync_write,
+	.aio_read = cifs_user_readv,
+	.aio_write = cifs_user_writev,
 	.open = cifs_open,
 	.release = cifs_close,
 	.fsync = cifs_fsync,
@@ -981,10 +987,10 @@ init_cifs(void)
 	int rc = 0;
 	cifs_proc_init();
 	INIT_LIST_HEAD(&cifs_tcp_ses_list);
-#ifdef CONFIG_CIFS_EXPERIMENTAL
+#ifdef CONFIG_CIFS_DNOTIFY_EXPERIMENTAL /* unused temporarily */
 	INIT_LIST_HEAD(&GlobalDnotifyReqList);
 	INIT_LIST_HEAD(&GlobalDnotifyRsp_Q);
-#endif
+#endif /* was needed for dnotify, and will be needed for inotify when VFS fix */
 /*
  *  Initialize Global counters
  */
@@ -1033,22 +1039,33 @@ init_cifs(void)
 	if (rc)
 		goto out_destroy_mids;
 
-	rc = register_filesystem(&cifs_fs_type);
-	if (rc)
-		goto out_destroy_request_bufs;
 #ifdef CONFIG_CIFS_UPCALL
 	rc = register_key_type(&cifs_spnego_key_type);
 	if (rc)
-		goto out_unregister_filesystem;
-#endif
+		goto out_destroy_request_bufs;
+#endif /* CONFIG_CIFS_UPCALL */
+
+#ifdef CONFIG_CIFS_ACL
+	rc = init_cifs_idmap();
+	if (rc)
+		goto out_register_key_type;
+#endif /* CONFIG_CIFS_ACL */
+
+	rc = register_filesystem(&cifs_fs_type);
+	if (rc)
+		goto out_init_cifs_idmap;
 
 	return 0;
 
-#ifdef CONFIG_CIFS_UPCALL
-out_unregister_filesystem:
-	unregister_filesystem(&cifs_fs_type);
+out_init_cifs_idmap:
+#ifdef CONFIG_CIFS_ACL
+	exit_cifs_idmap();
+out_register_key_type:
 #endif
+#ifdef CONFIG_CIFS_UPCALL
+	unregister_key_type(&cifs_spnego_key_type);
 out_destroy_request_bufs:
+#endif
 	cifs_destroy_request_bufs();
 out_destroy_mids:
 	cifs_destroy_mids();
@@ -1070,6 +1087,10 @@ exit_cifs(void)
 #ifdef CONFIG_CIFS_DFS_UPCALL
 	cifs_dfs_release_automount_timer();
 #endif
+#ifdef CONFIG_CIFS_ACL
+	cifs_destroy_idmaptrees();
+	exit_cifs_idmap();
+#endif
 #ifdef CONFIG_CIFS_UPCALL
 	unregister_key_type(&cifs_spnego_key_type);
 #endif
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index a9371b6578c..64313f778eb 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -47,7 +47,7 @@ extern void cifs_sb_deactive(struct super_block *sb);
 
 /* Functions related to inodes */
 extern const struct inode_operations cifs_dir_inode_ops;
-extern struct inode *cifs_root_iget(struct super_block *, unsigned long);
+extern struct inode *cifs_root_iget(struct super_block *);
 extern int cifs_create(struct inode *, struct dentry *, int,
 		       struct nameidata *);
 extern struct dentry *cifs_lookup(struct inode *, struct dentry *,
@@ -59,9 +59,11 @@ extern int cifs_mkdir(struct inode *, struct dentry *, int);
 extern int cifs_rmdir(struct inode *, struct dentry *);
 extern int cifs_rename(struct inode *, struct dentry *, struct inode *,
 		       struct dentry *);
+extern int cifs_revalidate_file_attr(struct file *filp);
+extern int cifs_revalidate_dentry_attr(struct dentry *);
 extern int cifs_revalidate_file(struct file *filp);
 extern int cifs_revalidate_dentry(struct dentry *);
-extern void cifs_invalidate_mapping(struct inode *inode);
+extern int cifs_invalidate_mapping(struct inode *inode);
 extern int cifs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 extern int cifs_setattr(struct dentry *, struct iattr *);
 
@@ -80,12 +82,12 @@ extern const struct file_operations cifs_file_strict_nobrl_ops;
 extern int cifs_open(struct inode *inode, struct file *file);
 extern int cifs_close(struct inode *inode, struct file *file);
 extern int cifs_closedir(struct inode *inode, struct file *file);
-extern ssize_t cifs_user_read(struct file *file, char __user *read_data,
-			      size_t read_size, loff_t *poffset);
+extern ssize_t cifs_user_readv(struct kiocb *iocb, const struct iovec *iov,
+			       unsigned long nr_segs, loff_t pos);
 extern ssize_t cifs_strict_readv(struct kiocb *iocb, const struct iovec *iov,
 				 unsigned long nr_segs, loff_t pos);
-extern ssize_t cifs_user_write(struct file *file, const char __user *write_data,
-			       size_t write_size, loff_t *poffset);
+extern ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov,
+				unsigned long nr_segs, loff_t pos);
 extern ssize_t cifs_strict_writev(struct kiocb *iocb, const struct iovec *iov,
 				  unsigned long nr_segs, loff_t pos);
 extern int cifs_lock(struct file *, int, struct file_lock *);
@@ -123,9 +125,9 @@ extern ssize_t	cifs_getxattr(struct dentry *, const char *, void *, size_t);
 extern ssize_t	cifs_listxattr(struct dentry *, char *, size_t);
 extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
 
-#ifdef CONFIG_CIFS_EXPERIMENTAL
+#ifdef CIFS_NFSD_EXPORT
 extern const struct export_operations cifs_export_ops;
-#endif /* EXPERIMENTAL */
+#endif /* CIFS_NFSD_EXPORT */
 
-#define CIFS_VERSION   "1.71"
+#define CIFS_VERSION   "1.72"
 #endif				/* _CIFSFS_H */
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index a5d1106fcbd..76b4517e74b 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -274,7 +274,8 @@ struct cifsSesInfo {
 	int capabilities;
 	char serverName[SERVER_NAME_LEN_WITH_NULL * 2];	/* BB make bigger for
 				TCP names - will ipv6 and sctp addresses fit? */
-	char *user_name;
+	char *user_name;	/* must not be null except during init of sess
+				   and after mount option parsing we fill it */
 	char *domainName;
 	char *password;
 	struct session_key auth_key;
@@ -780,10 +781,12 @@ GLOBAL_EXTERN spinlock_t		cifs_tcp_ses_lock;
  */
 GLOBAL_EXTERN spinlock_t	cifs_file_list_lock;
 
+#ifdef CONFIG_CIFS_DNOTIFY_EXPERIMENTAL /* unused temporarily */
 /* Outstanding dir notify requests */
 GLOBAL_EXTERN struct list_head GlobalDnotifyReqList;
 /* DirNotify response queue */
 GLOBAL_EXTERN struct list_head GlobalDnotifyRsp_Q;
+#endif /* was needed for dnotify, and will be needed for inotify when VFS fix */
 
 /*
  * Global transaction id (XID) information
@@ -830,6 +833,11 @@ GLOBAL_EXTERN unsigned int cifs_max_pending; /* MAX requests at once to server*/
 /* reconnect after this many failed echo attempts */
 GLOBAL_EXTERN unsigned short echo_retries;
 
+GLOBAL_EXTERN struct rb_root uidtree;
+GLOBAL_EXTERN struct rb_root gidtree;
+GLOBAL_EXTERN spinlock_t siduidlock;
+GLOBAL_EXTERN spinlock_t sidgidlock;
+
 void cifs_oplock_break(struct work_struct *work);
 void cifs_oplock_break_get(struct cifsFileInfo *cfile);
 void cifs_oplock_break_put(struct cifsFileInfo *cfile);
diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h
index b5c8cc5d7a7..de3aa285de0 100644
--- a/fs/cifs/cifspdu.h
+++ b/fs/cifs/cifspdu.h
@@ -397,9 +397,9 @@
 #define GETU32(var)  (*((__u32 *)var))	/* BB check for endian issues */
 
 struct smb_hdr {
-	__u32 smb_buf_length;	/* big endian on wire *//* BB length is only two
-		or three bytes - with one or two byte type preceding it that are
-		zero - we could mask the type byte off just in case BB */
+	__be32 smb_buf_length;	/* BB length is only two (rarely three) bytes,
+		with one or two byte "type" preceding it that will be
+		zero - we could mask the type byte off */
 	__u8 Protocol[4];
 	__u8 Command;
 	union {
@@ -428,43 +428,28 @@ struct smb_hdr {
 	__u8 WordCount;
 } __attribute__((packed));
 
-/* given a pointer to an smb_hdr retrieve a char pointer to the byte count */
-#define BCC(smb_var) ((unsigned char *)(smb_var) + sizeof(struct smb_hdr) + \
-			 (2 * (smb_var)->WordCount))
+/* given a pointer to an smb_hdr, retrieve a void pointer to the ByteCount */
+static inline void *
+BCC(struct smb_hdr *smb)
+{
+	return (void *)smb + sizeof(*smb) + 2 * smb->WordCount;
+}
 
 /* given a pointer to an smb_hdr retrieve the pointer to the byte area */
 #define pByteArea(smb_var) (BCC(smb_var) + 2)
 
-/* get the converted ByteCount for a SMB packet and return it */
-static inline __u16
-get_bcc(struct smb_hdr *hdr)
-{
-	__u16 *bc_ptr = (__u16 *)BCC(hdr);
-
-	return get_unaligned(bc_ptr);
-}
-
 /* get the unconverted ByteCount for a SMB packet and return it */
 static inline __u16
-get_bcc_le(struct smb_hdr *hdr)
+get_bcc(struct smb_hdr *hdr)
 {
 	__le16 *bc_ptr = (__le16 *)BCC(hdr);
 
 	return get_unaligned_le16(bc_ptr);
 }
 
-/* set the ByteCount for a SMB packet in host-byte order */
-static inline void
-put_bcc(__u16 count, struct smb_hdr *hdr)
-{
-	__u16 *bc_ptr = (__u16 *)BCC(hdr);
-
-	put_unaligned(count, bc_ptr);
-}
-
 /* set the ByteCount for a SMB packet in little-endian */
 static inline void
-put_bcc_le(__u16 count, struct smb_hdr *hdr)
+put_bcc(__u16 count, struct smb_hdr *hdr)
 {
 	__le16 *bc_ptr = (__le16 *)BCC(hdr);
 
diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h
index 8096f27ad9a..6e69e06a30b 100644
--- a/fs/cifs/cifsproto.h
+++ b/fs/cifs/cifsproto.h
@@ -53,6 +53,9 @@ do {								\
 	cFYI(1, "CIFS VFS: leaving %s (xid = %d) rc = %d",	\
 	     __func__, curr_xid, (int)rc);			\
 } while (0)
+extern int init_cifs_idmap(void);
+extern void exit_cifs_idmap(void);
+extern void cifs_destroy_idmaptrees(void);
 extern char *build_path_from_dentry(struct dentry *);
 extern char *cifs_build_path_to_root(struct cifs_sb_info *cifs_sb,
 					struct cifsTconInfo *tcon);
@@ -90,7 +93,6 @@ extern void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset,
 extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, bool);
 extern struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *, bool);
 extern unsigned int smbCalcSize(struct smb_hdr *ptr);
-extern unsigned int smbCalcSize_LE(struct smb_hdr *ptr);
 extern int decode_negTokenInit(unsigned char *security_blob, int length,
 			struct TCP_Server_Info *server);
 extern int cifs_convert_address(struct sockaddr *dst, const char *src, int len);
@@ -143,8 +145,10 @@ extern int cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb,
 extern int mode_to_cifs_acl(struct inode *inode, const char *path, __u64);
 extern struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *, struct inode *,
 					const char *, u32 *);
+extern int set_cifs_acl(struct cifs_ntsd *, __u32, struct inode *,
+				const char *);
 
-extern int cifs_mount(struct super_block *, struct cifs_sb_info *, char *,
+extern int cifs_mount(struct super_block *, struct cifs_sb_info *,
 			const char *);
 extern int cifs_umount(struct super_block *, struct cifs_sb_info *);
 extern void cifs_dfs_release_automount_timer(void);
@@ -304,12 +308,13 @@ extern int CIFSSMBUnixQuerySymLink(const int xid,
 			struct cifsTconInfo *tcon,
 			const unsigned char *searchName, char **syminfo,
 			const struct nls_table *nls_codepage);
+#ifdef CONFIG_CIFS_SYMLINK_EXPERIMENTAL
 extern int CIFSSMBQueryReparseLinkInfo(const int xid,
 			struct cifsTconInfo *tcon,
 			const unsigned char *searchName,
 			char *symlinkinfo, const int buflen, __u16 fid,
 			const struct nls_table *nls_codepage);
-
+#endif /* temporarily unused until cifs_symlink fixed */
 extern int CIFSSMBOpen(const int xid, struct cifsTconInfo *tcon,
 			const char *fileName, const int disposition,
 			const int access_flags, const int omode,
@@ -348,8 +353,6 @@ extern int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon,
 			const unsigned char *searchName, __u64 *inode_number,
 			const struct nls_table *nls_codepage,
 			int remap_special_chars);
-extern int cifsConvertToUCS(__le16 *target, const char *source, int maxlen,
-			const struct nls_table *cp, int mapChars);
 
 extern int CIFSSMBLock(const int xid, struct cifsTconInfo *tcon,
 			const __u16 netfid, const __u64 len,
@@ -383,9 +386,15 @@ extern void cifs_crypto_shash_release(struct TCP_Server_Info *);
 extern int calc_seckey(struct cifsSesInfo *);
 
 #ifdef CONFIG_CIFS_WEAK_PW_HASH
-extern void calc_lanman_hash(const char *password, const char *cryptkey,
+extern int calc_lanman_hash(const char *password, const char *cryptkey,
 				bool encrypt, char *lnm_session_key);
 #endif /* CIFS_WEAK_PW_HASH */
+#ifdef CONFIG_CIFS_DNOTIFY_EXPERIMENTAL /* unused temporarily */
+extern int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon,
+			const int notify_subdirs, const __u16 netfid,
+			__u32 filter, struct file *file, int multishot,
+			const struct nls_table *nls_codepage);
+#endif /* was needed for dnotify, and will be needed for inotify when VFS fix */
 extern int CIFSSMBCopy(int xid,
 			struct cifsTconInfo *source_tcon,
 			const char *fromName,
@@ -393,10 +402,6 @@ extern int CIFSSMBCopy(int xid,
 			const char *toName, const int flags,
 			const struct nls_table *nls_codepage,
 			int remap_special_chars);
-extern int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon,
-			const int notify_subdirs, const __u16 netfid,
-			__u32 filter, struct file *file, int multishot,
-			const struct nls_table *nls_codepage);
 extern ssize_t CIFSSMBQAllEAs(const int xid, struct cifsTconInfo *tcon,
 			const unsigned char *searchName,
 			const unsigned char *ea_name, char *EAData,
@@ -427,9 +432,6 @@ extern int CIFSCheckMFSymlink(struct cifs_fattr *fattr,
 		struct cifs_sb_info *cifs_sb, int xid);
 extern int mdfour(unsigned char *, unsigned char *, int);
 extern int E_md4hash(const unsigned char *passwd, unsigned char *p16);
-extern void SMBencrypt(unsigned char *passwd, const unsigned char *c8,
-			unsigned char *p24);
-extern void E_P16(unsigned char *p14, unsigned char *p16);
-extern void E_P24(unsigned char *p21, const unsigned char *c8,
+extern int SMBencrypt(unsigned char *passwd, const unsigned char *c8,
 			unsigned char *p24);
 #endif			/* _CIFSPROTO_H */
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index df959bae672..83df937b814 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -339,12 +339,13 @@ static int validate_t2(struct smb_t2_rsp *pSMB)
 	    get_unaligned_le16(&pSMB->t2_rsp.DataOffset) > 1024)
 		goto vt2_err;
 
-	/* check that bcc is at least as big as parms + data */
-	/* check that bcc is less than negotiated smb buffer */
 	total_size = get_unaligned_le16(&pSMB->t2_rsp.ParameterCount);
 	if (total_size >= 512)
 		goto vt2_err;
 
+	/* check that bcc is at least as big as parms + data, and that it is
+	 * less than negotiated smb buffer
+	 */
 	total_size += get_unaligned_le16(&pSMB->t2_rsp.DataCount);
 	if (total_size > get_bcc(&pSMB->hdr) ||
 	    total_size >= CIFSMaxBufSize + MAX_CIFS_HDR_SIZE)
@@ -357,6 +358,13 @@ vt2_err:
 	return -EINVAL;
 }
 
+static inline void inc_rfc1001_len(void *pSMB, int count)
+{
+	struct smb_hdr *hdr = (struct smb_hdr *)pSMB;
+
+	be32_add_cpu(&hdr->smb_buf_length, count);
+}
+
 int
 CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses)
 {
@@ -409,7 +417,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses)
 		count += strlen(protocols[i].name) + 1;
 		/* null at end of source and target buffers anyway */
 	}
-	pSMB->hdr.smb_buf_length += count;
+	inc_rfc1001_len(pSMB, count);
 	pSMB->ByteCount = cpu_to_le16(count);
 
 	rc = SendReceive(xid, ses, (struct smb_hdr *) pSMB,
@@ -541,10 +549,6 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses)
 		server->secType = RawNTLMSSP;
 	else if (secFlags & CIFSSEC_MAY_LANMAN)
 		server->secType = LANMAN;
-/* #ifdef CONFIG_CIFS_EXPERIMENTAL
-	else if (secFlags & CIFSSEC_MAY_PLNTXT)
-		server->secType = ??
-#endif */
 	else {
 		rc = -EOPNOTSUPP;
 		cERROR(1, "Invalid security type");
@@ -578,7 +582,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses)
 
 	if ((pSMBr->hdr.Flags2 & SMBFLG2_EXT_SEC) &&
 		(server->capabilities & CAP_EXTENDED_SECURITY)) {
-		count = pSMBr->ByteCount;
+		count = get_bcc(&pSMBr->hdr);
 		if (count < 16) {
 			rc = -EIO;
 			goto neg_err_exit;
@@ -732,9 +736,9 @@ CIFSSMBEcho(struct TCP_Server_Info *server)
 	smb->hdr.Tid = 0xffff;
 	smb->hdr.WordCount = 1;
 	put_unaligned_le16(1, &smb->EchoCount);
-	put_bcc_le(1, &smb->hdr);
+	put_bcc(1, &smb->hdr);
 	smb->Data[0] = 'a';
-	smb->hdr.smb_buf_length += 3;
+	inc_rfc1001_len(smb, 3);
 
 	rc = cifs_call_async(server, (struct smb_hdr *)smb,
 				cifs_echo_callback, server);
@@ -852,7 +856,7 @@ PsxDelete:
 	pSMB->TotalParameterCount = pSMB->ParameterCount;
 	pSMB->InformationLevel = cpu_to_le16(SMB_POSIX_UNLINK);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -898,7 +902,7 @@ DelFileRetry:
 	pSMB->SearchAttributes =
 	    cpu_to_le16(ATTR_READONLY | ATTR_HIDDEN | ATTR_SYSTEM);
 	pSMB->BufferFormat = 0x04;
-	pSMB->hdr.smb_buf_length += name_len + 1;
+	inc_rfc1001_len(pSMB, name_len + 1);
 	pSMB->ByteCount = cpu_to_le16(name_len + 1);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -942,7 +946,7 @@ RmDirRetry:
 	}
 
 	pSMB->BufferFormat = 0x04;
-	pSMB->hdr.smb_buf_length += name_len + 1;
+	inc_rfc1001_len(pSMB, name_len + 1);
 	pSMB->ByteCount = cpu_to_le16(name_len + 1);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -985,7 +989,7 @@ MkDirRetry:
 	}
 
 	pSMB->BufferFormat = 0x04;
-	pSMB->hdr.smb_buf_length += name_len + 1;
+	inc_rfc1001_len(pSMB, name_len + 1);
 	pSMB->ByteCount = cpu_to_le16(name_len + 1);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -1063,7 +1067,7 @@ PsxCreat:
 	pSMB->TotalParameterCount = pSMB->ParameterCount;
 	pSMB->InformationLevel = cpu_to_le16(SMB_POSIX_OPEN);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -1075,7 +1079,7 @@ PsxCreat:
 	cFYI(1, "copying inode info");
 	rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-	if (rc || (pSMBr->ByteCount < sizeof(OPEN_PSX_RSP))) {
+	if (rc || get_bcc(&pSMBr->hdr) < sizeof(OPEN_PSX_RSP)) {
 		rc = -EIO;	/* bad smb */
 		goto psx_create_err;
 	}
@@ -1096,7 +1100,7 @@ PsxCreat:
 		pRetData->Type = cpu_to_le32(-1); /* unknown */
 		cFYI(DBG2, "unknown type");
 	} else {
-		if (pSMBr->ByteCount < sizeof(OPEN_PSX_RSP)
+		if (get_bcc(&pSMBr->hdr) < sizeof(OPEN_PSX_RSP)
 					+ sizeof(FILE_UNIX_BASIC_INFO)) {
 			cERROR(1, "Open response data too small");
 			pRetData->Type = cpu_to_le32(-1);
@@ -1228,7 +1232,7 @@ OldOpenRetry:
 	pSMB->Sattr = cpu_to_le16(ATTR_HIDDEN | ATTR_SYSTEM | ATTR_DIRECTORY);
 	pSMB->OpenFunction = cpu_to_le16(convert_disposition(openDisposition));
 	count += name_len;
-	pSMB->hdr.smb_buf_length += count;
+	inc_rfc1001_len(pSMB, count);
 
 	pSMB->ByteCount = cpu_to_le16(count);
 	/* long_op set to 1 to allow for oplock break timeouts */
@@ -1341,7 +1345,7 @@ openRetry:
 	    SECURITY_CONTEXT_TRACKING | SECURITY_EFFECTIVE_ONLY;
 
 	count += name_len;
-	pSMB->hdr.smb_buf_length += count;
+	inc_rfc1001_len(pSMB, count);
 
 	pSMB->ByteCount = cpu_to_le16(count);
 	/* long_op set to 1 to allow for oplock break timeouts */
@@ -1426,7 +1430,7 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, const int netfid,
 	}
 
 	iov[0].iov_base = (char *)pSMB;
-	iov[0].iov_len = pSMB->hdr.smb_buf_length + 4;
+	iov[0].iov_len = be32_to_cpu(pSMB->hdr.smb_buf_length) + 4;
 	rc = SendReceive2(xid, tcon->ses, iov, 1 /* num iovecs */,
 			 &resp_buf_type, CIFS_LOG_ERROR);
 	cifs_stats_inc(&tcon->num_reads);
@@ -1560,7 +1564,7 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon,
 
 	pSMB->DataLengthLow = cpu_to_le16(bytes_sent & 0xFFFF);
 	pSMB->DataLengthHigh = cpu_to_le16(bytes_sent >> 16);
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 
 	if (wct == 14)
 		pSMB->ByteCount = cpu_to_le16(byte_count);
@@ -1644,11 +1648,12 @@ CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon,
 
 	pSMB->DataLengthLow = cpu_to_le16(count & 0xFFFF);
 	pSMB->DataLengthHigh = cpu_to_le16(count >> 16);
-	smb_hdr_len = pSMB->hdr.smb_buf_length + 1; /* hdr + 1 byte pad */
+	/* header + 1 byte pad */
+	smb_hdr_len = be32_to_cpu(pSMB->hdr.smb_buf_length) + 1;
 	if (wct == 14)
-		pSMB->hdr.smb_buf_length += count+1;
+		inc_rfc1001_len(pSMB, count + 1);
 	else /* wct == 12 */
-		pSMB->hdr.smb_buf_length += count+5; /* smb data starts later */
+		inc_rfc1001_len(pSMB, count + 5); /* smb data starts later */
 	if (wct == 14)
 		pSMB->ByteCount = cpu_to_le16(count + 1);
 	else /* wct == 12 */ /* bigger pad, smaller smb hdr, keep offset ok */ {
@@ -1748,7 +1753,7 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon,
 		/* oplock break */
 		count = 0;
 	}
-	pSMB->hdr.smb_buf_length += count;
+	inc_rfc1001_len(pSMB, count);
 	pSMB->ByteCount = cpu_to_le16(count);
 
 	if (waitFlag) {
@@ -1839,14 +1844,14 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon,
 	pSMB->Fid = smb_file_id;
 	pSMB->InformationLevel = cpu_to_le16(SMB_SET_POSIX_LOCK);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	if (waitFlag) {
 		rc = SendReceiveBlockingLock(xid, tcon, (struct smb_hdr *) pSMB,
 			(struct smb_hdr *) pSMBr, &bytes_returned);
 	} else {
 		iov[0].iov_base = (char *)pSMB;
-		iov[0].iov_len = pSMB->hdr.smb_buf_length + 4;
+		iov[0].iov_len = be32_to_cpu(pSMB->hdr.smb_buf_length) + 4;
 		rc = SendReceive2(xid, tcon->ses, iov, 1 /* num iovecs */,
 				&resp_buf_type, timeout);
 		pSMB = NULL; /* request buf already freed by SendReceive2. Do
@@ -1862,7 +1867,7 @@ CIFSSMBPosixLock(const int xid, struct cifsTconInfo *tcon,
 		__u16 data_count;
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < sizeof(struct cifs_posix_lock))) {
+		if (rc || get_bcc(&pSMBr->hdr) < sizeof(*parm_data)) {
 			rc = -EIO;      /* bad smb */
 			goto plk_err_exit;
 		}
@@ -2012,7 +2017,7 @@ renameRetry:
 	}
 
 	count = 1 /* 1st signature byte */  + name_len + name_len2;
-	pSMB->hdr.smb_buf_length += count;
+	inc_rfc1001_len(pSMB, count);
 	pSMB->ByteCount = cpu_to_le16(count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -2092,7 +2097,7 @@ int CIFSSMBRenameOpenFile(const int xid, struct cifsTconInfo *pTcon,
 	pSMB->InformationLevel =
 		cpu_to_le16(SMB_SET_FILE_RENAME_INFORMATION);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, pTcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -2159,7 +2164,7 @@ copyRetry:
 	}
 
 	count = 1 /* 1st signature byte */  + name_len + name_len2;
-	pSMB->hdr.smb_buf_length += count;
+	inc_rfc1001_len(pSMB, count);
 	pSMB->ByteCount = cpu_to_le16(count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -2249,7 +2254,7 @@ createSymLinkRetry:
 	pSMB->DataOffset = cpu_to_le16(offset);
 	pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_UNIX_LINK);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -2335,7 +2340,7 @@ createHardLinkRetry:
 	pSMB->DataOffset = cpu_to_le16(offset);
 	pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_UNIX_HLINK);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -2406,7 +2411,7 @@ winCreateHardLinkRetry:
 	}
 
 	count = 1 /* string type byte */  + name_len + name_len2;
-	pSMB->hdr.smb_buf_length += count;
+	inc_rfc1001_len(pSMB, count);
 	pSMB->ByteCount = cpu_to_le16(count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -2477,7 +2482,7 @@ querySymLinkRetry:
 	pSMB->ParameterCount = pSMB->TotalParameterCount;
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_UNIX_LINK);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -2489,7 +2494,7 @@ querySymLinkRetry:
 
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 		/* BB also check enough total bytes returned */
-		if (rc || (pSMBr->ByteCount < 2))
+		if (rc || get_bcc(&pSMBr->hdr) < 2)
 			rc = -EIO;
 		else {
 			bool is_unicode;
@@ -2516,7 +2521,17 @@ querySymLinkRetry:
 	return rc;
 }
 
-#ifdef CONFIG_CIFS_EXPERIMENTAL
+#ifdef CONFIG_CIFS_SYMLINK_EXPERIMENTAL
+/*
+ *	Recent Windows versions now create symlinks more frequently
+ *	and they use the "reparse point" mechanism below.  We can of course
+ *	do symlinks nicely to Samba and other servers which support the
+ *	CIFS Unix Extensions and we can also do SFU symlinks and "client only"
+ *	"MF" symlinks optionally, but for recent Windows we really need to
+ *	reenable the code below and fix the cifs_symlink callers to handle this.
+ *	In the interim this code has been moved to its own config option so
+ *	it is not compiled in by default until callers fixed up and more tested.
+ */
 int
 CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon,
 			const unsigned char *searchName,
@@ -2561,14 +2576,14 @@ CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon,
 	} else {		/* decode response */
 		__u32 data_offset = le32_to_cpu(pSMBr->DataOffset);
 		__u32 data_count = le32_to_cpu(pSMBr->DataCount);
-		if ((pSMBr->ByteCount < 2) || (data_offset > 512)) {
-		/* BB also check enough total bytes returned */
+		if (get_bcc(&pSMBr->hdr) < 2 || data_offset > 512) {
+			/* BB also check enough total bytes returned */
 			rc = -EIO;	/* bad smb */
 			goto qreparse_out;
 		}
 		if (data_count && (data_count < 2048)) {
 			char *end_of_smb = 2 /* sizeof byte count */ +
-				pSMBr->ByteCount + (char *)&pSMBr->ByteCount;
+			       get_bcc(&pSMBr->hdr) + (char *)&pSMBr->ByteCount;
 
 			struct reparse_data *reparse_buf =
 						(struct reparse_data *)
@@ -2618,7 +2633,7 @@ qreparse_out:
 
 	return rc;
 }
-#endif /* CIFS_EXPERIMENTAL */
+#endif /* CIFS_SYMLINK_EXPERIMENTAL */ /* BB temporarily unused */
 
 #ifdef CONFIG_CIFS_POSIX
 
@@ -2814,7 +2829,7 @@ queryAclRetry:
 	pSMB->ParameterCount = pSMB->TotalParameterCount;
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_POSIX_ACL);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -2826,8 +2841,8 @@ queryAclRetry:
 		/* decode response */
 
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
-		if (rc || (pSMBr->ByteCount < 2))
 		/* BB also check enough total bytes returned */
+		if (rc || get_bcc(&pSMBr->hdr) < 2)
 			rc = -EIO;      /* bad smb */
 		else {
 			__u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset);
@@ -2908,7 +2923,7 @@ setAclRetry:
 	pSMB->ParameterCount = cpu_to_le16(params);
 	pSMB->TotalParameterCount = pSMB->ParameterCount;
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -2966,7 +2981,7 @@ GetExtAttrRetry:
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_ATTR_FLAGS);
 	pSMB->Pad = 0;
 	pSMB->Fid = netfid;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->t2.ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -2976,8 +2991,8 @@ GetExtAttrRetry:
 	} else {
 		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
-		if (rc || (pSMBr->ByteCount < 2))
 		/* BB also check enough total bytes returned */
+		if (rc || get_bcc(&pSMBr->hdr) < 2)
 			/* If rc should we check for EOPNOSUPP and
 			   disable the srvino flag? or in caller? */
 			rc = -EIO;      /* bad smb */
@@ -3052,6 +3067,7 @@ validate_ntransact(char *buf, char **ppparm, char **ppdata,
 	char *end_of_smb;
 	__u32 data_count, data_offset, parm_count, parm_offset;
 	struct smb_com_ntransact_rsp *pSMBr;
+	u16 bcc;
 
 	*pdatalen = 0;
 	*pparmlen = 0;
@@ -3061,8 +3077,8 @@ validate_ntransact(char *buf, char **ppparm, char **ppdata,
 
 	pSMBr = (struct smb_com_ntransact_rsp *)buf;
 
-	/* ByteCount was converted from little endian in SendReceive */
-	end_of_smb = 2 /* sizeof byte count */ + pSMBr->ByteCount +
+	bcc = get_bcc(&pSMBr->hdr);
+	end_of_smb = 2 /* sizeof byte count */ + bcc +
 			(char *)&pSMBr->ByteCount;
 
 	data_offset = le32_to_cpu(pSMBr->DataOffset);
@@ -3088,7 +3104,7 @@ validate_ntransact(char *buf, char **ppparm, char **ppdata,
 			*ppdata, data_count, (data_count + *ppdata),
 			end_of_smb, pSMBr);
 		return -EINVAL;
-	} else if (parm_count + data_count > pSMBr->ByteCount) {
+	} else if (parm_count + data_count > bcc) {
 		cFYI(1, "parm count and data count larger than SMB");
 		return -EINVAL;
 	}
@@ -3124,9 +3140,9 @@ CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid,
 	pSMB->AclFlags = cpu_to_le32(CIFS_ACL_OWNER | CIFS_ACL_GROUP |
 				     CIFS_ACL_DACL);
 	pSMB->ByteCount = cpu_to_le16(11); /* 3 bytes pad + 8 bytes parm */
-	pSMB->hdr.smb_buf_length += 11;
+	inc_rfc1001_len(pSMB, 11);
 	iov[0].iov_base = (char *)pSMB;
-	iov[0].iov_len = pSMB->hdr.smb_buf_length + 4;
+	iov[0].iov_len = be32_to_cpu(pSMB->hdr.smb_buf_length) + 4;
 
 	rc = SendReceive2(xid, tcon->ses, iov, 1 /* num iovec */, &buf_type,
 			 0);
@@ -3235,10 +3251,9 @@ setCifsAclRetry:
 		memcpy((char *) &pSMBr->hdr.Protocol + data_offset,
 			(char *) pntsd,
 			acllen);
-		pSMB->hdr.smb_buf_length += (byte_count + data_count);
-
+		inc_rfc1001_len(pSMB, byte_count + data_count);
 	} else
-		pSMB->hdr.smb_buf_length += byte_count;
+		inc_rfc1001_len(pSMB, byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 		(struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -3289,7 +3304,7 @@ QInfRetry:
 	}
 	pSMB->BufferFormat = 0x04;
 	name_len++; /* account for buffer type byte */
-	pSMB->hdr.smb_buf_length += (__u16) name_len;
+	inc_rfc1001_len(pSMB, (__u16)name_len);
 	pSMB->ByteCount = cpu_to_le16(name_len);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -3364,7 +3379,7 @@ QFileInfoRetry:
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_ALL_INFO);
 	pSMB->Pad = 0;
 	pSMB->Fid = netfid;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -3375,7 +3390,7 @@ QFileInfoRetry:
 
 		if (rc) /* BB add auto retry on EOPNOTSUPP? */
 			rc = -EIO;
-		else if (pSMBr->ByteCount < 40)
+		else if (get_bcc(&pSMBr->hdr) < 40)
 			rc = -EIO;	/* bad smb */
 		else if (pFindData) {
 			__u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset);
@@ -3451,7 +3466,7 @@ QPathInfoRetry:
 	else
 		pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_ALL_INFO);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -3463,9 +3478,9 @@ QPathInfoRetry:
 
 		if (rc) /* BB add auto retry on EOPNOTSUPP? */
 			rc = -EIO;
-		else if (!legacy && (pSMBr->ByteCount < 40))
+		else if (!legacy && get_bcc(&pSMBr->hdr) < 40)
 			rc = -EIO;	/* bad smb */
-		else if (legacy && (pSMBr->ByteCount < 24))
+		else if (legacy && get_bcc(&pSMBr->hdr) < 24)
 			rc = -EIO;  /* 24 or 26 expected but we do not read
 					last field */
 		else if (pFindData) {
@@ -3532,7 +3547,7 @@ UnixQFileInfoRetry:
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_UNIX_BASIC);
 	pSMB->Pad = 0;
 	pSMB->Fid = netfid;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -3541,7 +3556,7 @@ UnixQFileInfoRetry:
 	} else {		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < sizeof(FILE_UNIX_BASIC_INFO))) {
+		if (rc || get_bcc(&pSMBr->hdr) < sizeof(FILE_UNIX_BASIC_INFO)) {
 			cERROR(1, "Malformed FILE_UNIX_BASIC_INFO response.\n"
 				   "Unix Extensions can be disabled on mount "
 				   "by specifying the nosfu mount option.");
@@ -3617,7 +3632,7 @@ UnixQPathInfoRetry:
 	pSMB->ParameterCount = pSMB->TotalParameterCount;
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_UNIX_BASIC);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -3627,7 +3642,7 @@ UnixQPathInfoRetry:
 	} else {		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < sizeof(FILE_UNIX_BASIC_INFO))) {
+		if (rc || get_bcc(&pSMBr->hdr) < sizeof(FILE_UNIX_BASIC_INFO)) {
 			cERROR(1, "Malformed FILE_UNIX_BASIC_INFO response.\n"
 				   "Unix Extensions can be disabled on mount "
 				   "by specifying the nosfu mount option.");
@@ -3731,7 +3746,7 @@ findFirstRetry:
 
 	/* BB what should we set StorageType to? Does it matter? BB */
 	pSMB->SearchStorageType = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -3860,7 +3875,7 @@ int CIFSFindNext(const int xid, struct cifsTconInfo *tcon,
 	byte_count = params + 1 /* pad */ ;
 	pSMB->TotalParameterCount = cpu_to_le16(params);
 	pSMB->ParameterCount = pSMB->TotalParameterCount;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4022,7 +4037,7 @@ GetInodeNumberRetry:
 	pSMB->ParameterCount = pSMB->TotalParameterCount;
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_INTERNAL_INFO);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4032,8 +4047,8 @@ GetInodeNumberRetry:
 	} else {
 		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
-		if (rc || (pSMBr->ByteCount < 2))
 		/* BB also check enough total bytes returned */
+		if (rc || get_bcc(&pSMBr->hdr) < 2)
 			/* If rc should we check for EOPNOSUPP and
 			disable the srvino flag? or in caller? */
 			rc = -EIO;      /* bad smb */
@@ -4246,7 +4261,7 @@ getDFSRetry:
 	pSMB->ParameterCount = cpu_to_le16(params);
 	pSMB->TotalParameterCount = pSMB->ParameterCount;
 	pSMB->MaxReferralLevel = cpu_to_le16(3);
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, ses, (struct smb_hdr *) pSMB,
@@ -4258,13 +4273,13 @@ getDFSRetry:
 	rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
 	/* BB Also check if enough total bytes returned? */
-	if (rc || (pSMBr->ByteCount < 17)) {
+	if (rc || get_bcc(&pSMBr->hdr) < 17) {
 		rc = -EIO;      /* bad smb */
 		goto GetDFSRefExit;
 	}
 
 	cFYI(1, "Decoding GetDFSRefer response BCC: %d  Offset %d",
-				pSMBr->ByteCount,
+				get_bcc(&pSMBr->hdr),
 				le16_to_cpu(pSMBr->t2.DataOffset));
 
 	/* parse returned result into more usable form */
@@ -4320,7 +4335,7 @@ oldQFSInfoRetry:
 	pSMB->Reserved3 = 0;
 	pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION);
 	pSMB->InformationLevel = cpu_to_le16(SMB_INFO_ALLOCATION);
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4330,12 +4345,12 @@ oldQFSInfoRetry:
 	} else {                /* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < 18))
+		if (rc || get_bcc(&pSMBr->hdr) < 18)
 			rc = -EIO;      /* bad smb */
 		else {
 			__u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset);
 			cFYI(1, "qfsinf resp BCC: %d  Offset %d",
-				 pSMBr->ByteCount, data_offset);
+				 get_bcc(&pSMBr->hdr), data_offset);
 
 			response_data = (FILE_SYSTEM_ALLOC_INFO *)
 				(((char *) &pSMBr->hdr.Protocol) + data_offset);
@@ -4399,7 +4414,7 @@ QFSInfoRetry:
 	pSMB->Reserved3 = 0;
 	pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION);
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FS_SIZE_INFO);
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4409,7 +4424,7 @@ QFSInfoRetry:
 	} else {		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < 24))
+		if (rc || get_bcc(&pSMBr->hdr) < 24)
 			rc = -EIO;	/* bad smb */
 		else {
 			__u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset);
@@ -4479,7 +4494,7 @@ QFSAttributeRetry:
 	pSMB->Reserved3 = 0;
 	pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION);
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FS_ATTRIBUTE_INFO);
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4489,7 +4504,7 @@ QFSAttributeRetry:
 	} else {		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < 13)) {
+		if (rc || get_bcc(&pSMBr->hdr) < 13) {
 			/* BB also check if enough bytes returned */
 			rc = -EIO;	/* bad smb */
 		} else {
@@ -4550,7 +4565,7 @@ QFSDeviceRetry:
 	pSMB->Reserved3 = 0;
 	pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION);
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FS_DEVICE_INFO);
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4560,7 +4575,8 @@ QFSDeviceRetry:
 	} else {		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < sizeof(FILE_SYSTEM_DEVICE_INFO)))
+		if (rc || get_bcc(&pSMBr->hdr) <
+			  sizeof(FILE_SYSTEM_DEVICE_INFO))
 			rc = -EIO;	/* bad smb */
 		else {
 			__u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset);
@@ -4619,7 +4635,7 @@ QFSUnixRetry:
 	pSMB->Reserved3 = 0;
 	pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION);
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_CIFS_UNIX_INFO);
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4629,7 +4645,7 @@ QFSUnixRetry:
 	} else {		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < 13)) {
+		if (rc || get_bcc(&pSMBr->hdr) < 13) {
 			rc = -EIO;	/* bad smb */
 		} else {
 			__u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset);
@@ -4702,7 +4718,7 @@ SETFSUnixRetry:
 	pSMB->ClientUnixMinor = cpu_to_le16(CIFS_UNIX_MINOR_VERSION);
 	pSMB->ClientUnixCap = cpu_to_le64(cap);
 
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4764,7 +4780,7 @@ QFSPosixRetry:
 	pSMB->Reserved3 = 0;
 	pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION);
 	pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_POSIX_FS_INFO);
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4774,7 +4790,7 @@ QFSPosixRetry:
 	} else {		/* decode response */
 		rc = validate_t2((struct smb_t2_rsp *)pSMBr);
 
-		if (rc || (pSMBr->ByteCount < 13)) {
+		if (rc || get_bcc(&pSMBr->hdr) < 13) {
 			rc = -EIO;	/* bad smb */
 		} else {
 			__u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset);
@@ -4890,7 +4906,7 @@ SetEOFRetry:
 	pSMB->ParameterCount = cpu_to_le16(params);
 	pSMB->TotalParameterCount = pSMB->ParameterCount;
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	parm_data->FileSize = cpu_to_le64(size);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -4969,7 +4985,7 @@ CIFSSMBSetFileSize(const int xid, struct cifsTconInfo *tcon, __u64 size,
 				cpu_to_le16(SMB_SET_FILE_END_OF_FILE_INFO);
 	}
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0);
 	if (rc) {
@@ -5037,7 +5053,7 @@ CIFSSMBSetFileInfo(const int xid, struct cifsTconInfo *tcon,
 	else
 		pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_BASIC_INFO);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	memcpy(data_offset, data, sizeof(FILE_BASIC_INFO));
 	rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0);
@@ -5096,7 +5112,7 @@ CIFSSMBSetFileDisposition(const int xid, struct cifsTconInfo *tcon,
 	pSMB->Fid = fid;
 	pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_DISPOSITION_INFO);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	*data_offset = delete_file ? 1 : 0;
 	rc = SendReceiveNoRsp(xid, tcon->ses, (struct smb_hdr *) pSMB, 0);
@@ -5169,7 +5185,7 @@ SetTimesRetry:
 	else
 		pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_BASIC_INFO);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	memcpy(data_offset, data, sizeof(FILE_BASIC_INFO));
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -5221,7 +5237,7 @@ SetAttrLgcyRetry:
 	}
 	pSMB->attr = cpu_to_le16(dos_attrs);
 	pSMB->BufferFormat = 0x04;
-	pSMB->hdr.smb_buf_length += name_len + 1;
+	inc_rfc1001_len(pSMB, name_len + 1);
 	pSMB->ByteCount = cpu_to_le16(name_len + 1);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -5326,7 +5342,7 @@ CIFSSMBUnixSetFileInfo(const int xid, struct cifsTconInfo *tcon,
 	pSMB->Fid = fid;
 	pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_UNIX_BASIC);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	cifs_fill_unix_set_info(data_offset, args);
@@ -5402,7 +5418,7 @@ setPermsRetry:
 	pSMB->TotalDataCount = pSMB->DataCount;
 	pSMB->InformationLevel = cpu_to_le16(SMB_SET_FILE_UNIX_BASIC);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 
 	cifs_fill_unix_set_info(data_offset, args);
 
@@ -5418,79 +5434,6 @@ setPermsRetry:
 	return rc;
 }
 
-int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon,
-		  const int notify_subdirs, const __u16 netfid,
-		  __u32 filter, struct file *pfile, int multishot,
-		  const struct nls_table *nls_codepage)
-{
-	int rc = 0;
-	struct smb_com_transaction_change_notify_req *pSMB = NULL;
-	struct smb_com_ntransaction_change_notify_rsp *pSMBr = NULL;
-	struct dir_notify_req *dnotify_req;
-	int bytes_returned;
-
-	cFYI(1, "In CIFSSMBNotify for file handle %d", (int)netfid);
-	rc = smb_init(SMB_COM_NT_TRANSACT, 23, tcon, (void **) &pSMB,
-		      (void **) &pSMBr);
-	if (rc)
-		return rc;
-
-	pSMB->TotalParameterCount = 0 ;
-	pSMB->TotalDataCount = 0;
-	pSMB->MaxParameterCount = cpu_to_le32(2);
-	/* BB find exact data count max from sess structure BB */
-	pSMB->MaxDataCount = 0; /* same in little endian or be */
-/* BB VERIFY verify which is correct for above BB */
-	pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf -
-					     MAX_CIFS_HDR_SIZE) & 0xFFFFFF00);
-
-	pSMB->MaxSetupCount = 4;
-	pSMB->Reserved = 0;
-	pSMB->ParameterOffset = 0;
-	pSMB->DataCount = 0;
-	pSMB->DataOffset = 0;
-	pSMB->SetupCount = 4; /* single byte does not need le conversion */
-	pSMB->SubCommand = cpu_to_le16(NT_TRANSACT_NOTIFY_CHANGE);
-	pSMB->ParameterCount = pSMB->TotalParameterCount;
-	if (notify_subdirs)
-		pSMB->WatchTree = 1; /* one byte - no le conversion needed */
-	pSMB->Reserved2 = 0;
-	pSMB->CompletionFilter = cpu_to_le32(filter);
-	pSMB->Fid = netfid; /* file handle always le */
-	pSMB->ByteCount = 0;
-
-	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
-			 (struct smb_hdr *)pSMBr, &bytes_returned,
-			 CIFS_ASYNC_OP);
-	if (rc) {
-		cFYI(1, "Error in Notify = %d", rc);
-	} else {
-		/* Add file to outstanding requests */
-		/* BB change to kmem cache alloc */
-		dnotify_req = kmalloc(
-						sizeof(struct dir_notify_req),
-						 GFP_KERNEL);
-		if (dnotify_req) {
-			dnotify_req->Pid = pSMB->hdr.Pid;
-			dnotify_req->PidHigh = pSMB->hdr.PidHigh;
-			dnotify_req->Mid = pSMB->hdr.Mid;
-			dnotify_req->Tid = pSMB->hdr.Tid;
-			dnotify_req->Uid = pSMB->hdr.Uid;
-			dnotify_req->netfid = netfid;
-			dnotify_req->pfile = pfile;
-			dnotify_req->filter = filter;
-			dnotify_req->multishot = multishot;
-			spin_lock(&GlobalMid_Lock);
-			list_add_tail(&dnotify_req->lhead,
-					&GlobalDnotifyReqList);
-			spin_unlock(&GlobalMid_Lock);
-		} else
-			rc = -ENOMEM;
-	}
-	cifs_buf_release(pSMB);
-	return rc;
-}
-
 #ifdef CONFIG_CIFS_XATTR
 /*
  * Do a path-based QUERY_ALL_EAS call and parse the result. This is a common
@@ -5560,7 +5503,7 @@ QAllEAsRetry:
 	pSMB->ParameterCount = pSMB->TotalParameterCount;
 	pSMB->InformationLevel = cpu_to_le16(SMB_INFO_QUERY_ALL_EAS);
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
@@ -5576,7 +5519,7 @@ QAllEAsRetry:
 	of these trans2 responses */
 
 	rc = validate_t2((struct smb_t2_rsp *)pSMBr);
-	if (rc || (pSMBr->ByteCount < 4)) {
+	if (rc || get_bcc(&pSMBr->hdr) < 4) {
 		rc = -EIO;	/* bad smb */
 		goto QAllEAsOut;
 	}
@@ -5773,7 +5716,7 @@ SetEARetry:
 	pSMB->ParameterCount = cpu_to_le16(params);
 	pSMB->TotalParameterCount = pSMB->ParameterCount;
 	pSMB->Reserved4 = 0;
-	pSMB->hdr.smb_buf_length += byte_count;
+	inc_rfc1001_len(pSMB, byte_count);
 	pSMB->ByteCount = cpu_to_le16(byte_count);
 	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
 			 (struct smb_hdr *) pSMBr, &bytes_returned, 0);
@@ -5787,5 +5730,99 @@ SetEARetry:
 
 	return rc;
 }
-
 #endif
+
+#ifdef CONFIG_CIFS_DNOTIFY_EXPERIMENTAL /* BB unused temporarily */
+/*
+ *	Years ago the kernel added a "dnotify" function for Samba server,
+ *	to allow network clients (such as Windows) to display updated
+ *	lists of files in directory listings automatically when
+ *	files are added by one user when another user has the
+ *	same directory open on their desktop.  The Linux cifs kernel
+ *	client hooked into the kernel side of this interface for
+ *	the same reason, but ironically when the VFS moved from
+ *	"dnotify" to "inotify" it became harder to plug in Linux
+ *	network file system clients (the most obvious use case
+ *	for notify interfaces is when multiple users can update
+ *	the contents of the same directory - exactly what network
+ *	file systems can do) although the server (Samba) could
+ *	still use it.  For the short term we leave the worker
+ *	function ifdeffed out (below) until inotify is fixed
+ *	in the VFS to make it easier to plug in network file
+ *	system clients.  If inotify turns out to be permanently
+ *	incompatible for network fs clients, we could instead simply
+ *	expose this config flag by adding a future cifs (and smb2) notify ioctl.
+ */
+int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon,
+		  const int notify_subdirs, const __u16 netfid,
+		  __u32 filter, struct file *pfile, int multishot,
+		  const struct nls_table *nls_codepage)
+{
+	int rc = 0;
+	struct smb_com_transaction_change_notify_req *pSMB = NULL;
+	struct smb_com_ntransaction_change_notify_rsp *pSMBr = NULL;
+	struct dir_notify_req *dnotify_req;
+	int bytes_returned;
+
+	cFYI(1, "In CIFSSMBNotify for file handle %d", (int)netfid);
+	rc = smb_init(SMB_COM_NT_TRANSACT, 23, tcon, (void **) &pSMB,
+		      (void **) &pSMBr);
+	if (rc)
+		return rc;
+
+	pSMB->TotalParameterCount = 0 ;
+	pSMB->TotalDataCount = 0;
+	pSMB->MaxParameterCount = cpu_to_le32(2);
+	/* BB find exact data count max from sess structure BB */
+	pSMB->MaxDataCount = 0; /* same in little endian or be */
+/* BB VERIFY verify which is correct for above BB */
+	pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf -
+					     MAX_CIFS_HDR_SIZE) & 0xFFFFFF00);
+
+	pSMB->MaxSetupCount = 4;
+	pSMB->Reserved = 0;
+	pSMB->ParameterOffset = 0;
+	pSMB->DataCount = 0;
+	pSMB->DataOffset = 0;
+	pSMB->SetupCount = 4; /* single byte does not need le conversion */
+	pSMB->SubCommand = cpu_to_le16(NT_TRANSACT_NOTIFY_CHANGE);
+	pSMB->ParameterCount = pSMB->TotalParameterCount;
+	if (notify_subdirs)
+		pSMB->WatchTree = 1; /* one byte - no le conversion needed */
+	pSMB->Reserved2 = 0;
+	pSMB->CompletionFilter = cpu_to_le32(filter);
+	pSMB->Fid = netfid; /* file handle always le */
+	pSMB->ByteCount = 0;
+
+	rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB,
+			 (struct smb_hdr *)pSMBr, &bytes_returned,
+			 CIFS_ASYNC_OP);
+	if (rc) {
+		cFYI(1, "Error in Notify = %d", rc);
+	} else {
+		/* Add file to outstanding requests */
+		/* BB change to kmem cache alloc */
+		dnotify_req = kmalloc(
+						sizeof(struct dir_notify_req),
+						 GFP_KERNEL);
+		if (dnotify_req) {
+			dnotify_req->Pid = pSMB->hdr.Pid;
+			dnotify_req->PidHigh = pSMB->hdr.PidHigh;
+			dnotify_req->Mid = pSMB->hdr.Mid;
+			dnotify_req->Tid = pSMB->hdr.Tid;
+			dnotify_req->Uid = pSMB->hdr.Uid;
+			dnotify_req->netfid = netfid;
+			dnotify_req->pfile = pfile;
+			dnotify_req->filter = filter;
+			dnotify_req->multishot = multishot;
+			spin_lock(&GlobalMid_Lock);
+			list_add_tail(&dnotify_req->lhead,
+					&GlobalDnotifyReqList);
+			spin_unlock(&GlobalMid_Lock);
+		} else
+			rc = -ENOMEM;
+	}
+	cifs_buf_release(pSMB);
+	return rc;
+}
+#endif /* was needed for dnotify, and will be needed for inotify when VFS fix */
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index db9d55b507d..da284e3cb65 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -102,6 +102,7 @@ struct smb_vol {
 	bool fsc:1;	/* enable fscache */
 	bool mfsymlinks:1; /* use Minshall+French Symlinks */
 	bool multiuser:1;
+	bool use_smb2:1; /* force smb2 use on mount instead of cifs */
 	unsigned int rsize;
 	unsigned int wsize;
 	bool sockopt_tcp_nodelay:1;
@@ -274,7 +275,8 @@ static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB)
 	char *data_area_of_target;
 	char *data_area_of_buf2;
 	int remaining;
-	__u16 byte_count, total_data_size, total_in_buf, total_in_buf2;
+	unsigned int byte_count, total_in_buf;
+	__u16 total_data_size, total_in_buf2;
 
 	total_data_size = get_unaligned_le16(&pSMBt->t2_rsp.TotalDataCount);
 
@@ -287,7 +289,7 @@ static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB)
 	remaining = total_data_size - total_in_buf;
 
 	if (remaining < 0)
-		return -EINVAL;
+		return -EPROTO;
 
 	if (remaining == 0) /* nothing to do, ignore */
 		return 0;
@@ -308,19 +310,28 @@ static int coalesce_t2(struct smb_hdr *psecond, struct smb_hdr *pTargetSMB)
 	data_area_of_target += total_in_buf;
 
 	/* copy second buffer into end of first buffer */
-	memcpy(data_area_of_target, data_area_of_buf2, total_in_buf2);
 	total_in_buf += total_in_buf2;
+	/* is the result too big for the field? */
+	if (total_in_buf > USHRT_MAX)
+		return -EPROTO;
 	put_unaligned_le16(total_in_buf, &pSMBt->t2_rsp.DataCount);
-	byte_count = get_bcc_le(pTargetSMB);
-	byte_count += total_in_buf2;
-	put_bcc_le(byte_count, pTargetSMB);
 
-	byte_count = pTargetSMB->smb_buf_length;
+	/* fix up the BCC */
+	byte_count = get_bcc(pTargetSMB);
 	byte_count += total_in_buf2;
+	/* is the result too big for the field? */
+	if (byte_count > USHRT_MAX)
+		return -EPROTO;
+	put_bcc(byte_count, pTargetSMB);
 
-	/* BB also add check that we are not beyond maximum buffer size */
+	byte_count = be32_to_cpu(pTargetSMB->smb_buf_length);
+	byte_count += total_in_buf2;
+	/* don't allow buffer to overflow */
+	if (byte_count > CIFSMaxBufSize)
+		return -ENOBUFS;
+	pTargetSMB->smb_buf_length = cpu_to_be32(byte_count);
 
-	pTargetSMB->smb_buf_length = byte_count;
+	memcpy(data_area_of_target, data_area_of_buf2, total_in_buf2);
 
 	if (remaining == total_in_buf2) {
 		cFYI(1, "found the last secondary response");
@@ -485,8 +496,7 @@ incomplete_rcv:
 		/* Note that FC 1001 length is big endian on the wire,
 		but we convert it here so it is always manipulated
 		as host byte order */
-		pdu_length = be32_to_cpu((__force __be32)smb_buffer->smb_buf_length);
-		smb_buffer->smb_buf_length = pdu_length;
+		pdu_length = be32_to_cpu(smb_buffer->smb_buf_length);
 
 		cFYI(1, "rfc1002 length 0x%x", pdu_length+4);
 
@@ -607,59 +617,63 @@ incomplete_rcv:
 		list_for_each_safe(tmp, tmp2, &server->pending_mid_q) {
 			mid_entry = list_entry(tmp, struct mid_q_entry, qhead);
 
-			if ((mid_entry->mid == smb_buffer->Mid) &&
-			    (mid_entry->midState == MID_REQUEST_SUBMITTED) &&
-			    (mid_entry->command == smb_buffer->Command)) {
-				if (length == 0 &&
-				   check2ndT2(smb_buffer, server->maxBuf) > 0) {
-					/* We have a multipart transact2 resp */
-					isMultiRsp = true;
-					if (mid_entry->resp_buf) {
-						/* merge response - fix up 1st*/
-						if (coalesce_t2(smb_buffer,
-							mid_entry->resp_buf)) {
-							mid_entry->multiRsp =
-								 true;
-							break;
-						} else {
-							/* all parts received */
-							mid_entry->multiEnd =
-								 true;
-							goto multi_t2_fnd;
-						}
+			if (mid_entry->mid != smb_buffer->Mid ||
+			    mid_entry->midState != MID_REQUEST_SUBMITTED ||
+			    mid_entry->command != smb_buffer->Command) {
+				mid_entry = NULL;
+				continue;
+			}
+
+			if (length == 0 &&
+			    check2ndT2(smb_buffer, server->maxBuf) > 0) {
+				/* We have a multipart transact2 resp */
+				isMultiRsp = true;
+				if (mid_entry->resp_buf) {
+					/* merge response - fix up 1st*/
+					length = coalesce_t2(smb_buffer,
+							mid_entry->resp_buf);
+					if (length > 0) {
+						length = 0;
+						mid_entry->multiRsp = true;
+						break;
 					} else {
-						if (!isLargeBuf) {
-							cERROR(1, "1st trans2 resp needs bigbuf");
-					/* BB maybe we can fix this up,  switch
-					   to already allocated large buffer? */
-						} else {
-							/* Have first buffer */
-							mid_entry->resp_buf =
-								 smb_buffer;
-							mid_entry->largeBuf =
-								 true;
-							bigbuf = NULL;
-						}
+						/* all parts received or
+						 * packet is malformed
+						 */
+						mid_entry->multiEnd = true;
+						goto multi_t2_fnd;
+					}
+				} else {
+					if (!isLargeBuf) {
+						/*
+						 * FIXME: switch to already
+						 *        allocated largebuf?
+						 */
+						cERROR(1, "1st trans2 resp "
+							  "needs bigbuf");
+					} else {
+						/* Have first buffer */
+						mid_entry->resp_buf =
+							 smb_buffer;
+						mid_entry->largeBuf = true;
+						bigbuf = NULL;
 					}
-					break;
 				}
-				mid_entry->resp_buf = smb_buffer;
-				mid_entry->largeBuf = isLargeBuf;
+				break;
+			}
+			mid_entry->resp_buf = smb_buffer;
+			mid_entry->largeBuf = isLargeBuf;
 multi_t2_fnd:
-				if (length == 0)
-					mid_entry->midState =
-							MID_RESPONSE_RECEIVED;
-				else
-					mid_entry->midState =
-							MID_RESPONSE_MALFORMED;
+			if (length == 0)
+				mid_entry->midState = MID_RESPONSE_RECEIVED;
+			else
+				mid_entry->midState = MID_RESPONSE_MALFORMED;
 #ifdef CONFIG_CIFS_STATS2
-				mid_entry->when_received = jiffies;
+			mid_entry->when_received = jiffies;
 #endif
-				list_del_init(&mid_entry->qhead);
-				mid_entry->callback(mid_entry);
-				break;
-			}
-			mid_entry = NULL;
+			list_del_init(&mid_entry->qhead);
+			mid_entry->callback(mid_entry);
+			break;
 		}
 		spin_unlock(&GlobalMid_Lock);
 
@@ -721,7 +735,7 @@ multi_t2_fnd:
 		sock_release(csocket);
 		server->ssocket = NULL;
 	}
-	/* buffer usuallly freed in free_mid - need to free it here on exit */
+	/* buffer usually freed in free_mid - need to free it here on exit */
 	cifs_buf_release(bigbuf);
 	if (smallbuf) /* no sense logging a debug message if NULL */
 		cifs_small_buf_release(smallbuf);
@@ -804,11 +818,11 @@ extract_hostname(const char *unc)
 }
 
 static int
-cifs_parse_mount_options(char *options, const char *devname,
+cifs_parse_mount_options(const char *mountdata, const char *devname,
 			 struct smb_vol *vol)
 {
-	char *value;
-	char *data;
+	char *value, *data, *end;
+	char *mountdata_copy, *options;
 	unsigned int  temp_len, i, j;
 	char separator[2];
 	short int override_uid = -1;
@@ -848,9 +862,15 @@ cifs_parse_mount_options(char *options, const char *devname,
 
 	vol->actimeo = CIFS_DEF_ACTIMEO;
 
-	if (!options)
-		return 1;
+	if (!mountdata)
+		goto cifs_parse_mount_err;
 
+	mountdata_copy = kstrndup(mountdata, PAGE_SIZE, GFP_KERNEL);
+	if (!mountdata_copy)
+		goto cifs_parse_mount_err;
+
+	options = mountdata_copy;
+	end = options + strlen(options);
 	if (strncmp(options, "sep=", 4) == 0) {
 		if (options[4] != 0) {
 			separator[0] = options[4];
@@ -875,17 +895,22 @@ cifs_parse_mount_options(char *options, const char *devname,
 			if (!value) {
 				printk(KERN_WARNING
 				       "CIFS: invalid or missing username\n");
-				return 1;	/* needs_arg; */
+				goto cifs_parse_mount_err;
 			} else if (!*value) {
 				/* null user, ie anonymous, authentication */
 				vol->nullauth = 1;
 			}
 			if (strnlen(value, MAX_USERNAME_SIZE) <
 						MAX_USERNAME_SIZE) {
-				vol->username = value;
+				vol->username = kstrdup(value, GFP_KERNEL);
+				if (!vol->username) {
+					printk(KERN_WARNING "CIFS: no memory "
+							    "for username\n");
+					goto cifs_parse_mount_err;
+				}
 			} else {
 				printk(KERN_WARNING "CIFS: username too long\n");
-				return 1;
+				goto cifs_parse_mount_err;
 			}
 		} else if (strnicmp(data, "pass", 4) == 0) {
 			if (!value) {
@@ -916,6 +941,7 @@ cifs_parse_mount_options(char *options, const char *devname,
 			the only illegal character in a password is null */
 
 			if ((value[temp_len] == 0) &&
+			    (value + temp_len < end) &&
 			    (value[temp_len+1] == separator[0])) {
 				/* reinsert comma */
 				value[temp_len] = separator[0];
@@ -948,7 +974,7 @@ cifs_parse_mount_options(char *options, const char *devname,
 				if (vol->password == NULL) {
 					printk(KERN_WARNING "CIFS: no memory "
 							    "for password\n");
-					return 1;
+					goto cifs_parse_mount_err;
 				}
 				for (i = 0, j = 0; i < temp_len; i++, j++) {
 					vol->password[j] = value[i];
@@ -964,7 +990,7 @@ cifs_parse_mount_options(char *options, const char *devname,
 				if (vol->password == NULL) {
 					printk(KERN_WARNING "CIFS: no memory "
 							    "for password\n");
-					return 1;
+					goto cifs_parse_mount_err;
 				}
 				strcpy(vol->password, value);
 			}
@@ -974,11 +1000,16 @@ cifs_parse_mount_options(char *options, const char *devname,
 				vol->UNCip = NULL;
 			} else if (strnlen(value, INET6_ADDRSTRLEN) <
 							INET6_ADDRSTRLEN) {
-				vol->UNCip = value;
+				vol->UNCip = kstrdup(value, GFP_KERNEL);
+				if (!vol->UNCip) {
+					printk(KERN_WARNING "CIFS: no memory "
+							    "for UNC IP\n");
+					goto cifs_parse_mount_err;
+				}
 			} else {
 				printk(KERN_WARNING "CIFS: ip address "
 						    "too long\n");
-				return 1;
+				goto cifs_parse_mount_err;
 			}
 		} else if (strnicmp(data, "sec", 3) == 0) {
 			if (!value || !*value) {
@@ -991,7 +1022,7 @@ cifs_parse_mount_options(char *options, const char *devname,
 				/* vol->secFlg |= CIFSSEC_MUST_SEAL |
 					CIFSSEC_MAY_KRB5; */
 				cERROR(1, "Krb5 cifs privacy not supported");
-				return 1;
+				goto cifs_parse_mount_err;
 			} else if (strnicmp(value, "krb5", 4) == 0) {
 				vol->secFlg |= CIFSSEC_MAY_KRB5;
 			} else if (strnicmp(value, "ntlmsspi", 8) == 0) {
@@ -1021,7 +1052,23 @@ cifs_parse_mount_options(char *options, const char *devname,
 				vol->nullauth = 1;
 			} else {
 				cERROR(1, "bad security option: %s", value);
-				return 1;
+				goto cifs_parse_mount_err;
+			}
+		} else if (strnicmp(data, "vers", 3) == 0) {
+			if (!value || !*value) {
+				cERROR(1, "no protocol version specified"
+					  " after vers= mount option");
+			} else if ((strnicmp(value, "cifs", 4) == 0) ||
+				   (strnicmp(value, "1", 1) == 0)) {
+				/* this is the default */
+				continue;
+			} else if ((strnicmp(value, "smb2", 4) == 0) ||
+				   (strnicmp(value, "2", 1) == 0)) {
+#ifdef CONFIG_CIFS_SMB2
+				vol->use_smb2 = true;
+#else
+				cERROR(1, "smb2 support not enabled");
+#endif /* CONFIG_CIFS_SMB2 */
 			}
 		} else if ((strnicmp(data, "unc", 3) == 0)
 			   || (strnicmp(data, "target", 6) == 0)
@@ -1029,12 +1076,12 @@ cifs_parse_mount_options(char *options, const char *devname,
 			if (!value || !*value) {
 				printk(KERN_WARNING "CIFS: invalid path to "
 						    "network resource\n");
-				return 1;	/* needs_arg; */
+				goto cifs_parse_mount_err;
 			}
 			if ((temp_len = strnlen(value, 300)) < 300) {
 				vol->UNC = kmalloc(temp_len+1, GFP_KERNEL);
 				if (vol->UNC == NULL)
-					return 1;
+					goto cifs_parse_mount_err;
 				strcpy(vol->UNC, value);
 				if (strncmp(vol->UNC, "//", 2) == 0) {
 					vol->UNC[0] = '\\';
@@ -1043,27 +1090,32 @@ cifs_parse_mount_options(char *options, const char *devname,
 					printk(KERN_WARNING
 					       "CIFS: UNC Path does not begin "
 					       "with // or \\\\ \n");
-					return 1;
+					goto cifs_parse_mount_err;
 				}
 			} else {
 				printk(KERN_WARNING "CIFS: UNC name too long\n");
-				return 1;
+				goto cifs_parse_mount_err;
 			}
 		} else if ((strnicmp(data, "domain", 3) == 0)
 			   || (strnicmp(data, "workgroup", 5) == 0)) {
 			if (!value || !*value) {
 				printk(KERN_WARNING "CIFS: invalid domain name\n");
-				return 1;	/* needs_arg; */
+				goto cifs_parse_mount_err;
 			}
 			/* BB are there cases in which a comma can be valid in
 			a domain name and need special handling? */
 			if (strnlen(value, 256) < 256) {
-				vol->domainname = value;
+				vol->domainname = kstrdup(value, GFP_KERNEL);
+				if (!vol->domainname) {
+					printk(KERN_WARNING "CIFS: no memory "
+							    "for domainname\n");
+					goto cifs_parse_mount_err;
+				}
 				cFYI(1, "Domain name set");
 			} else {
 				printk(KERN_WARNING "CIFS: domain name too "
 						    "long\n");
-				return 1;
+				goto cifs_parse_mount_err;
 			}
 		} else if (strnicmp(data, "srcaddr", 7) == 0) {
 			vol->srcaddr.ss_family = AF_UNSPEC;
@@ -1071,7 +1123,7 @@ cifs_parse_mount_options(char *options, const char *devname,
 			if (!value || !*value) {
 				printk(KERN_WARNING "CIFS: srcaddr value"
 				       " not specified.\n");
-				return 1;	/* needs_arg; */
+				goto cifs_parse_mount_err;
 			}
 			i = cifs_convert_address((struct sockaddr *)&vol->srcaddr,
 						 value, strlen(value));
@@ -1079,20 +1131,20 @@ cifs_parse_mount_options(char *options, const char *devname,
 				printk(KERN_WARNING "CIFS:  Could not parse"
 				       " srcaddr: %s\n",
 				       value);
-				return 1;
+				goto cifs_parse_mount_err;
 			}
 		} else if (strnicmp(data, "prefixpath", 10) == 0) {
 			if (!value || !*value) {
 				printk(KERN_WARNING
 					"CIFS: invalid path prefix\n");
-				return 1;       /* needs_argument */
+				goto cifs_parse_mount_err;
 			}
 			if ((temp_len = strnlen(value, 1024)) < 1024) {
 				if (value[0] != '/')
 					temp_len++;  /* missing leading slash */
 				vol->prepath = kmalloc(temp_len+1, GFP_KERNEL);
 				if (vol->prepath == NULL)
-					return 1;
+					goto cifs_parse_mount_err;
 				if (value[0] != '/') {
 					vol->prepath[0] = '/';
 					strcpy(vol->prepath+1, value);
@@ -1101,24 +1153,33 @@ cifs_parse_mount_options(char *options, const char *devname,
 				cFYI(1, "prefix path %s", vol->prepath);
 			} else {
 				printk(KERN_WARNING "CIFS: prefix too long\n");
-				return 1;
+				goto cifs_parse_mount_err;
 			}
 		} else if (strnicmp(data, "iocharset", 9) == 0) {
 			if (!value || !*value) {
 				printk(KERN_WARNING "CIFS: invalid iocharset "
 						    "specified\n");
-				return 1;	/* needs_arg; */
+				goto cifs_parse_mount_err;
 			}
 			if (strnlen(value, 65) < 65) {
-				if (strnicmp(value, "default", 7))
-					vol->iocharset = value;
+				if (strnicmp(value, "default", 7)) {
+					vol->iocharset = kstrdup(value,
+								 GFP_KERNEL);
+
+					if (!vol->iocharset) {
+						printk(KERN_WARNING "CIFS: no "
+								   "memory for"
+								   "charset\n");
+						goto cifs_parse_mount_err;
+					}
+				}
 				/* if iocharset not set then load_nls_default
 				   is used by caller */
 				cFYI(1, "iocharset set to %s", value);
 			} else {
 				printk(KERN_WARNING "CIFS: iocharset name "
 						    "too long.\n");
-				return 1;
+				goto cifs_parse_mount_err;
 			}
 		} else if (!strnicmp(data, "uid", 3) && value && *value) {
 			vol->linux_uid = simple_strtoul(value, &value, 0);
@@ -1231,7 +1292,7 @@ cifs_parse_mount_options(char *options, const char *devname,
 				if (vol->actimeo > CIFS_MAX_ACTIMEO) {
 					cERROR(1, "CIFS: attribute cache"
 							"timeout too large");
-					return 1;
+					goto cifs_parse_mount_err;
 				}
 			}
 		} else if (strnicmp(data, "credentials", 4) == 0) {
@@ -1375,7 +1436,7 @@ cifs_parse_mount_options(char *options, const char *devname,
 #ifndef CONFIG_CIFS_FSCACHE
 			cERROR(1, "FS-Cache support needs CONFIG_CIFS_FSCACHE"
 				  "kernel config option set");
-			return 1;
+			goto cifs_parse_mount_err;
 #endif
 			vol->fsc = true;
 		} else if (strnicmp(data, "mfsymlinks", 10) == 0) {
@@ -1390,12 +1451,12 @@ cifs_parse_mount_options(char *options, const char *devname,
 		if (devname == NULL) {
 			printk(KERN_WARNING "CIFS: Missing UNC name for mount "
 						"target\n");
-			return 1;
+			goto cifs_parse_mount_err;
 		}
 		if ((temp_len = strnlen(devname, 300)) < 300) {
 			vol->UNC = kmalloc(temp_len+1, GFP_KERNEL);
 			if (vol->UNC == NULL)
-				return 1;
+				goto cifs_parse_mount_err;
 			strcpy(vol->UNC, devname);
 			if (strncmp(vol->UNC, "//", 2) == 0) {
 				vol->UNC[0] = '\\';
@@ -1403,21 +1464,21 @@ cifs_parse_mount_options(char *options, const char *devname,
 			} else if (strncmp(vol->UNC, "\\\\", 2) != 0) {
 				printk(KERN_WARNING "CIFS: UNC Path does not "
 						    "begin with // or \\\\ \n");
-				return 1;
+				goto cifs_parse_mount_err;
 			}
 			value = strpbrk(vol->UNC+2, "/\\");
 			if (value)
 				*value = '\\';
 		} else {
 			printk(KERN_WARNING "CIFS: UNC name too long\n");
-			return 1;
+			goto cifs_parse_mount_err;
 		}
 	}
 
 	if (vol->multiuser && !(vol->secFlg & CIFSSEC_MAY_KRB5)) {
 		cERROR(1, "Multiuser mounts currently require krb5 "
 			  "authentication!");
-		return 1;
+		goto cifs_parse_mount_err;
 	}
 
 	if (vol->UNCip == NULL)
@@ -1435,7 +1496,12 @@ cifs_parse_mount_options(char *options, const char *devname,
 		printk(KERN_NOTICE "CIFS: ignoring forcegid mount option "
 				   "specified with no gid= option.\n");
 
+	kfree(mountdata_copy);
 	return 0;
+
+cifs_parse_mount_err:
+	kfree(mountdata_copy);
+	return 1;
 }
 
 /** Returns true if srcaddr isn't specified and rhs isn't
@@ -2265,7 +2331,7 @@ ip_rfc1001_connect(struct TCP_Server_Info *server)
 		smb_buf = (struct smb_hdr *)ses_init_buf;
 
 		/* sizeof RFC1002_SESSION_REQUEST with no scope */
-		smb_buf->smb_buf_length = 0x81000044;
+		smb_buf->smb_buf_length = cpu_to_be32(0x81000044);
 		rc = smb_send(server, smb_buf, 0x44);
 		kfree(ses_init_buf);
 		/*
@@ -2658,6 +2724,11 @@ is_path_accessible(int xid, struct cifsTconInfo *tcon,
 			      0 /* not legacy */, cifs_sb->local_nls,
 			      cifs_sb->mnt_cifs_flags &
 				CIFS_MOUNT_MAP_SPECIAL_CHR);
+
+	if (rc == -EOPNOTSUPP || rc == -EINVAL)
+		rc = SMBQueryInformation(xid, tcon, full_path, pfile_info,
+				cifs_sb->local_nls, cifs_sb->mnt_cifs_flags &
+				  CIFS_MOUNT_MAP_SPECIAL_CHR);
 	kfree(pfile_info);
 	return rc;
 }
@@ -2671,8 +2742,12 @@ cleanup_volume_info(struct smb_vol **pvolume_info)
 		return;
 
 	volume_info = *pvolume_info;
+	kfree(volume_info->username);
 	kzfree(volume_info->password);
 	kfree(volume_info->UNC);
+	kfree(volume_info->UNCip);
+	kfree(volume_info->domainname);
+	kfree(volume_info->iocharset);
 	kfree(volume_info->prepath);
 	kfree(volume_info);
 	*pvolume_info = NULL;
@@ -2709,11 +2784,65 @@ build_unc_path_to_root(const struct smb_vol *volume_info,
 	full_path[unc_len + cifs_sb->prepathlen] = 0; /* add trailing null */
 	return full_path;
 }
+
+/*
+ * Perform a dfs referral query for a share and (optionally) prefix
+ *
+ * If a referral is found, cifs_sb->mountdata will be (re-)allocated
+ * to a string containing updated options for the submount.  Otherwise it
+ * will be left untouched.
+ *
+ * Returns the rc from get_dfs_path to the caller, which can be used to
+ * determine whether there were referrals.
+ */
+static int
+expand_dfs_referral(int xid, struct cifsSesInfo *pSesInfo,
+		    struct smb_vol *volume_info, struct cifs_sb_info *cifs_sb,
+		    int check_prefix)
+{
+	int rc;
+	unsigned int num_referrals = 0;
+	struct dfs_info3_param *referrals = NULL;
+	char *full_path = NULL, *ref_path = NULL, *mdata = NULL;
+
+	full_path = build_unc_path_to_root(volume_info, cifs_sb);
+	if (IS_ERR(full_path))
+		return PTR_ERR(full_path);
+
+	/* For DFS paths, skip the first '\' of the UNC */
+	ref_path = check_prefix ? full_path + 1 : volume_info->UNC + 1;
+
+	rc = get_dfs_path(xid, pSesInfo , ref_path, cifs_sb->local_nls,
+			  &num_referrals, &referrals,
+			  cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
+
+	if (!rc && num_referrals > 0) {
+		char *fake_devname = NULL;
+
+		mdata = cifs_compose_mount_options(cifs_sb->mountdata,
+						   full_path + 1, referrals,
+						   &fake_devname);
+
+		free_dfs_info_array(referrals, num_referrals);
+		kfree(fake_devname);
+
+		if (cifs_sb->mountdata != NULL)
+			kfree(cifs_sb->mountdata);
+
+		if (IS_ERR(mdata)) {
+			rc = PTR_ERR(mdata);
+			mdata = NULL;
+		}
+		cifs_sb->mountdata = mdata;
+	}
+	kfree(full_path);
+	return rc;
+}
 #endif
 
 int
 cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb,
-		char *mount_data_global, const char *devname)
+		const char *devname)
 {
 	int rc;
 	int xid;
@@ -2722,13 +2851,20 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb,
 	struct cifsTconInfo *tcon;
 	struct TCP_Server_Info *srvTcp;
 	char   *full_path;
-	char *mount_data = mount_data_global;
 	struct tcon_link *tlink;
 #ifdef CONFIG_CIFS_DFS_UPCALL
-	struct dfs_info3_param *referrals = NULL;
-	unsigned int num_referrals = 0;
 	int referral_walks_count = 0;
 try_mount_again:
+	/* cleanup activities if we're chasing a referral */
+	if (referral_walks_count) {
+		if (tcon)
+			cifs_put_tcon(tcon);
+		else if (pSesInfo)
+			cifs_put_smb_ses(pSesInfo);
+
+		cleanup_volume_info(&volume_info);
+		FreeXid(xid);
+	}
 #endif
 	rc = 0;
 	tcon = NULL;
@@ -2745,7 +2881,8 @@ try_mount_again:
 		goto out;
 	}
 
-	if (cifs_parse_mount_options(mount_data, devname, volume_info)) {
+	if (cifs_parse_mount_options(cifs_sb->mountdata, devname,
+				     volume_info)) {
 		rc = -EINVAL;
 		goto out;
 	}
@@ -2841,6 +2978,24 @@ try_mount_again:
 			       (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE));
 
 remote_path_check:
+#ifdef CONFIG_CIFS_DFS_UPCALL
+	/*
+	 * Perform an unconditional check for whether there are DFS
+	 * referrals for this path without prefix, to provide support
+	 * for DFS referrals from w2k8 servers which don't seem to respond
+	 * with PATH_NOT_COVERED to requests that include the prefix.
+	 * Chase the referral if found, otherwise continue normally.
+	 */
+	if (referral_walks_count == 0) {
+		int refrc = expand_dfs_referral(xid, pSesInfo, volume_info,
+						cifs_sb, false);
+		if (!refrc) {
+			referral_walks_count++;
+			goto try_mount_again;
+		}
+	}
+#endif
+
 	/* check if a whole path (including prepath) is not remote */
 	if (!rc && tcon) {
 		/* build_path_to_root works only when we have a valid tcon */
@@ -2874,46 +3029,15 @@ remote_path_check:
 		if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) == 0)
 			convert_delimiter(cifs_sb->prepath,
 					CIFS_DIR_SEP(cifs_sb));
-		full_path = build_unc_path_to_root(volume_info, cifs_sb);
-		if (IS_ERR(full_path)) {
-			rc = PTR_ERR(full_path);
-			goto mount_fail_check;
-		}
-
-		cFYI(1, "Getting referral for: %s", full_path);
-		rc = get_dfs_path(xid, pSesInfo , full_path + 1,
-			cifs_sb->local_nls, &num_referrals, &referrals,
-			cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
-		if (!rc && num_referrals > 0) {
-			char *fake_devname = NULL;
-
-			if (mount_data != mount_data_global)
-				kfree(mount_data);
 
-			mount_data = cifs_compose_mount_options(
-					cifs_sb->mountdata, full_path + 1,
-					referrals, &fake_devname);
+		rc = expand_dfs_referral(xid, pSesInfo, volume_info, cifs_sb,
+					 true);
 
-			free_dfs_info_array(referrals, num_referrals);
-			kfree(fake_devname);
-			kfree(full_path);
-
-			if (IS_ERR(mount_data)) {
-				rc = PTR_ERR(mount_data);
-				mount_data = NULL;
-				goto mount_fail_check;
-			}
-
-			if (tcon)
-				cifs_put_tcon(tcon);
-			else if (pSesInfo)
-				cifs_put_smb_ses(pSesInfo);
-
-			cleanup_volume_info(&volume_info);
+		if (!rc) {
 			referral_walks_count++;
-			FreeXid(xid);
 			goto try_mount_again;
 		}
+		goto mount_fail_check;
 #else /* No DFS support, return error on mount */
 		rc = -EOPNOTSUPP;
 #endif
@@ -2946,8 +3070,6 @@ remote_path_check:
 mount_fail_check:
 	/* on error free sesinfo and tcon struct if needed */
 	if (rc) {
-		if (mount_data != mount_data_global)
-			kfree(mount_data);
 		/* If find_unc succeeded then rc == 0 so we can not end */
 		/* up accidentally freeing someone elses tcon struct */
 		if (tcon)
@@ -3063,7 +3185,8 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses,
 	bcc_ptr += strlen("?????");
 	bcc_ptr += 1;
 	count = bcc_ptr - &pSMB->Password[0];
-	pSMB->hdr.smb_buf_length += count;
+	pSMB->hdr.smb_buf_length = cpu_to_be32(be32_to_cpu(
+					pSMB->hdr.smb_buf_length) + count);
 	pSMB->ByteCount = cpu_to_le16(count);
 
 	rc = SendReceive(xid, ses, smb_buffer, smb_buffer_response, &length,
@@ -3238,7 +3361,9 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, uid_t fsuid)
 	struct cifsSesInfo *ses;
 	struct cifsTconInfo *tcon = NULL;
 	struct smb_vol *vol_info;
-	char username[MAX_USERNAME_SIZE + 1];
+	char username[28]; /* big enough for "krb50x" + hex of ULONG_MAX 6+16 */
+			   /* We used to have this as MAX_USERNAME which is   */
+			   /* way too big now (256 instead of 32) */
 
 	vol_info = kzalloc(sizeof(*vol_info), GFP_KERNEL);
 	if (vol_info == NULL) {
diff --git a/fs/cifs/export.c b/fs/cifs/export.c
index 993f82045bf..55d87ac5200 100644
--- a/fs/cifs/export.c
+++ b/fs/cifs/export.c
@@ -45,7 +45,7 @@
 #include "cifs_debug.h"
 #include "cifsfs.h"
 
-#ifdef CONFIG_CIFS_EXPERIMENTAL
+#ifdef CIFS_NFSD_EXPORT
 static struct dentry *cifs_get_parent(struct dentry *dentry)
 {
 	/* BB need to add code here eventually to enable export via NFSD */
@@ -63,5 +63,5 @@ const struct export_operations cifs_export_ops = {
 	.encode_fs =  */
 };
 
-#endif /* EXPERIMENTAL */
+#endif /* CIFS_NFSD_EXPORT */
 
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index faf59529e84..c672afef0c0 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -857,95 +857,6 @@ cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset,
 		cifsi->server_eof = end_of_write;
 }
 
-ssize_t cifs_user_write(struct file *file, const char __user *write_data,
-	size_t write_size, loff_t *poffset)
-{
-	struct inode *inode = file->f_path.dentry->d_inode;
-	int rc = 0;
-	unsigned int bytes_written = 0;
-	unsigned int total_written;
-	struct cifs_sb_info *cifs_sb;
-	struct cifsTconInfo *pTcon;
-	int xid;
-	struct cifsFileInfo *open_file;
-	struct cifsInodeInfo *cifsi = CIFS_I(inode);
-
-	cifs_sb = CIFS_SB(file->f_path.dentry->d_sb);
-
-	/* cFYI(1, " write %d bytes to offset %lld of %s", write_size,
-	   *poffset, file->f_path.dentry->d_name.name); */
-
-	if (file->private_data == NULL)
-		return -EBADF;
-
-	open_file = file->private_data;
-	pTcon = tlink_tcon(open_file->tlink);
-
-	rc = generic_write_checks(file, poffset, &write_size, 0);
-	if (rc)
-		return rc;
-
-	xid = GetXid();
-
-	for (total_written = 0; write_size > total_written;
-	     total_written += bytes_written) {
-		rc = -EAGAIN;
-		while (rc == -EAGAIN) {
-			if (file->private_data == NULL) {
-				/* file has been closed on us */
-				FreeXid(xid);
-			/* if we have gotten here we have written some data
-			   and blocked, and the file has been freed on us while
-			   we blocked so return what we managed to write */
-				return total_written;
-			}
-			if (open_file->invalidHandle) {
-				/* we could deadlock if we called
-				   filemap_fdatawait from here so tell
-				   reopen_file not to flush data to server
-				   now */
-				rc = cifs_reopen_file(open_file, false);
-				if (rc != 0)
-					break;
-			}
-
-			rc = CIFSSMBWrite(xid, pTcon,
-				open_file->netfid,
-				min_t(const int, cifs_sb->wsize,
-				      write_size - total_written),
-				*poffset, &bytes_written,
-				NULL, write_data + total_written, 0);
-		}
-		if (rc || (bytes_written == 0)) {
-			if (total_written)
-				break;
-			else {
-				FreeXid(xid);
-				return rc;
-			}
-		} else {
-			cifs_update_eof(cifsi, *poffset, bytes_written);
-			*poffset += bytes_written;
-		}
-	}
-
-	cifs_stats_bytes_written(pTcon, total_written);
-
-/* Do not update local mtime - server will set its actual value on write
- *	inode->i_ctime = inode->i_mtime =
- * 		current_fs_time(inode->i_sb);*/
-	if (total_written > 0) {
-		spin_lock(&inode->i_lock);
-		if (*poffset > inode->i_size)
-			i_size_write(inode, *poffset);
-		spin_unlock(&inode->i_lock);
-	}
-	mark_inode_dirty_sync(inode);
-
-	FreeXid(xid);
-	return total_written;
-}
-
 static ssize_t cifs_write(struct cifsFileInfo *open_file,
 			  const char *write_data, size_t write_size,
 			  loff_t *poffset)
@@ -1420,9 +1331,10 @@ retry_write:
 	return rc;
 }
 
-static int cifs_writepage(struct page *page, struct writeback_control *wbc)
+static int
+cifs_writepage_locked(struct page *page, struct writeback_control *wbc)
 {
-	int rc = -EFAULT;
+	int rc;
 	int xid;
 
 	xid = GetXid();
@@ -1442,15 +1354,29 @@ static int cifs_writepage(struct page *page, struct writeback_control *wbc)
 	 * to fail to update with the state of the page correctly.
 	 */
 	set_page_writeback(page);
+retry_write:
 	rc = cifs_partialpagewrite(page, 0, PAGE_CACHE_SIZE);
-	SetPageUptodate(page); /* BB add check for error and Clearuptodate? */
-	unlock_page(page);
+	if (rc == -EAGAIN && wbc->sync_mode == WB_SYNC_ALL)
+		goto retry_write;
+	else if (rc == -EAGAIN)
+		redirty_page_for_writepage(wbc, page);
+	else if (rc != 0)
+		SetPageError(page);
+	else
+		SetPageUptodate(page);
 	end_page_writeback(page);
 	page_cache_release(page);
 	FreeXid(xid);
 	return rc;
 }
 
+static int cifs_writepage(struct page *page, struct writeback_control *wbc)
+{
+	int rc = cifs_writepage_locked(page, wbc);
+	unlock_page(page);
+	return rc;
+}
+
 static int cifs_write_end(struct file *file, struct address_space *mapping,
 			loff_t pos, unsigned len, unsigned copied,
 			struct page *page, void *fsdata)
@@ -1519,8 +1445,13 @@ int cifs_strict_fsync(struct file *file, int datasync)
 	cFYI(1, "Sync file - name: %s datasync: 0x%x",
 		file->f_path.dentry->d_name.name, datasync);
 
-	if (!CIFS_I(inode)->clientCanCacheRead)
-		cifs_invalidate_mapping(inode);
+	if (!CIFS_I(inode)->clientCanCacheRead) {
+		rc = cifs_invalidate_mapping(inode);
+		if (rc) {
+			cFYI(1, "rc: %d during invalidate phase", rc);
+			rc = 0; /* don't care about it in fsync */
+		}
+	}
 
 	tcon = tlink_tcon(smbfile->tlink);
 	if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC))
@@ -1726,7 +1657,7 @@ cifs_iovec_write(struct file *file, const struct iovec *iov,
 	return total_written;
 }
 
-static ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov,
+ssize_t cifs_user_writev(struct kiocb *iocb, const struct iovec *iov,
 				unsigned long nr_segs, loff_t pos)
 {
 	ssize_t written;
@@ -1849,17 +1780,7 @@ cifs_iovec_read(struct file *file, const struct iovec *iov,
 	return total_read;
 }
 
-ssize_t cifs_user_read(struct file *file, char __user *read_data,
-		       size_t read_size, loff_t *poffset)
-{
-	struct iovec iov;
-	iov.iov_base = read_data;
-	iov.iov_len = read_size;
-
-	return cifs_iovec_read(file, &iov, 1, poffset);
-}
-
-static ssize_t cifs_user_readv(struct kiocb *iocb, const struct iovec *iov,
+ssize_t cifs_user_readv(struct kiocb *iocb, const struct iovec *iov,
 			       unsigned long nr_segs, loff_t pos)
 {
 	ssize_t read;
@@ -1987,8 +1908,11 @@ int cifs_file_strict_mmap(struct file *file, struct vm_area_struct *vma)
 
 	xid = GetXid();
 
-	if (!CIFS_I(inode)->clientCanCacheRead)
-		cifs_invalidate_mapping(inode);
+	if (!CIFS_I(inode)->clientCanCacheRead) {
+		rc = cifs_invalidate_mapping(inode);
+		if (rc)
+			return rc;
+	}
 
 	rc = generic_file_mmap(file, vma);
 	if (rc == 0)
@@ -2415,6 +2339,27 @@ static void cifs_invalidate_page(struct page *page, unsigned long offset)
 		cifs_fscache_invalidate_page(page, &cifsi->vfs_inode);
 }
 
+static int cifs_launder_page(struct page *page)
+{
+	int rc = 0;
+	loff_t range_start = page_offset(page);
+	loff_t range_end = range_start + (loff_t)(PAGE_CACHE_SIZE - 1);
+	struct writeback_control wbc = {
+		.sync_mode = WB_SYNC_ALL,
+		.nr_to_write = 0,
+		.range_start = range_start,
+		.range_end = range_end,
+	};
+
+	cFYI(1, "Launder page: %p", page);
+
+	if (clear_page_dirty_for_io(page))
+		rc = cifs_writepage_locked(page, &wbc);
+
+	cifs_fscache_invalidate_page(page, page->mapping->host);
+	return rc;
+}
+
 void cifs_oplock_break(struct work_struct *work)
 {
 	struct cifsFileInfo *cfile = container_of(work, struct cifsFileInfo,
@@ -2486,7 +2431,7 @@ const struct address_space_operations cifs_addr_ops = {
 	.set_page_dirty = __set_page_dirty_nobuffers,
 	.releasepage = cifs_release_page,
 	.invalidatepage = cifs_invalidate_page,
-	/* .direct_IO = */
+	.launder_page = cifs_launder_page,
 };
 
 /*
@@ -2503,5 +2448,5 @@ const struct address_space_operations cifs_addr_ops_smallbuf = {
 	.set_page_dirty = __set_page_dirty_nobuffers,
 	.releasepage = cifs_release_page,
 	.invalidatepage = cifs_invalidate_page,
-	/* .direct_IO = */
+	.launder_page = cifs_launder_page,
 };
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 8852470b4fb..de02ed5e25c 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -878,7 +878,7 @@ retry_iget5_locked:
 }
 
 /* gets root inode */
-struct inode *cifs_root_iget(struct super_block *sb, unsigned long ino)
+struct inode *cifs_root_iget(struct super_block *sb)
 {
 	int xid;
 	struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
@@ -1683,71 +1683,70 @@ cifs_inode_needs_reval(struct inode *inode)
 /*
  * Zap the cache. Called when invalid_mapping flag is set.
  */
-void
+int
 cifs_invalidate_mapping(struct inode *inode)
 {
-	int rc;
+	int rc = 0;
 	struct cifsInodeInfo *cifs_i = CIFS_I(inode);
 
 	cifs_i->invalid_mapping = false;
 
-	/* write back any cached data */
 	if (inode->i_mapping && inode->i_mapping->nrpages != 0) {
-		rc = filemap_write_and_wait(inode->i_mapping);
-		mapping_set_error(inode->i_mapping, rc);
+		rc = invalidate_inode_pages2(inode->i_mapping);
+		if (rc) {
+			cERROR(1, "%s: could not invalidate inode %p", __func__,
+			       inode);
+			cifs_i->invalid_mapping = true;
+		}
 	}
-	invalidate_remote_inode(inode);
+
 	cifs_fscache_reset_inode_cookie(inode);
+	return rc;
 }
 
-int cifs_revalidate_file(struct file *filp)
+int cifs_revalidate_file_attr(struct file *filp)
 {
 	int rc = 0;
 	struct inode *inode = filp->f_path.dentry->d_inode;
 	struct cifsFileInfo *cfile = (struct cifsFileInfo *) filp->private_data;
 
 	if (!cifs_inode_needs_reval(inode))
-		goto check_inval;
+		return rc;
 
 	if (tlink_tcon(cfile->tlink)->unix_ext)
 		rc = cifs_get_file_info_unix(filp);
 	else
 		rc = cifs_get_file_info(filp);
 
-check_inval:
-	if (CIFS_I(inode)->invalid_mapping)
-		cifs_invalidate_mapping(inode);
-
 	return rc;
 }
 
-/* revalidate a dentry's inode attributes */
-int cifs_revalidate_dentry(struct dentry *dentry)
+int cifs_revalidate_dentry_attr(struct dentry *dentry)
 {
 	int xid;
 	int rc = 0;
-	char *full_path = NULL;
 	struct inode *inode = dentry->d_inode;
 	struct super_block *sb = dentry->d_sb;
+	char *full_path = NULL;
 
 	if (inode == NULL)
 		return -ENOENT;
 
-	xid = GetXid();
-
 	if (!cifs_inode_needs_reval(inode))
-		goto check_inval;
+		return rc;
+
+	xid = GetXid();
 
 	/* can not safely grab the rename sem here if rename calls revalidate
 	   since that would deadlock */
 	full_path = build_path_from_dentry(dentry);
 	if (full_path == NULL) {
 		rc = -ENOMEM;
-		goto check_inval;
+		goto out;
 	}
 
-	cFYI(1, "Revalidate: %s inode 0x%p count %d dentry: 0x%p d_time %ld "
-		 "jiffies %ld", full_path, inode, inode->i_count.counter,
+	cFYI(1, "Update attributes: %s inode 0x%p count %d dentry: 0x%p d_time "
+		 "%ld jiffies %ld", full_path, inode, inode->i_count.counter,
 		 dentry, dentry->d_time, jiffies);
 
 	if (cifs_sb_master_tcon(CIFS_SB(sb))->unix_ext)
@@ -1756,41 +1755,83 @@ int cifs_revalidate_dentry(struct dentry *dentry)
 		rc = cifs_get_inode_info(&inode, full_path, NULL, sb,
 					 xid, NULL);
 
-check_inval:
-	if (CIFS_I(inode)->invalid_mapping)
-		cifs_invalidate_mapping(inode);
-
+out:
 	kfree(full_path);
 	FreeXid(xid);
 	return rc;
 }
 
+int cifs_revalidate_file(struct file *filp)
+{
+	int rc;
+	struct inode *inode = filp->f_path.dentry->d_inode;
+
+	rc = cifs_revalidate_file_attr(filp);
+	if (rc)
+		return rc;
+
+	if (CIFS_I(inode)->invalid_mapping)
+		rc = cifs_invalidate_mapping(inode);
+	return rc;
+}
+
+/* revalidate a dentry's inode attributes */
+int cifs_revalidate_dentry(struct dentry *dentry)
+{
+	int rc;
+	struct inode *inode = dentry->d_inode;
+
+	rc = cifs_revalidate_dentry_attr(dentry);
+	if (rc)
+		return rc;
+
+	if (CIFS_I(inode)->invalid_mapping)
+		rc = cifs_invalidate_mapping(inode);
+	return rc;
+}
+
 int cifs_getattr(struct vfsmount *mnt, struct dentry *dentry,
 		 struct kstat *stat)
 {
 	struct cifs_sb_info *cifs_sb = CIFS_SB(dentry->d_sb);
 	struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb);
-	int err = cifs_revalidate_dentry(dentry);
-
-	if (!err) {
-		generic_fillattr(dentry->d_inode, stat);
-		stat->blksize = CIFS_MAX_MSGSIZE;
-		stat->ino = CIFS_I(dentry->d_inode)->uniqueid;
+	struct inode *inode = dentry->d_inode;
+	int rc;
 
-		/*
-		 * If on a multiuser mount without unix extensions, and the
-		 * admin hasn't overridden them, set the ownership to the
-		 * fsuid/fsgid of the current process.
-		 */
-		if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER) &&
-		    !tcon->unix_ext) {
-			if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID))
-				stat->uid = current_fsuid();
-			if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID))
-				stat->gid = current_fsgid();
+	/*
+	 * We need to be sure that all dirty pages are written and the server
+	 * has actual ctime, mtime and file length.
+	 */
+	if (!CIFS_I(inode)->clientCanCacheRead && inode->i_mapping &&
+	    inode->i_mapping->nrpages != 0) {
+		rc = filemap_fdatawait(inode->i_mapping);
+		if (rc) {
+			mapping_set_error(inode->i_mapping, rc);
+			return rc;
 		}
 	}
-	return err;
+
+	rc = cifs_revalidate_dentry_attr(dentry);
+	if (rc)
+		return rc;
+
+	generic_fillattr(inode, stat);
+	stat->blksize = CIFS_MAX_MSGSIZE;
+	stat->ino = CIFS_I(inode)->uniqueid;
+
+	/*
+	 * If on a multiuser mount without unix extensions, and the admin hasn't
+	 * overridden them, set the ownership to the fsuid/fsgid of the current
+	 * process.
+	 */
+	if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER) &&
+	    !tcon->unix_ext) {
+		if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_UID))
+			stat->uid = current_fsuid();
+		if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID))
+			stat->gid = current_fsgid();
+	}
+	return rc;
 }
 
 static int cifs_truncate_page(struct address_space *mapping, loff_t from)
diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c
index 0c684ae4c07..907531ac588 100644
--- a/fs/cifs/misc.c
+++ b/fs/cifs/misc.c
@@ -304,12 +304,10 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ ,
 
 	memset(temp, 0, 256); /* bigger than MAX_CIFS_HDR_SIZE */
 
-	buffer->smb_buf_length =
+	buffer->smb_buf_length = cpu_to_be32(
 	    (2 * word_count) + sizeof(struct smb_hdr) -
 	    4 /*  RFC 1001 length field does not count */  +
-	    2 /* for bcc field itself */ ;
-	/* Note that this is the only network field that has to be converted
-	   to big endian and it is done just before we send it */
+	    2 /* for bcc field itself */) ;
 
 	buffer->Protocol[0] = 0xFF;
 	buffer->Protocol[1] = 'S';
@@ -424,7 +422,7 @@ check_smb_hdr(struct smb_hdr *smb, __u16 mid)
 int
 checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length)
 {
-	__u32 len = smb->smb_buf_length;
+	__u32 len = be32_to_cpu(smb->smb_buf_length);
 	__u32 clc_len;  /* calculated length */
 	cFYI(0, "checkSMB Length: 0x%x, smb_buf_length: 0x%x", length, len);
 
@@ -464,7 +462,7 @@ checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length)
 
 	if (check_smb_hdr(smb, mid))
 		return 1;
-	clc_len = smbCalcSize_LE(smb);
+	clc_len = smbCalcSize(smb);
 
 	if (4 + len != length) {
 		cERROR(1, "Length read does not match RFC1001 length %d",
@@ -521,7 +519,7 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv)
 			(struct smb_com_transaction_change_notify_rsp *)buf;
 		struct file_notify_information *pnotify;
 		__u32 data_offset = 0;
-		if (get_bcc_le(buf) > sizeof(struct file_notify_information)) {
+		if (get_bcc(buf) > sizeof(struct file_notify_information)) {
 			data_offset = le32_to_cpu(pSMBr->DataOffset);
 
 			pnotify = (struct file_notify_information *)
diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c
index 79f641eeda3..79b71c2c7c9 100644
--- a/fs/cifs/netmisc.c
+++ b/fs/cifs/netmisc.c
@@ -919,13 +919,6 @@ smbCalcSize(struct smb_hdr *ptr)
 		2 /* size of the bcc field */ + get_bcc(ptr));
 }
 
-unsigned int
-smbCalcSize_LE(struct smb_hdr *ptr)
-{
-	return (sizeof(struct smb_hdr) + (2 * ptr->WordCount) +
-		2 /* size of the bcc field */ + get_bcc_le(ptr));
-}
-
 /* The following are taken from fs/ntfs/util.c */
 
 #define NTFS_TIME_OFFSET ((u64)(369*365 + 89) * 24 * 3600 * 10000000)
diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c
index f6728eb6f4b..7dd46210037 100644
--- a/fs/cifs/sess.c
+++ b/fs/cifs/sess.c
@@ -276,7 +276,7 @@ static void ascii_ssetup_strings(char **pbcc_area, struct cifsSesInfo *ses,
 }
 
 static void
-decode_unicode_ssetup(char **pbcc_area, __u16 bleft, struct cifsSesInfo *ses,
+decode_unicode_ssetup(char **pbcc_area, int bleft, struct cifsSesInfo *ses,
 		      const struct nls_table *nls_cp)
 {
 	int len;
@@ -284,19 +284,6 @@ decode_unicode_ssetup(char **pbcc_area, __u16 bleft, struct cifsSesInfo *ses,
 
 	cFYI(1, "bleft %d", bleft);
 
-	/*
-	 * Windows servers do not always double null terminate their final
-	 * Unicode string. Check to see if there are an uneven number of bytes
-	 * left. If so, then add an extra NULL pad byte to the end of the
-	 * response.
-	 *
-	 * See section 2.7.2 in "Implementing CIFS" for details
-	 */
-	if (bleft % 2) {
-		data[bleft] = 0;
-		++bleft;
-	}
-
 	kfree(ses->serverOS);
 	ses->serverOS = cifs_strndup_from_ucs(data, bleft, true, nls_cp);
 	cFYI(1, "serverOS=%s", ses->serverOS);
@@ -634,7 +621,7 @@ ssetup_ntlmssp_authenticate:
 	and rest of bcc area. This allows us to avoid
 	a large buffer 17K allocation */
 	iov[0].iov_base = (char *)pSMB;
-	iov[0].iov_len = smb_buf->smb_buf_length + 4;
+	iov[0].iov_len = be32_to_cpu(smb_buf->smb_buf_length) + 4;
 
 	/* setting this here allows the code at the end of the function
 	   to free the request buffer if there's an error */
@@ -669,7 +656,7 @@ ssetup_ntlmssp_authenticate:
 		 * to use challenge/response method (i.e. Password bit is 1).
 		 */
 
-		calc_lanman_hash(ses->password, ses->server->cryptkey,
+		rc = calc_lanman_hash(ses->password, ses->server->cryptkey,
 				 ses->server->secMode & SECMODE_PW_ENCRYPT ?
 					true : false, lnm_session_key);
 
@@ -872,9 +859,10 @@ ssetup_ntlmssp_authenticate:
 	iov[2].iov_len = (long) bcc_ptr - (long) str_area;
 
 	count = iov[1].iov_len + iov[2].iov_len;
-	smb_buf->smb_buf_length += count;
+	smb_buf->smb_buf_length =
+		cpu_to_be32(be32_to_cpu(smb_buf->smb_buf_length) + count);
 
-	put_bcc_le(count, smb_buf);
+	put_bcc(count, smb_buf);
 
 	rc = SendReceive2(xid, ses, iov, 3 /* num_iovecs */, &resp_buf_type,
 			  CIFS_LOG_ERROR);
@@ -929,7 +917,9 @@ ssetup_ntlmssp_authenticate:
 	}
 
 	/* BB check if Unicode and decode strings */
-	if (smb_buf->Flags2 & SMBFLG2_UNICODE) {
+	if (bytes_remaining == 0) {
+		/* no string area to decode, do nothing */
+	} else if (smb_buf->Flags2 & SMBFLG2_UNICODE) {
 		/* unicode string area must be word-aligned */
 		if (((unsigned long) bcc_ptr - (unsigned long) smb_buf) % 2) {
 			++bcc_ptr;
diff --git a/fs/cifs/smbdes.c b/fs/cifs/smbdes.c
deleted file mode 100644
index 04721485925..00000000000
--- a/fs/cifs/smbdes.c
+++ /dev/null
@@ -1,418 +0,0 @@
-/*
-   Unix SMB/Netbios implementation.
-   Version 1.9.
-
-   a partial implementation of DES designed for use in the
-   SMB authentication protocol
-
-   Copyright (C) Andrew Tridgell 1998
-   Modified by Steve French (sfrench@us.ibm.com) 2002,2004
-
-   This program is free software; you can redistribute it and/or modify
-   it under the terms of the GNU General Public License as published by
-   the Free Software Foundation; either version 2 of the License, or
-   (at your option) any later version.
-
-   This program is distributed in the hope that it will be useful,
-   but WITHOUT ANY WARRANTY; without even the implied warranty of
-   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-   GNU General Public License for more details.
-
-   You should have received a copy of the GNU General Public License
-   along with this program; if not, write to the Free Software
-   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-*/
-
-/* NOTES:
-
-   This code makes no attempt to be fast! In fact, it is a very
-   slow implementation
-
-   This code is NOT a complete DES implementation. It implements only
-   the minimum necessary for SMB authentication, as used by all SMB
-   products (including every copy of Microsoft Windows95 ever sold)
-
-   In particular, it can only do a unchained forward DES pass. This
-   means it is not possible to use this code for encryption/decryption
-   of data, instead it is only useful as a "hash" algorithm.
-
-   There is no entry point into this code that allows normal DES operation.
-
-   I believe this means that this code does not come under ITAR
-   regulations but this is NOT a legal opinion. If you are concerned
-   about the applicability of ITAR regulations to this code then you
-   should confirm it for yourself (and maybe let me know if you come
-   up with a different answer to the one above)
-*/
-#include <linux/slab.h>
-#define uchar unsigned char
-
-static uchar perm1[56] = { 57, 49, 41, 33, 25, 17, 9,
-	1, 58, 50, 42, 34, 26, 18,
-	10, 2, 59, 51, 43, 35, 27,
-	19, 11, 3, 60, 52, 44, 36,
-	63, 55, 47, 39, 31, 23, 15,
-	7, 62, 54, 46, 38, 30, 22,
-	14, 6, 61, 53, 45, 37, 29,
-	21, 13, 5, 28, 20, 12, 4
-};
-
-static uchar perm2[48] = { 14, 17, 11, 24, 1, 5,
-	3, 28, 15, 6, 21, 10,
-	23, 19, 12, 4, 26, 8,
-	16, 7, 27, 20, 13, 2,
-	41, 52, 31, 37, 47, 55,
-	30, 40, 51, 45, 33, 48,
-	44, 49, 39, 56, 34, 53,
-	46, 42, 50, 36, 29, 32
-};
-
-static uchar perm3[64] = { 58, 50, 42, 34, 26, 18, 10, 2,
-	60, 52, 44, 36, 28, 20, 12, 4,
-	62, 54, 46, 38, 30, 22, 14, 6,
-	64, 56, 48, 40, 32, 24, 16, 8,
-	57, 49, 41, 33, 25, 17, 9, 1,
-	59, 51, 43, 35, 27, 19, 11, 3,
-	61, 53, 45, 37, 29, 21, 13, 5,
-	63, 55, 47, 39, 31, 23, 15, 7
-};
-
-static uchar perm4[48] = { 32, 1, 2, 3, 4, 5,
-	4, 5, 6, 7, 8, 9,
-	8, 9, 10, 11, 12, 13,
-	12, 13, 14, 15, 16, 17,
-	16, 17, 18, 19, 20, 21,
-	20, 21, 22, 23, 24, 25,
-	24, 25, 26, 27, 28, 29,
-	28, 29, 30, 31, 32, 1
-};
-
-static uchar perm5[32] = { 16, 7, 20, 21,
-	29, 12, 28, 17,
-	1, 15, 23, 26,
-	5, 18, 31, 10,
-	2, 8, 24, 14,
-	32, 27, 3, 9,
-	19, 13, 30, 6,
-	22, 11, 4, 25
-};
-
-static uchar perm6[64] = { 40, 8, 48, 16, 56, 24, 64, 32,
-	39, 7, 47, 15, 55, 23, 63, 31,
-	38, 6, 46, 14, 54, 22, 62, 30,
-	37, 5, 45, 13, 53, 21, 61, 29,
-	36, 4, 44, 12, 52, 20, 60, 28,
-	35, 3, 43, 11, 51, 19, 59, 27,
-	34, 2, 42, 10, 50, 18, 58, 26,
-	33, 1, 41, 9, 49, 17, 57, 25
-};
-
-static uchar sc[16] = { 1, 1, 2, 2, 2, 2, 2, 2, 1, 2, 2, 2, 2, 2, 2, 1 };
-
-static uchar sbox[8][4][16] = {
-	{{14, 4, 13, 1, 2, 15, 11, 8, 3, 10, 6, 12, 5, 9, 0, 7},
-	 {0, 15, 7, 4, 14, 2, 13, 1, 10, 6, 12, 11, 9, 5, 3, 8},
-	 {4, 1, 14, 8, 13, 6, 2, 11, 15, 12, 9, 7, 3, 10, 5, 0},
-	 {15, 12, 8, 2, 4, 9, 1, 7, 5, 11, 3, 14, 10, 0, 6, 13} },
-
-	{{15, 1, 8, 14, 6, 11, 3, 4, 9, 7, 2, 13, 12, 0, 5, 10},
-	 {3, 13, 4, 7, 15, 2, 8, 14, 12, 0, 1, 10, 6, 9, 11, 5},
-	 {0, 14, 7, 11, 10, 4, 13, 1, 5, 8, 12, 6, 9, 3, 2, 15},
-	 {13, 8, 10, 1, 3, 15, 4, 2, 11, 6, 7, 12, 0, 5, 14, 9} },
-
-	{{10, 0, 9, 14, 6, 3, 15, 5, 1, 13, 12, 7, 11, 4, 2, 8},
-	 {13, 7, 0, 9, 3, 4, 6, 10, 2, 8, 5, 14, 12, 11, 15, 1},
-	 {13, 6, 4, 9, 8, 15, 3, 0, 11, 1, 2, 12, 5, 10, 14, 7},
-	 {1, 10, 13, 0, 6, 9, 8, 7, 4, 15, 14, 3, 11, 5, 2, 12} },
-
-	{{7, 13, 14, 3, 0, 6, 9, 10, 1, 2, 8, 5, 11, 12, 4, 15},
-	 {13, 8, 11, 5, 6, 15, 0, 3, 4, 7, 2, 12, 1, 10, 14, 9},
-	 {10, 6, 9, 0, 12, 11, 7, 13, 15, 1, 3, 14, 5, 2, 8, 4},
-	 {3, 15, 0, 6, 10, 1, 13, 8, 9, 4, 5, 11, 12, 7, 2, 14} },
-
-	{{2, 12, 4, 1, 7, 10, 11, 6, 8, 5, 3, 15, 13, 0, 14, 9},
-	 {14, 11, 2, 12, 4, 7, 13, 1, 5, 0, 15, 10, 3, 9, 8, 6},
-	 {4, 2, 1, 11, 10, 13, 7, 8, 15, 9, 12, 5, 6, 3, 0, 14},
-	 {11, 8, 12, 7, 1, 14, 2, 13, 6, 15, 0, 9, 10, 4, 5, 3} },
-
-	{{12, 1, 10, 15, 9, 2, 6, 8, 0, 13, 3, 4, 14, 7, 5, 11},
-	 {10, 15, 4, 2, 7, 12, 9, 5, 6, 1, 13, 14, 0, 11, 3, 8},
-	 {9, 14, 15, 5, 2, 8, 12, 3, 7, 0, 4, 10, 1, 13, 11, 6},
-	 {4, 3, 2, 12, 9, 5, 15, 10, 11, 14, 1, 7, 6, 0, 8, 13} },
-
-	{{4, 11, 2, 14, 15, 0, 8, 13, 3, 12, 9, 7, 5, 10, 6, 1},
-	 {13, 0, 11, 7, 4, 9, 1, 10, 14, 3, 5, 12, 2, 15, 8, 6},
-	 {1, 4, 11, 13, 12, 3, 7, 14, 10, 15, 6, 8, 0, 5, 9, 2},
-	 {6, 11, 13, 8, 1, 4, 10, 7, 9, 5, 0, 15, 14, 2, 3, 12} },
-
-	{{13, 2, 8, 4, 6, 15, 11, 1, 10, 9, 3, 14, 5, 0, 12, 7},
-	 {1, 15, 13, 8, 10, 3, 7, 4, 12, 5, 6, 11, 0, 14, 9, 2},
-	 {7, 11, 4, 1, 9, 12, 14, 2, 0, 6, 10, 13, 15, 3, 5, 8},
-	 {2, 1, 14, 7, 4, 10, 8, 13, 15, 12, 9, 0, 3, 5, 6, 11} }
-};
-
-static void
-permute(char *out, char *in, uchar *p, int n)
-{
-	int i;
-	for (i = 0; i < n; i++)
-		out[i] = in[p[i] - 1];
-}
-
-static void
-lshift(char *d, int count, int n)
-{
-	char out[64];
-	int i;
-	for (i = 0; i < n; i++)
-		out[i] = d[(i + count) % n];
-	for (i = 0; i < n; i++)
-		d[i] = out[i];
-}
-
-static void
-concat(char *out, char *in1, char *in2, int l1, int l2)
-{
-	while (l1--)
-		*out++ = *in1++;
-	while (l2--)
-		*out++ = *in2++;
-}
-
-static void
-xor(char *out, char *in1, char *in2, int n)
-{
-	int i;
-	for (i = 0; i < n; i++)
-		out[i] = in1[i] ^ in2[i];
-}
-
-static void
-dohash(char *out, char *in, char *key, int forw)
-{
-	int i, j, k;
-	char *pk1;
-	char c[28];
-	char d[28];
-	char *cd;
-	char (*ki)[48];
-	char *pd1;
-	char l[32], r[32];
-	char *rl;
-
-	/* Have to reduce stack usage */
-	pk1 = kmalloc(56+56+64+64, GFP_KERNEL);
-	if (pk1 == NULL)
-		return;
-
-	ki = kmalloc(16*48, GFP_KERNEL);
-	if (ki == NULL) {
-		kfree(pk1);
-		return;
-	}
-
-	cd = pk1 + 56;
-	pd1 = cd  + 56;
-	rl = pd1 + 64;
-
-	permute(pk1, key, perm1, 56);
-
-	for (i = 0; i < 28; i++)
-		c[i] = pk1[i];
-	for (i = 0; i < 28; i++)
-		d[i] = pk1[i + 28];
-
-	for (i = 0; i < 16; i++) {
-		lshift(c, sc[i], 28);
-		lshift(d, sc[i], 28);
-
-		concat(cd, c, d, 28, 28);
-		permute(ki[i], cd, perm2, 48);
-	}
-
-	permute(pd1, in, perm3, 64);
-
-	for (j = 0; j < 32; j++) {
-		l[j] = pd1[j];
-		r[j] = pd1[j + 32];
-	}
-
-	for (i = 0; i < 16; i++) {
-		char *er;  /* er[48]  */
-		char *erk; /* erk[48] */
-		char b[8][6];
-		char *cb;  /* cb[32]  */
-		char *pcb; /* pcb[32] */
-		char *r2;  /* r2[32]  */
-
-		er = kmalloc(48+48+32+32+32, GFP_KERNEL);
-		if (er == NULL) {
-			kfree(pk1);
-			kfree(ki);
-			return;
-		}
-		erk = er+48;
-		cb  = erk+48;
-		pcb = cb+32;
-		r2  = pcb+32;
-
-		permute(er, r, perm4, 48);
-
-		xor(erk, er, ki[forw ? i : 15 - i], 48);
-
-		for (j = 0; j < 8; j++)
-			for (k = 0; k < 6; k++)
-				b[j][k] = erk[j * 6 + k];
-
-		for (j = 0; j < 8; j++) {
-			int m, n;
-			m = (b[j][0] << 1) | b[j][5];
-
-			n = (b[j][1] << 3) | (b[j][2] << 2) | (b[j][3] <<
-							       1) | b[j][4];
-
-			for (k = 0; k < 4; k++)
-				b[j][k] =
-				    (sbox[j][m][n] & (1 << (3 - k))) ? 1 : 0;
-		}
-
-		for (j = 0; j < 8; j++)
-			for (k = 0; k < 4; k++)
-				cb[j * 4 + k] = b[j][k];
-		permute(pcb, cb, perm5, 32);
-
-		xor(r2, l, pcb, 32);
-
-		for (j = 0; j < 32; j++)
-			l[j] = r[j];
-
-		for (j = 0; j < 32; j++)
-			r[j] = r2[j];
-
-		kfree(er);
-	}
-
-	concat(rl, r, l, 32, 32);
-
-	permute(out, rl, perm6, 64);
-	kfree(pk1);
-	kfree(ki);
-}
-
-static void
-str_to_key(unsigned char *str, unsigned char *key)
-{
-	int i;
-
-	key[0] = str[0] >> 1;
-	key[1] = ((str[0] & 0x01) << 6) | (str[1] >> 2);
-	key[2] = ((str[1] & 0x03) << 5) | (str[2] >> 3);
-	key[3] = ((str[2] & 0x07) << 4) | (str[3] >> 4);
-	key[4] = ((str[3] & 0x0F) << 3) | (str[4] >> 5);
-	key[5] = ((str[4] & 0x1F) << 2) | (str[5] >> 6);
-	key[6] = ((str[5] & 0x3F) << 1) | (str[6] >> 7);
-	key[7] = str[6] & 0x7F;
-	for (i = 0; i < 8; i++)
-		key[i] = (key[i] << 1);
-}
-
-static void
-smbhash(unsigned char *out, const unsigned char *in, unsigned char *key,
-	int forw)
-{
-	int i;
-	char *outb; /* outb[64] */
-	char *inb;  /* inb[64]  */
-	char *keyb; /* keyb[64] */
-	unsigned char key2[8];
-
-	outb = kmalloc(64 * 3, GFP_KERNEL);
-	if (outb == NULL)
-		return;
-
-	inb  = outb + 64;
-	keyb = inb +  64;
-
-	str_to_key(key, key2);
-
-	for (i = 0; i < 64; i++) {
-		inb[i] = (in[i / 8] & (1 << (7 - (i % 8)))) ? 1 : 0;
-		keyb[i] = (key2[i / 8] & (1 << (7 - (i % 8)))) ? 1 : 0;
-		outb[i] = 0;
-	}
-
-	dohash(outb, inb, keyb, forw);
-
-	for (i = 0; i < 8; i++)
-		out[i] = 0;
-
-	for (i = 0; i < 64; i++) {
-		if (outb[i])
-			out[i / 8] |= (1 << (7 - (i % 8)));
-	}
-	kfree(outb);
-}
-
-void
-E_P16(unsigned char *p14, unsigned char *p16)
-{
-	unsigned char sp8[8] =
-	    { 0x4b, 0x47, 0x53, 0x21, 0x40, 0x23, 0x24, 0x25 };
-	smbhash(p16, sp8, p14, 1);
-	smbhash(p16 + 8, sp8, p14 + 7, 1);
-}
-
-void
-E_P24(unsigned char *p21, const unsigned char *c8, unsigned char *p24)
-{
-	smbhash(p24, c8, p21, 1);
-	smbhash(p24 + 8, c8, p21 + 7, 1);
-	smbhash(p24 + 16, c8, p21 + 14, 1);
-}
-
-#if 0 /* currently unused */
-static void
-D_P16(unsigned char *p14, unsigned char *in, unsigned char *out)
-{
-	smbhash(out, in, p14, 0);
-	smbhash(out + 8, in + 8, p14 + 7, 0);
-}
-
-static void
-E_old_pw_hash(unsigned char *p14, unsigned char *in, unsigned char *out)
-{
-	smbhash(out, in, p14, 1);
-	smbhash(out + 8, in + 8, p14 + 7, 1);
-}
-/* these routines are currently unneeded, but may be
-	needed later */
-void
-cred_hash1(unsigned char *out, unsigned char *in, unsigned char *key)
-{
-	unsigned char buf[8];
-
-	smbhash(buf, in, key, 1);
-	smbhash(out, buf, key + 9, 1);
-}
-
-void
-cred_hash2(unsigned char *out, unsigned char *in, unsigned char *key)
-{
-	unsigned char buf[8];
-	static unsigned char key2[8];
-
-	smbhash(buf, in, key, 1);
-	key2[0] = key[7];
-	smbhash(out, buf, key2, 1);
-}
-
-void
-cred_hash3(unsigned char *out, unsigned char *in, unsigned char *key, int forw)
-{
-	static unsigned char key2[8];
-
-	smbhash(out, in, key, forw);
-	key2[0] = key[7];
-	smbhash(out + 8, in + 8, key2, forw);
-}
-#endif /* unneeded routines */
diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c
index b5041c84998..1525d5e662b 100644
--- a/fs/cifs/smbencrypt.c
+++ b/fs/cifs/smbencrypt.c
@@ -47,6 +47,88 @@
 #define SSVALX(buf,pos,val) (CVAL(buf,pos)=(val)&0xFF,CVAL(buf,pos+1)=(val)>>8)
 #define SSVAL(buf,pos,val) SSVALX((buf),(pos),((__u16)(val)))
 
+static void
+str_to_key(unsigned char *str, unsigned char *key)
+{
+	int i;
+
+	key[0] = str[0] >> 1;
+	key[1] = ((str[0] & 0x01) << 6) | (str[1] >> 2);
+	key[2] = ((str[1] & 0x03) << 5) | (str[2] >> 3);
+	key[3] = ((str[2] & 0x07) << 4) | (str[3] >> 4);
+	key[4] = ((str[3] & 0x0F) << 3) | (str[4] >> 5);
+	key[5] = ((str[4] & 0x1F) << 2) | (str[5] >> 6);
+	key[6] = ((str[5] & 0x3F) << 1) | (str[6] >> 7);
+	key[7] = str[6] & 0x7F;
+	for (i = 0; i < 8; i++)
+		key[i] = (key[i] << 1);
+}
+
+static int
+smbhash(unsigned char *out, const unsigned char *in, unsigned char *key)
+{
+	int rc;
+	unsigned char key2[8];
+	struct crypto_blkcipher *tfm_des;
+	struct scatterlist sgin, sgout;
+	struct blkcipher_desc desc;
+
+	str_to_key(key, key2);
+
+	tfm_des = crypto_alloc_blkcipher("ecb(des)", 0, CRYPTO_ALG_ASYNC);
+	if (IS_ERR(tfm_des)) {
+		rc = PTR_ERR(tfm_des);
+		cERROR(1, "could not allocate des crypto API\n");
+		goto smbhash_err;
+	}
+
+	desc.tfm = tfm_des;
+
+	crypto_blkcipher_setkey(tfm_des, key2, 8);
+
+	sg_init_one(&sgin, in, 8);
+	sg_init_one(&sgout, out, 8);
+
+	rc = crypto_blkcipher_encrypt(&desc, &sgout, &sgin, 8);
+	if (rc) {
+		cERROR(1, "could not encrypt crypt key rc: %d\n", rc);
+		crypto_free_blkcipher(tfm_des);
+		goto smbhash_err;
+	}
+
+smbhash_err:
+	return rc;
+}
+
+static int
+E_P16(unsigned char *p14, unsigned char *p16)
+{
+	int rc;
+	unsigned char sp8[8] =
+	    { 0x4b, 0x47, 0x53, 0x21, 0x40, 0x23, 0x24, 0x25 };
+
+	rc = smbhash(p16, sp8, p14);
+	if (rc)
+		return rc;
+	rc = smbhash(p16 + 8, sp8, p14 + 7);
+	return rc;
+}
+
+static int
+E_P24(unsigned char *p21, const unsigned char *c8, unsigned char *p24)
+{
+	int rc;
+
+	rc = smbhash(p24, c8, p21);
+	if (rc)
+		return rc;
+	rc = smbhash(p24 + 8, c8, p21 + 7);
+	if (rc)
+		return rc;
+	rc = smbhash(p24 + 16, c8, p21 + 14);
+	return rc;
+}
+
 /* produce a md4 message digest from data of length n bytes */
 int
 mdfour(unsigned char *md4_hash, unsigned char *link_str, int link_len)
@@ -87,40 +169,30 @@ mdfour_err:
 	return rc;
 }
 
-/* Does the des encryption from the NT or LM MD4 hash. */
-static void
-SMBOWFencrypt(unsigned char passwd[16], const unsigned char *c8,
-	      unsigned char p24[24])
-{
-	unsigned char p21[21];
-
-	memset(p21, '\0', 21);
-
-	memcpy(p21, passwd, 16);
-	E_P24(p21, c8, p24);
-}
-
 /*
    This implements the X/Open SMB password encryption
    It takes a password, a 8 byte "crypt key" and puts 24 bytes of
    encrypted password into p24 */
 /* Note that password must be uppercased and null terminated */
-void
+int
 SMBencrypt(unsigned char *passwd, const unsigned char *c8, unsigned char *p24)
 {
-	unsigned char p14[15], p21[21];
+	int rc;
+	unsigned char p14[14], p16[16], p21[21];
 
-	memset(p21, '\0', 21);
 	memset(p14, '\0', 14);
-	strncpy((char *) p14, (char *) passwd, 14);
+	memset(p16, '\0', 16);
+	memset(p21, '\0', 21);
 
-/*	strupper((char *)p14); *//* BB at least uppercase the easy range */
-	E_P16(p14, p21);
+	memcpy(p14, passwd, 14);
+	rc = E_P16(p14, p16);
+	if (rc)
+		return rc;
 
-	SMBOWFencrypt(p21, c8, p24);
+	memcpy(p21, p16, 16);
+	rc = E_P24(p21, c8, p24);
 
-	memset(p14, 0, 15);
-	memset(p21, 0, 21);
+	return rc;
 }
 
 /* Routines for Windows NT MD4 Hash functions. */
@@ -279,16 +351,18 @@ int
 SMBNTencrypt(unsigned char *passwd, unsigned char *c8, unsigned char *p24)
 {
 	int rc;
-	unsigned char p21[21];
+	unsigned char p16[16], p21[21];
 
+	memset(p16, '\0', 16);
 	memset(p21, '\0', 21);
 
-	rc = E_md4hash(passwd, p21);
+	rc = E_md4hash(passwd, p16);
 	if (rc) {
 		cFYI(1, "%s Can't generate NT hash, error: %d", __func__, rc);
 		return rc;
 	}
-	SMBOWFencrypt(p21, c8, p24);
+	memcpy(p21, p16, 16);
+	rc = E_P24(p21, c8, p24);
 	return rc;
 }
 
diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c
index 46d8756f2b2..f2513fb8c39 100644
--- a/fs/cifs/transport.c
+++ b/fs/cifs/transport.c
@@ -129,7 +129,7 @@ smb_sendv(struct TCP_Server_Info *server, struct kvec *iov, int n_vec)
 	unsigned int len = iov[0].iov_len;
 	unsigned int total_len;
 	int first_vec = 0;
-	unsigned int smb_buf_length = smb_buffer->smb_buf_length;
+	unsigned int smb_buf_length = be32_to_cpu(smb_buffer->smb_buf_length);
 	struct socket *ssocket = server->ssocket;
 
 	if (ssocket == NULL)
@@ -144,17 +144,10 @@ smb_sendv(struct TCP_Server_Info *server, struct kvec *iov, int n_vec)
 	else
 		smb_msg.msg_flags = MSG_NOSIGNAL;
 
-	/* smb header is converted in header_assemble. bcc and rest of SMB word
-	   area, and byte area if necessary, is converted to littleendian in
-	   cifssmb.c and RFC1001 len is converted to bigendian in smb_send
-	   Flags2 is converted in SendReceive */
-
-
 	total_len = 0;
 	for (i = 0; i < n_vec; i++)
 		total_len += iov[i].iov_len;
 
-	smb_buffer->smb_buf_length = cpu_to_be32(smb_buffer->smb_buf_length);
 	cFYI(1, "Sending smb:  total_len %d", total_len);
 	dump_smb(smb_buffer, len);
 
@@ -243,7 +236,7 @@ smb_sendv(struct TCP_Server_Info *server, struct kvec *iov, int n_vec)
 
 	/* Don't want to modify the buffer as a
 	   side effect of this call. */
-	smb_buffer->smb_buf_length = smb_buf_length;
+	smb_buffer->smb_buf_length = cpu_to_be32(smb_buf_length);
 
 	return rc;
 }
@@ -387,7 +380,7 @@ cifs_call_async(struct TCP_Server_Info *server, struct smb_hdr *in_buf,
 #ifdef CONFIG_CIFS_STATS2
 	atomic_inc(&server->inSend);
 #endif
-	rc = smb_send(server, in_buf, in_buf->smb_buf_length);
+	rc = smb_send(server, in_buf, be32_to_cpu(in_buf->smb_buf_length));
 #ifdef CONFIG_CIFS_STATS2
 	atomic_dec(&server->inSend);
 	mid->when_sent = jiffies;
@@ -422,7 +415,7 @@ SendReceiveNoRsp(const unsigned int xid, struct cifsSesInfo *ses,
 	int resp_buf_type;
 
 	iov[0].iov_base = (char *)in_buf;
-	iov[0].iov_len = in_buf->smb_buf_length + 4;
+	iov[0].iov_len = be32_to_cpu(in_buf->smb_buf_length) + 4;
 	flags |= CIFS_NO_RESP;
 	rc = SendReceive2(xid, ses, iov, 1, &resp_buf_type, flags);
 	cFYI(DBG2, "SendRcvNoRsp flags %d rc %d", flags, rc);
@@ -488,10 +481,10 @@ send_nt_cancel(struct TCP_Server_Info *server, struct smb_hdr *in_buf,
 	int rc = 0;
 
 	/* -4 for RFC1001 length and +2 for BCC field */
-	in_buf->smb_buf_length = sizeof(struct smb_hdr) - 4  + 2;
+	in_buf->smb_buf_length = cpu_to_be32(sizeof(struct smb_hdr) - 4  + 2);
 	in_buf->Command = SMB_COM_NT_CANCEL;
 	in_buf->WordCount = 0;
-	put_bcc_le(0, in_buf);
+	put_bcc(0, in_buf);
 
 	mutex_lock(&server->srv_mutex);
 	rc = cifs_sign_smb(in_buf, server, &mid->sequence_number);
@@ -499,7 +492,7 @@ send_nt_cancel(struct TCP_Server_Info *server, struct smb_hdr *in_buf,
 		mutex_unlock(&server->srv_mutex);
 		return rc;
 	}
-	rc = smb_send(server, in_buf, in_buf->smb_buf_length);
+	rc = smb_send(server, in_buf, be32_to_cpu(in_buf->smb_buf_length));
 	mutex_unlock(&server->srv_mutex);
 
 	cFYI(1, "issued NT_CANCEL for mid %u, rc = %d",
@@ -612,7 +605,7 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses,
 		return rc;
 	}
 
-	receive_len = midQ->resp_buf->smb_buf_length;
+	receive_len = be32_to_cpu(midQ->resp_buf->smb_buf_length);
 
 	if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) {
 		cERROR(1, "Frame too large received.  Length: %d  Xid: %d",
@@ -651,11 +644,6 @@ SendReceive2(const unsigned int xid, struct cifsSesInfo *ses,
 		rc = map_smb_to_linux_error(midQ->resp_buf,
 					    flags & CIFS_LOG_ERROR);
 
-		/* convert ByteCount if necessary */
-		if (receive_len >= sizeof(struct smb_hdr) - 4
-		    /* do not count RFC1001 header */  +
-		    (2 * midQ->resp_buf->WordCount) + 2 /* bcc */ )
-			put_bcc(get_bcc_le(midQ->resp_buf), midQ->resp_buf);
 		if ((flags & CIFS_NO_RESP) == 0)
 			midQ->resp_buf = NULL;  /* mark it so buf will
 						   not be freed by
@@ -698,9 +686,10 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses,
 	   to the same server. We may make this configurable later or
 	   use ses->maxReq */
 
-	if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) {
+	if (be32_to_cpu(in_buf->smb_buf_length) > CIFSMaxBufSize +
+			MAX_CIFS_HDR_SIZE - 4) {
 		cERROR(1, "Illegal length, greater than maximum frame, %d",
-			   in_buf->smb_buf_length);
+			   be32_to_cpu(in_buf->smb_buf_length));
 		return -EIO;
 	}
 
@@ -733,7 +722,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses,
 #ifdef CONFIG_CIFS_STATS2
 	atomic_inc(&ses->server->inSend);
 #endif
-	rc = smb_send(ses->server, in_buf, in_buf->smb_buf_length);
+	rc = smb_send(ses->server, in_buf, be32_to_cpu(in_buf->smb_buf_length));
 #ifdef CONFIG_CIFS_STATS2
 	atomic_dec(&ses->server->inSend);
 	midQ->when_sent = jiffies;
@@ -768,7 +757,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses,
 		return rc;
 	}
 
-	receive_len = midQ->resp_buf->smb_buf_length;
+	receive_len = be32_to_cpu(midQ->resp_buf->smb_buf_length);
 
 	if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) {
 		cERROR(1, "Frame too large received.  Length: %d  Xid: %d",
@@ -781,7 +770,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses,
 
 	if (midQ->resp_buf && out_buf
 	    && (midQ->midState == MID_RESPONSE_RECEIVED)) {
-		out_buf->smb_buf_length = receive_len;
+		out_buf->smb_buf_length = cpu_to_be32(receive_len);
 		memcpy((char *)out_buf + 4,
 		       (char *)midQ->resp_buf + 4,
 		       receive_len);
@@ -800,16 +789,10 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses,
 			}
 		}
 
-		*pbytes_returned = out_buf->smb_buf_length;
+		*pbytes_returned = be32_to_cpu(out_buf->smb_buf_length);
 
 		/* BB special case reconnect tid and uid here? */
 		rc = map_smb_to_linux_error(out_buf, 0 /* no log */ );
-
-		/* convert ByteCount if necessary */
-		if (receive_len >= sizeof(struct smb_hdr) - 4
-		    /* do not count RFC1001 header */  +
-		    (2 * out_buf->WordCount) + 2 /* bcc */ )
-			put_bcc(get_bcc_le(midQ->resp_buf), midQ->resp_buf);
 	} else {
 		rc = -EIO;
 		cERROR(1, "Bad MID state?");
@@ -877,9 +860,10 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon,
 	   to the same server. We may make this configurable later or
 	   use ses->maxReq */
 
-	if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) {
+	if (be32_to_cpu(in_buf->smb_buf_length) > CIFSMaxBufSize +
+			MAX_CIFS_HDR_SIZE - 4) {
 		cERROR(1, "Illegal length, greater than maximum frame, %d",
-			   in_buf->smb_buf_length);
+			   be32_to_cpu(in_buf->smb_buf_length));
 		return -EIO;
 	}
 
@@ -910,7 +894,7 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon,
 #ifdef CONFIG_CIFS_STATS2
 	atomic_inc(&ses->server->inSend);
 #endif
-	rc = smb_send(ses->server, in_buf, in_buf->smb_buf_length);
+	rc = smb_send(ses->server, in_buf, be32_to_cpu(in_buf->smb_buf_length));
 #ifdef CONFIG_CIFS_STATS2
 	atomic_dec(&ses->server->inSend);
 	midQ->when_sent = jiffies;
@@ -977,7 +961,7 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon,
 	if (rc != 0)
 		return rc;
 
-	receive_len = midQ->resp_buf->smb_buf_length;
+	receive_len = be32_to_cpu(midQ->resp_buf->smb_buf_length);
 	if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) {
 		cERROR(1, "Frame too large received.  Length: %d  Xid: %d",
 			receive_len, xid);
@@ -993,7 +977,7 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon,
 		goto out;
 	}
 
-	out_buf->smb_buf_length = receive_len;
+	out_buf->smb_buf_length = cpu_to_be32(receive_len);
 	memcpy((char *)out_buf + 4,
 	       (char *)midQ->resp_buf + 4,
 	       receive_len);
@@ -1012,17 +996,11 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifsTconInfo *tcon,
 		}
 	}
 
-	*pbytes_returned = out_buf->smb_buf_length;
+	*pbytes_returned = be32_to_cpu(out_buf->smb_buf_length);
 
 	/* BB special case reconnect tid and uid here? */
 	rc = map_smb_to_linux_error(out_buf, 0 /* no log */ );
 
-	/* convert ByteCount if necessary */
-	if (receive_len >= sizeof(struct smb_hdr) - 4
-	    /* do not count RFC1001 header */  +
-	    (2 * out_buf->WordCount) + 2 /* bcc */ )
-		put_bcc(get_bcc_le(out_buf), out_buf);
-
 out:
 	delete_mid(midQ);
 	if (rstart && rc == -EACCES)
diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c
index eae2a149160..912995e013e 100644
--- a/fs/cifs/xattr.c
+++ b/fs/cifs/xattr.c
@@ -112,6 +112,7 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name,
 	struct cifsTconInfo *pTcon;
 	struct super_block *sb;
 	char *full_path;
+	struct cifs_ntsd *pacl;
 
 	if (direntry == NULL)
 		return -EIO;
@@ -166,6 +167,25 @@ int cifs_setxattr(struct dentry *direntry, const char *ea_name,
 		rc = CIFSSMBSetEA(xid, pTcon, full_path, ea_name, ea_value,
 			(__u16)value_size, cifs_sb->local_nls,
 			cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR);
+	} else if (strncmp(ea_name, CIFS_XATTR_CIFS_ACL,
+			strlen(CIFS_XATTR_CIFS_ACL)) == 0) {
+		pacl = kmalloc(value_size, GFP_KERNEL);
+		if (!pacl) {
+			cFYI(1, "%s: Can't allocate memory for ACL",
+					__func__);
+			rc = -ENOMEM;
+		} else {
+#ifdef CONFIG_CIFS_ACL
+			memcpy(pacl, ea_value, value_size);
+			rc = set_cifs_acl(pacl, value_size,
+				direntry->d_inode, full_path);
+			if (rc == 0) /* force revalidate of the inode */
+				CIFS_I(direntry->d_inode)->time = 0;
+			kfree(pacl);
+#else
+			cFYI(1, "Set CIFS ACL not supported yet");
+#endif /* CONFIG_CIFS_ACL */
+		}
 	} else {
 		int temp;
 		temp = strncmp(ea_name, POSIX_ACL_XATTR_ACCESS,
diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
index 3313dd19f54..9a37a9b6de3 100644
--- a/fs/configfs/dir.c
+++ b/fs/configfs/dir.c
@@ -53,11 +53,14 @@ DEFINE_SPINLOCK(configfs_dirent_lock);
 static void configfs_d_iput(struct dentry * dentry,
 			    struct inode * inode)
 {
-	struct configfs_dirent * sd = dentry->d_fsdata;
+	struct configfs_dirent *sd = dentry->d_fsdata;
 
 	if (sd) {
 		BUG_ON(sd->s_dentry != dentry);
+		/* Coordinate with configfs_readdir */
+		spin_lock(&configfs_dirent_lock);
 		sd->s_dentry = NULL;
+		spin_unlock(&configfs_dirent_lock);
 		configfs_put(sd);
 	}
 	iput(inode);
@@ -689,7 +692,8 @@ static int create_default_group(struct config_group *parent_group,
 			sd = child->d_fsdata;
 			sd->s_type |= CONFIGFS_USET_DEFAULT;
 		} else {
-			d_delete(child);
+			BUG_ON(child->d_inode);
+			d_drop(child);
 			dput(child);
 		}
 	}
@@ -1545,7 +1549,7 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir
 	struct configfs_dirent * parent_sd = dentry->d_fsdata;
 	struct configfs_dirent *cursor = filp->private_data;
 	struct list_head *p, *q = &cursor->s_sibling;
-	ino_t ino;
+	ino_t ino = 0;
 	int i = filp->f_pos;
 
 	switch (i) {
@@ -1573,6 +1577,7 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir
 				struct configfs_dirent *next;
 				const char * name;
 				int len;
+				struct inode *inode = NULL;
 
 				next = list_entry(p, struct configfs_dirent,
 						   s_sibling);
@@ -1581,9 +1586,28 @@ static int configfs_readdir(struct file * filp, void * dirent, filldir_t filldir
 
 				name = configfs_get_name(next);
 				len = strlen(name);
-				if (next->s_dentry)
-					ino = next->s_dentry->d_inode->i_ino;
-				else
+
+				/*
+				 * We'll have a dentry and an inode for
+				 * PINNED items and for open attribute
+				 * files.  We lock here to prevent a race
+				 * with configfs_d_iput() clearing
+				 * s_dentry before calling iput().
+				 *
+				 * Why do we go to the trouble?  If
+				 * someone has an attribute file open,
+				 * the inode number should match until
+				 * they close it.  Beyond that, we don't
+				 * care.
+				 */
+				spin_lock(&configfs_dirent_lock);
+				dentry = next->s_dentry;
+				if (dentry)
+					inode = dentry->d_inode;
+				if (inode)
+					ino = inode->i_ino;
+				spin_unlock(&configfs_dirent_lock);
+				if (!inode)
 					ino = iunique(configfs_sb, 2);
 
 				if (filldir(dirent, name, len, filp->f_pos, ino,
@@ -1683,7 +1707,8 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys)
 		err = configfs_attach_group(sd->s_element, &group->cg_item,
 					    dentry);
 		if (err) {
-			d_delete(dentry);
+			BUG_ON(dentry->d_inode);
+			d_drop(dentry);
 			dput(dentry);
 		} else {
 			spin_lock(&configfs_dirent_lock);
diff --git a/fs/dcache.c b/fs/dcache.c
index 129a3573099..18b2a1f10ed 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -35,6 +35,7 @@
 #include <linux/hardirq.h>
 #include <linux/bit_spinlock.h>
 #include <linux/rculist_bl.h>
+#include <linux/prefetch.h>
 #include "internal.h"
 
 /*
@@ -99,12 +100,9 @@ static struct kmem_cache *dentry_cache __read_mostly;
 static unsigned int d_hash_mask __read_mostly;
 static unsigned int d_hash_shift __read_mostly;
 
-struct dcache_hash_bucket {
-	struct hlist_bl_head head;
-};
-static struct dcache_hash_bucket *dentry_hashtable __read_mostly;
+static struct hlist_bl_head *dentry_hashtable __read_mostly;
 
-static inline struct dcache_hash_bucket *d_hash(struct dentry *parent,
+static inline struct hlist_bl_head *d_hash(struct dentry *parent,
 					unsigned long hash)
 {
 	hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES;
@@ -112,16 +110,6 @@ static inline struct dcache_hash_bucket *d_hash(struct dentry *parent,
 	return dentry_hashtable + (hash & D_HASHMASK);
 }
 
-static inline void spin_lock_bucket(struct dcache_hash_bucket *b)
-{
-	bit_spin_lock(0, (unsigned long *)&b->head.first);
-}
-
-static inline void spin_unlock_bucket(struct dcache_hash_bucket *b)
-{
-	__bit_spin_unlock(0, (unsigned long *)&b->head.first);
-}
-
 /* Statistics gathering. */
 struct dentry_stat_t dentry_stat = {
 	.age_limit = 45,
@@ -167,8 +155,8 @@ static void d_free(struct dentry *dentry)
 	if (dentry->d_op && dentry->d_op->d_release)
 		dentry->d_op->d_release(dentry);
 
-	/* if dentry was never inserted into hash, immediate free is OK */
-	if (hlist_bl_unhashed(&dentry->d_hash))
+	/* if dentry was never visible to RCU, immediate free is OK */
+	if (!(dentry->d_flags & DCACHE_RCUACCESS))
 		__d_free(&dentry->d_u.d_rcu);
 	else
 		call_rcu(&dentry->d_u.d_rcu, __d_free);
@@ -330,28 +318,19 @@ static struct dentry *d_kill(struct dentry *dentry, struct dentry *parent)
  */
 void __d_drop(struct dentry *dentry)
 {
-	if (!(dentry->d_flags & DCACHE_UNHASHED)) {
-		if (unlikely(dentry->d_flags & DCACHE_DISCONNECTED)) {
-			bit_spin_lock(0,
-				(unsigned long *)&dentry->d_sb->s_anon.first);
-			dentry->d_flags |= DCACHE_UNHASHED;
-			hlist_bl_del_init(&dentry->d_hash);
-			__bit_spin_unlock(0,
-				(unsigned long *)&dentry->d_sb->s_anon.first);
-		} else {
-			struct dcache_hash_bucket *b;
+	if (!d_unhashed(dentry)) {
+		struct hlist_bl_head *b;
+		if (unlikely(dentry->d_flags & DCACHE_DISCONNECTED))
+			b = &dentry->d_sb->s_anon;
+		else
 			b = d_hash(dentry->d_parent, dentry->d_name.hash);
-			spin_lock_bucket(b);
-			/*
-			 * We may not actually need to put DCACHE_UNHASHED
-			 * manipulations under the hash lock, but follow
-			 * the principle of least surprise.
-			 */
-			dentry->d_flags |= DCACHE_UNHASHED;
-			hlist_bl_del_rcu(&dentry->d_hash);
-			spin_unlock_bucket(b);
-			dentry_rcuwalk_barrier(dentry);
-		}
+
+		hlist_bl_lock(b);
+		__hlist_bl_del(&dentry->d_hash);
+		dentry->d_hash.pprev = NULL;
+		hlist_bl_unlock(b);
+
+		dentry_rcuwalk_barrier(dentry);
 	}
 }
 EXPORT_SYMBOL(__d_drop);
@@ -1304,7 +1283,7 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name)
 	dname[name->len] = 0;
 
 	dentry->d_count = 1;
-	dentry->d_flags = DCACHE_UNHASHED;
+	dentry->d_flags = 0;
 	spin_lock_init(&dentry->d_lock);
 	seqcount_init(&dentry->d_seq);
 	dentry->d_inode = NULL;
@@ -1606,10 +1585,9 @@ struct dentry *d_obtain_alias(struct inode *inode)
 	tmp->d_inode = inode;
 	tmp->d_flags |= DCACHE_DISCONNECTED;
 	list_add(&tmp->d_alias, &inode->i_dentry);
-	bit_spin_lock(0, (unsigned long *)&tmp->d_sb->s_anon.first);
-	tmp->d_flags &= ~DCACHE_UNHASHED;
+	hlist_bl_lock(&tmp->d_sb->s_anon);
 	hlist_bl_add_head(&tmp->d_hash, &tmp->d_sb->s_anon);
-	__bit_spin_unlock(0, (unsigned long *)&tmp->d_sb->s_anon.first);
+	hlist_bl_unlock(&tmp->d_sb->s_anon);
 	spin_unlock(&tmp->d_lock);
 	spin_unlock(&inode->i_lock);
 	security_d_instantiate(tmp, inode);
@@ -1789,7 +1767,7 @@ struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name,
 	unsigned int len = name->len;
 	unsigned int hash = name->hash;
 	const unsigned char *str = name->name;
-	struct dcache_hash_bucket *b = d_hash(parent, hash);
+	struct hlist_bl_head *b = d_hash(parent, hash);
 	struct hlist_bl_node *node;
 	struct dentry *dentry;
 
@@ -1813,7 +1791,7 @@ struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name,
 	 *
 	 * See Documentation/filesystems/path-lookup.txt for more details.
 	 */
-	hlist_bl_for_each_entry_rcu(dentry, node, &b->head, d_hash) {
+	hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) {
 		struct inode *i;
 		const char *tname;
 		int tlen;
@@ -1908,7 +1886,7 @@ struct dentry *__d_lookup(struct dentry *parent, struct qstr *name)
 	unsigned int len = name->len;
 	unsigned int hash = name->hash;
 	const unsigned char *str = name->name;
-	struct dcache_hash_bucket *b = d_hash(parent, hash);
+	struct hlist_bl_head *b = d_hash(parent, hash);
 	struct hlist_bl_node *node;
 	struct dentry *found = NULL;
 	struct dentry *dentry;
@@ -1935,7 +1913,7 @@ struct dentry *__d_lookup(struct dentry *parent, struct qstr *name)
 	 */
 	rcu_read_lock();
 	
-	hlist_bl_for_each_entry_rcu(dentry, node, &b->head, d_hash) {
+	hlist_bl_for_each_entry_rcu(dentry, node, b, d_hash) {
 		const char *tname;
 		int tlen;
 
@@ -2086,13 +2064,13 @@ again:
 }
 EXPORT_SYMBOL(d_delete);
 
-static void __d_rehash(struct dentry * entry, struct dcache_hash_bucket *b)
+static void __d_rehash(struct dentry * entry, struct hlist_bl_head *b)
 {
 	BUG_ON(!d_unhashed(entry));
-	spin_lock_bucket(b);
- 	entry->d_flags &= ~DCACHE_UNHASHED;
-	hlist_bl_add_head_rcu(&entry->d_hash, &b->head);
-	spin_unlock_bucket(b);
+	hlist_bl_lock(b);
+	entry->d_flags |= DCACHE_RCUACCESS;
+	hlist_bl_add_head_rcu(&entry->d_hash, b);
+	hlist_bl_unlock(b);
 }
 
 static void _d_rehash(struct dentry * entry)
@@ -3025,7 +3003,7 @@ static void __init dcache_init_early(void)
 
 	dentry_hashtable =
 		alloc_large_system_hash("Dentry cache",
-					sizeof(struct dcache_hash_bucket),
+					sizeof(struct hlist_bl_head),
 					dhash_entries,
 					13,
 					HASH_EARLY,
@@ -3034,7 +3012,7 @@ static void __init dcache_init_early(void)
 					0);
 
 	for (loop = 0; loop < (1 << d_hash_shift); loop++)
-		INIT_HLIST_BL_HEAD(&dentry_hashtable[loop].head);
+		INIT_HLIST_BL_HEAD(dentry_hashtable + loop);
 }
 
 static void __init dcache_init(void)
@@ -3057,7 +3035,7 @@ static void __init dcache_init(void)
 
 	dentry_hashtable =
 		alloc_large_system_hash("Dentry cache",
-					sizeof(struct dcache_hash_bucket),
+					sizeof(struct hlist_bl_head),
 					dhash_entries,
 					13,
 					0,
@@ -3066,7 +3044,7 @@ static void __init dcache_init(void)
 					0);
 
 	for (loop = 0; loop < (1 << d_hash_shift); loop++)
-		INIT_HLIST_BL_HEAD(&dentry_hashtable[loop].head);
+		INIT_HLIST_BL_HEAD(dentry_hashtable + loop);
 }
 
 /* SLAB cache for __getname() consumers */
diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
index 89d394d8fe2..90f76575c05 100644
--- a/fs/debugfs/file.c
+++ b/fs/debugfs/file.c
@@ -428,26 +428,17 @@ static ssize_t write_file_bool(struct file *file, const char __user *user_buf,
 			       size_t count, loff_t *ppos)
 {
 	char buf[32];
-	int buf_size;
+	size_t buf_size;
+	bool bv;
 	u32 *val = file->private_data;
 
 	buf_size = min(count, (sizeof(buf)-1));
 	if (copy_from_user(buf, user_buf, buf_size))
 		return -EFAULT;
 
-	switch (buf[0]) {
-	case 'y':
-	case 'Y':
-	case '1':
-		*val = 1;
-		break;
-	case 'n':
-	case 'N':
-	case '0':
-		*val = 0;
-		break;
-	}
-	
+	if (strtobool(buf, &bv) == 0)
+		*val = bv;
+
 	return count;
 }
 
diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index d2a70a4561f..b8d5c809102 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -1452,6 +1452,25 @@ static void set_default_header_data(struct ecryptfs_crypt_stat *crypt_stat)
 	crypt_stat->metadata_size = ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE;
 }
 
+void ecryptfs_i_size_init(const char *page_virt, struct inode *inode)
+{
+	struct ecryptfs_mount_crypt_stat *mount_crypt_stat;
+	struct ecryptfs_crypt_stat *crypt_stat;
+	u64 file_size;
+
+	crypt_stat = &ecryptfs_inode_to_private(inode)->crypt_stat;
+	mount_crypt_stat =
+		&ecryptfs_superblock_to_private(inode->i_sb)->mount_crypt_stat;
+	if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) {
+		file_size = i_size_read(ecryptfs_inode_to_lower(inode));
+		if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR)
+			file_size += crypt_stat->metadata_size;
+	} else
+		file_size = get_unaligned_be64(page_virt);
+	i_size_write(inode, (loff_t)file_size);
+	crypt_stat->flags |= ECRYPTFS_I_SIZE_INITIALIZED;
+}
+
 /**
  * ecryptfs_read_headers_virt
  * @page_virt: The virtual address into which to read the headers
@@ -1482,6 +1501,8 @@ static int ecryptfs_read_headers_virt(char *page_virt,
 		rc = -EINVAL;
 		goto out;
 	}
+	if (!(crypt_stat->flags & ECRYPTFS_I_SIZE_INITIALIZED))
+		ecryptfs_i_size_init(page_virt, ecryptfs_dentry->d_inode);
 	offset += MAGIC_ECRYPTFS_MARKER_SIZE_BYTES;
 	rc = ecryptfs_process_flags(crypt_stat, (page_virt + offset),
 				    &bytes_read);
diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h
index bd3cafd0949..e70282775e2 100644
--- a/fs/ecryptfs/ecryptfs_kernel.h
+++ b/fs/ecryptfs/ecryptfs_kernel.h
@@ -269,6 +269,7 @@ struct ecryptfs_crypt_stat {
 #define ECRYPTFS_ENCFN_USE_MOUNT_FNEK 0x00000800
 #define ECRYPTFS_ENCFN_USE_FEK        0x00001000
 #define ECRYPTFS_UNLINK_SIGS          0x00002000
+#define ECRYPTFS_I_SIZE_INITIALIZED   0x00004000
 	u32 flags;
 	unsigned int file_version;
 	size_t iv_bytes;
@@ -295,6 +296,8 @@ struct ecryptfs_crypt_stat {
 struct ecryptfs_inode_info {
 	struct inode vfs_inode;
 	struct inode *wii_inode;
+	struct mutex lower_file_mutex;
+	atomic_t lower_file_count;
 	struct file *lower_file;
 	struct ecryptfs_crypt_stat crypt_stat;
 };
@@ -626,6 +629,7 @@ struct ecryptfs_open_req {
 int ecryptfs_interpose(struct dentry *hidden_dentry,
 		       struct dentry *this_dentry, struct super_block *sb,
 		       u32 flags);
+void ecryptfs_i_size_init(const char *page_virt, struct inode *inode);
 int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 					struct dentry *lower_dentry,
 					struct inode *ecryptfs_dir_inode);
@@ -757,7 +761,8 @@ int ecryptfs_privileged_open(struct file **lower_file,
 			     struct dentry *lower_dentry,
 			     struct vfsmount *lower_mnt,
 			     const struct cred *cred);
-int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry);
+int ecryptfs_get_lower_file(struct dentry *ecryptfs_dentry);
+void ecryptfs_put_lower_file(struct inode *inode);
 int
 ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes,
 			     size_t *packet_size,
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index cedc913d11b..566e5472f78 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -191,10 +191,10 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 				      | ECRYPTFS_ENCRYPTED);
 	}
 	mutex_unlock(&crypt_stat->cs_mutex);
-	rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
+	rc = ecryptfs_get_lower_file(ecryptfs_dentry);
 	if (rc) {
 		printk(KERN_ERR "%s: Error attempting to initialize "
-			"the persistent file for the dentry with name "
+			"the lower file for the dentry with name "
 			"[%s]; rc = [%d]\n", __func__,
 			ecryptfs_dentry->d_name.name, rc);
 		goto out_free;
@@ -202,9 +202,9 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 	if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_ACCMODE)
 	    == O_RDONLY && (file->f_flags & O_ACCMODE) != O_RDONLY) {
 		rc = -EPERM;
-		printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs "
+		printk(KERN_WARNING "%s: Lower file is RO; eCryptfs "
 		       "file must hence be opened RO\n", __func__);
-		goto out_free;
+		goto out_put;
 	}
 	ecryptfs_set_file_lower(
 		file, ecryptfs_inode_to_private(inode)->lower_file);
@@ -232,10 +232,11 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 				       "Plaintext passthrough mode is not "
 				       "enabled; returning -EIO\n");
 				mutex_unlock(&crypt_stat->cs_mutex);
-				goto out_free;
+				goto out_put;
 			}
 			rc = 0;
-			crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED);
+			crypt_stat->flags &= ~(ECRYPTFS_I_SIZE_INITIALIZED
+					       | ECRYPTFS_ENCRYPTED);
 			mutex_unlock(&crypt_stat->cs_mutex);
 			goto out;
 		}
@@ -245,6 +246,8 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 			"[0x%.16lx] size: [0x%.16llx]\n", inode, inode->i_ino,
 			(unsigned long long)i_size_read(inode));
 	goto out;
+out_put:
+	ecryptfs_put_lower_file(inode);
 out_free:
 	kmem_cache_free(ecryptfs_file_info_cache,
 			ecryptfs_file_to_private(file));
@@ -254,17 +257,13 @@ out:
 
 static int ecryptfs_flush(struct file *file, fl_owner_t td)
 {
-	int rc = 0;
-	struct file *lower_file = NULL;
-
-	lower_file = ecryptfs_file_to_lower(file);
-	if (lower_file->f_op && lower_file->f_op->flush)
-		rc = lower_file->f_op->flush(lower_file, td);
-	return rc;
+	return file->f_mode & FMODE_WRITE
+	       ? filemap_write_and_wait(file->f_mapping) : 0;
 }
 
 static int ecryptfs_release(struct inode *inode, struct file *file)
 {
+	ecryptfs_put_lower_file(inode);
 	kmem_cache_free(ecryptfs_file_info_cache,
 			ecryptfs_file_to_private(file));
 	return 0;
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index f99051b7ada..4d4cc6a90cd 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -168,19 +168,18 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry)
 				"context; rc = [%d]\n", rc);
 		goto out;
 	}
-	rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
+	rc = ecryptfs_get_lower_file(ecryptfs_dentry);
 	if (rc) {
 		printk(KERN_ERR "%s: Error attempting to initialize "
-			"the persistent file for the dentry with name "
+			"the lower file for the dentry with name "
 			"[%s]; rc = [%d]\n", __func__,
 			ecryptfs_dentry->d_name.name, rc);
 		goto out;
 	}
 	rc = ecryptfs_write_metadata(ecryptfs_dentry);
-	if (rc) {
+	if (rc)
 		printk(KERN_ERR "Error writing headers; rc = [%d]\n", rc);
-		goto out;
-	}
+	ecryptfs_put_lower_file(ecryptfs_dentry->d_inode);
 out:
 	return rc;
 }
@@ -226,11 +225,9 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 	struct dentry *lower_dir_dentry;
 	struct vfsmount *lower_mnt;
 	struct inode *lower_inode;
-	struct ecryptfs_mount_crypt_stat *mount_crypt_stat;
 	struct ecryptfs_crypt_stat *crypt_stat;
 	char *page_virt = NULL;
-	u64 file_size;
-	int rc = 0;
+	int put_lower = 0, rc = 0;
 
 	lower_dir_dentry = lower_dentry->d_parent;
 	lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(
@@ -277,14 +274,15 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 		rc = -ENOMEM;
 		goto out;
 	}
-	rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
+	rc = ecryptfs_get_lower_file(ecryptfs_dentry);
 	if (rc) {
 		printk(KERN_ERR "%s: Error attempting to initialize "
-			"the persistent file for the dentry with name "
+			"the lower file for the dentry with name "
 			"[%s]; rc = [%d]\n", __func__,
 			ecryptfs_dentry->d_name.name, rc);
 		goto out_free_kmem;
 	}
+	put_lower = 1;
 	crypt_stat = &ecryptfs_inode_to_private(
 					ecryptfs_dentry->d_inode)->crypt_stat;
 	/* TODO: lock for crypt_stat comparison */
@@ -302,18 +300,7 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 		}
 		crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR;
 	}
-	mount_crypt_stat = &ecryptfs_superblock_to_private(
-		ecryptfs_dentry->d_sb)->mount_crypt_stat;
-	if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) {
-		if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR)
-			file_size = (crypt_stat->metadata_size
-				     + i_size_read(lower_dentry->d_inode));
-		else
-			file_size = i_size_read(lower_dentry->d_inode);
-	} else {
-		file_size = get_unaligned_be64(page_virt);
-	}
-	i_size_write(ecryptfs_dentry->d_inode, (loff_t)file_size);
+	ecryptfs_i_size_init(page_virt, ecryptfs_dentry->d_inode);
 out_free_kmem:
 	kmem_cache_free(ecryptfs_header_cache_2, page_virt);
 	goto out;
@@ -322,6 +309,8 @@ out_put:
 	mntput(lower_mnt);
 	d_drop(ecryptfs_dentry);
 out:
+	if (put_lower)
+		ecryptfs_put_lower_file(ecryptfs_dentry->d_inode);
 	return rc;
 }
 
@@ -538,8 +527,6 @@ static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry)
 	dget(lower_dentry);
 	rc = vfs_rmdir(lower_dir_dentry->d_inode, lower_dentry);
 	dput(lower_dentry);
-	if (!rc)
-		d_delete(lower_dentry);
 	fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode);
 	dir->i_nlink = lower_dir_dentry->d_inode->i_nlink;
 	unlock_dir(lower_dir_dentry);
@@ -610,8 +597,8 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		fsstack_copy_attr_all(old_dir, lower_old_dir_dentry->d_inode);
 out_lock:
 	unlock_rename(lower_old_dir_dentry, lower_new_dir_dentry);
-	dput(lower_new_dentry->d_parent);
-	dput(lower_old_dentry->d_parent);
+	dput(lower_new_dir_dentry);
+	dput(lower_old_dir_dentry);
 	dput(lower_new_dentry);
 	dput(lower_old_dentry);
 	return rc;
@@ -759,8 +746,11 @@ static int truncate_upper(struct dentry *dentry, struct iattr *ia,
 
 	if (unlikely((ia->ia_size == i_size))) {
 		lower_ia->ia_valid &= ~ATTR_SIZE;
-		goto out;
+		return 0;
 	}
+	rc = ecryptfs_get_lower_file(dentry);
+	if (rc)
+		return rc;
 	crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat;
 	/* Switch on growing or shrinking file */
 	if (ia->ia_size > i_size) {
@@ -838,6 +828,7 @@ static int truncate_upper(struct dentry *dentry, struct iattr *ia,
 			lower_ia->ia_valid &= ~ATTR_SIZE;
 	}
 out:
+	ecryptfs_put_lower_file(inode);
 	return rc;
 }
 
@@ -913,7 +904,13 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia)
 
 		mount_crypt_stat = &ecryptfs_superblock_to_private(
 			dentry->d_sb)->mount_crypt_stat;
+		rc = ecryptfs_get_lower_file(dentry);
+		if (rc) {
+			mutex_unlock(&crypt_stat->cs_mutex);
+			goto out;
+		}
 		rc = ecryptfs_read_metadata(dentry);
+		ecryptfs_put_lower_file(inode);
 		if (rc) {
 			if (!(mount_crypt_stat->flags
 			      & ECRYPTFS_PLAINTEXT_PASSTHROUGH_ENABLED)) {
@@ -927,10 +924,17 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia)
 				goto out;
 			}
 			rc = 0;
-			crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED);
+			crypt_stat->flags &= ~(ECRYPTFS_I_SIZE_INITIALIZED
+					       | ECRYPTFS_ENCRYPTED);
 		}
 	}
 	mutex_unlock(&crypt_stat->cs_mutex);
+	if (S_ISREG(inode->i_mode)) {
+		rc = filemap_write_and_wait(inode->i_mapping);
+		if (rc)
+			goto out;
+		fsstack_copy_attr_all(inode, lower_inode);
+	}
 	memcpy(&lower_ia, ia, sizeof(lower_ia));
 	if (ia->ia_valid & ATTR_FILE)
 		lower_ia.ia_file = ecryptfs_file_to_lower(ia->ia_file);
diff --git a/fs/ecryptfs/kthread.c b/fs/ecryptfs/kthread.c
index 0851ab6980f..69f994a7d52 100644
--- a/fs/ecryptfs/kthread.c
+++ b/fs/ecryptfs/kthread.c
@@ -44,7 +44,7 @@ static struct task_struct *ecryptfs_kthread;
  * @ignored: ignored
  *
  * The eCryptfs kernel thread that has the responsibility of getting
- * the lower persistent file with RW permissions.
+ * the lower file with RW permissions.
  *
  * Returns zero on success; non-zero otherwise
  */
@@ -141,8 +141,8 @@ int ecryptfs_privileged_open(struct file **lower_file,
 	int rc = 0;
 
 	/* Corresponding dput() and mntput() are done when the
-	 * persistent file is fput() when the eCryptfs inode is
-	 * destroyed. */
+	 * lower file is fput() when all eCryptfs files for the inode are
+	 * released. */
 	dget(lower_dentry);
 	mntget(lower_mnt);
 	flags |= IS_RDONLY(lower_dentry->d_inode) ? O_RDONLY : O_RDWR;
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index fdb2eb0ad09..89b93389af8 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -96,7 +96,7 @@ void __ecryptfs_printk(const char *fmt, ...)
 }
 
 /**
- * ecryptfs_init_persistent_file
+ * ecryptfs_init_lower_file
  * @ecryptfs_dentry: Fully initialized eCryptfs dentry object, with
  *                   the lower dentry and the lower mount set
  *
@@ -104,42 +104,70 @@ void __ecryptfs_printk(const char *fmt, ...)
  * inode. All I/O operations to the lower inode occur through that
  * file. When the first eCryptfs dentry that interposes with the first
  * lower dentry for that inode is created, this function creates the
- * persistent file struct and associates it with the eCryptfs
- * inode. When the eCryptfs inode is destroyed, the file is closed.
+ * lower file struct and associates it with the eCryptfs
+ * inode. When all eCryptfs files associated with the inode are released, the
+ * file is closed.
  *
- * The persistent file will be opened with read/write permissions, if
+ * The lower file will be opened with read/write permissions, if
  * possible. Otherwise, it is opened read-only.
  *
- * This function does nothing if a lower persistent file is already
+ * This function does nothing if a lower file is already
  * associated with the eCryptfs inode.
  *
  * Returns zero on success; non-zero otherwise
  */
-int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
+static int ecryptfs_init_lower_file(struct dentry *dentry,
+				    struct file **lower_file)
 {
 	const struct cred *cred = current_cred();
-	struct ecryptfs_inode_info *inode_info =
-		ecryptfs_inode_to_private(ecryptfs_dentry->d_inode);
-	int rc = 0;
+	struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry);
+	struct vfsmount *lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry);
+	int rc;
 
-	if (!inode_info->lower_file) {
-		struct dentry *lower_dentry;
-		struct vfsmount *lower_mnt =
-			ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry);
+	rc = ecryptfs_privileged_open(lower_file, lower_dentry, lower_mnt,
+				      cred);
+	if (rc) {
+		printk(KERN_ERR "Error opening lower file "
+		       "for lower_dentry [0x%p] and lower_mnt [0x%p]; "
+		       "rc = [%d]\n", lower_dentry, lower_mnt, rc);
+		(*lower_file) = NULL;
+	}
+	return rc;
+}
 
-		lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry);
-		rc = ecryptfs_privileged_open(&inode_info->lower_file,
-					      lower_dentry, lower_mnt, cred);
-		if (rc) {
-			printk(KERN_ERR "Error opening lower persistent file "
-			       "for lower_dentry [0x%p] and lower_mnt [0x%p]; "
-			       "rc = [%d]\n", lower_dentry, lower_mnt, rc);
-			inode_info->lower_file = NULL;
-		}
+int ecryptfs_get_lower_file(struct dentry *dentry)
+{
+	struct ecryptfs_inode_info *inode_info =
+		ecryptfs_inode_to_private(dentry->d_inode);
+	int count, rc = 0;
+
+	mutex_lock(&inode_info->lower_file_mutex);
+	count = atomic_inc_return(&inode_info->lower_file_count);
+	if (WARN_ON_ONCE(count < 1))
+		rc = -EINVAL;
+	else if (count == 1) {
+		rc = ecryptfs_init_lower_file(dentry,
+					      &inode_info->lower_file);
+		if (rc)
+			atomic_set(&inode_info->lower_file_count, 0);
 	}
+	mutex_unlock(&inode_info->lower_file_mutex);
 	return rc;
 }
 
+void ecryptfs_put_lower_file(struct inode *inode)
+{
+	struct ecryptfs_inode_info *inode_info;
+
+	inode_info = ecryptfs_inode_to_private(inode);
+	if (atomic_dec_and_mutex_lock(&inode_info->lower_file_count,
+				      &inode_info->lower_file_mutex)) {
+		fput(inode_info->lower_file);
+		inode_info->lower_file = NULL;
+		mutex_unlock(&inode_info->lower_file_mutex);
+	}
+}
+
 static struct inode *ecryptfs_get_inode(struct inode *lower_inode,
 		       struct super_block *sb)
 {
diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c
index bacc882e1ae..245b517bf1b 100644
--- a/fs/ecryptfs/super.c
+++ b/fs/ecryptfs/super.c
@@ -55,6 +55,8 @@ static struct inode *ecryptfs_alloc_inode(struct super_block *sb)
 	if (unlikely(!inode_info))
 		goto out;
 	ecryptfs_init_crypt_stat(&inode_info->crypt_stat);
+	mutex_init(&inode_info->lower_file_mutex);
+	atomic_set(&inode_info->lower_file_count, 0);
 	inode_info->lower_file = NULL;
 	inode = &inode_info->vfs_inode;
 out:
@@ -77,8 +79,7 @@ static void ecryptfs_i_callback(struct rcu_head *head)
  *
  * This is used during the final destruction of the inode.  All
  * allocation of memory related to the inode, including allocated
- * memory in the crypt_stat struct, will be released here. This
- * function also fput()'s the persistent file for the lower inode.
+ * memory in the crypt_stat struct, will be released here.
  * There should be no chance that this deallocation will be missed.
  */
 static void ecryptfs_destroy_inode(struct inode *inode)
@@ -86,16 +87,7 @@ static void ecryptfs_destroy_inode(struct inode *inode)
 	struct ecryptfs_inode_info *inode_info;
 
 	inode_info = ecryptfs_inode_to_private(inode);
-	if (inode_info->lower_file) {
-		struct dentry *lower_dentry =
-			inode_info->lower_file->f_dentry;
-
-		BUG_ON(!lower_dentry);
-		if (lower_dentry->d_inode) {
-			fput(inode_info->lower_file);
-			inode_info->lower_file = NULL;
-		}
-	}
+	BUG_ON(inode_info->lower_file);
 	ecryptfs_destroy_crypt_stat(&inode_info->crypt_stat);
 	call_rcu(&inode->i_rcu, ecryptfs_i_callback);
 }
diff --git a/fs/exec.c b/fs/exec.c
index 5e62d26a4fe..8328beb9016 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1659,6 +1659,7 @@ static int zap_process(struct task_struct *start, int exit_code)
 
 	t = start;
 	do {
+		task_clear_group_stop_pending(t);
 		if (t != current && t->mm) {
 			sigaddset(&t->pending.signal, SIGKILL);
 			signal_wake_up(t, 1);
diff --git a/fs/file.c b/fs/file.c
index 0be344755c0..4c6992d8f3b 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -9,6 +9,7 @@
 #include <linux/module.h>
 #include <linux/fs.h>
 #include <linux/mm.h>
+#include <linux/mmzone.h>
 #include <linux/time.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
@@ -39,14 +40,17 @@ int sysctl_nr_open_max = 1024 * 1024; /* raised later */
  */
 static DEFINE_PER_CPU(struct fdtable_defer, fdtable_defer_list);
 
-static inline void *alloc_fdmem(unsigned int size)
+static void *alloc_fdmem(unsigned int size)
 {
-	void *data;
-
-	data = kmalloc(size, GFP_KERNEL|__GFP_NOWARN);
-	if (data != NULL)
-		return data;
-
+	/*
+	 * Very large allocations can stress page reclaim, so fall back to
+	 * vmalloc() if the allocation size will be considered "large" by the VM.
+	 */
+	if (size <= (PAGE_SIZE << PAGE_ALLOC_COSTLY_ORDER)) {
+		void *data = kmalloc(size, GFP_KERNEL|__GFP_NOWARN);
+		if (data != NULL)
+			return data;
+	}
 	return vmalloc(size);
 }
 
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index c6ba49bd95b..b32eb29a4e6 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -174,7 +174,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd)
 		if (!inode)
 			return 0;
 
-		if (nd->flags & LOOKUP_RCU)
+		if (nd && (nd->flags & LOOKUP_RCU))
 			return -ECHILD;
 
 		fc = get_fuse_conn(inode);
diff --git a/fs/gfs2/Makefile b/fs/gfs2/Makefile
index f3d23ef4e87..86128202384 100644
--- a/fs/gfs2/Makefile
+++ b/fs/gfs2/Makefile
@@ -1,9 +1,9 @@
 ccflags-y := -I$(src)
 obj-$(CONFIG_GFS2_FS) += gfs2.o
 gfs2-y := acl.o bmap.o dir.o xattr.o glock.o \
-	glops.o inode.o log.o lops.o main.o meta_io.o \
+	glops.o log.o lops.o main.o meta_io.o \
 	aops.o dentry.o export.o file.o \
-	ops_fstype.o ops_inode.o quota.o \
+	ops_fstype.o inode.o quota.o \
 	recovery.o rgrp.o super.o sys.o trans.o util.o
 
 gfs2-$(CONFIG_GFS2_FS_LOCKING_DLM) += lock_dlm.o
diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index c71995b111b..802ac5eeba2 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -884,8 +884,8 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping,
 	}
 
 	brelse(dibh);
-	gfs2_trans_end(sdp);
 failed:
+	gfs2_trans_end(sdp);
 	if (al) {
 		gfs2_inplace_release(ip);
 		gfs2_quota_unlock(ip);
@@ -1076,8 +1076,8 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask)
 		bd = bh->b_private;
 		if (bd && bd->bd_ail)
 			goto cannot_release;
-		gfs2_assert_warn(sdp, !buffer_pinned(bh));
-		gfs2_assert_warn(sdp, !buffer_dirty(bh));
+		if (buffer_pinned(bh) || buffer_dirty(bh))
+			goto not_possible;
 		bh = bh->b_this_page;
 	} while(bh != head);
 	gfs2_log_unlock(sdp);
@@ -1107,6 +1107,10 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask)
 	} while (bh != head);
 
 	return try_to_free_buffers(page);
+
+not_possible: /* Should never happen */
+	WARN_ON(buffer_dirty(bh));
+	WARN_ON(buffer_pinned(bh));
 cannot_release:
 	gfs2_log_unlock(sdp);
 	return 0;
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index 5c356d09c32..091ee477953 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -82,12 +82,9 @@
 struct qstr gfs2_qdot __read_mostly;
 struct qstr gfs2_qdotdot __read_mostly;
 
-typedef int (*leaf_call_t) (struct gfs2_inode *dip, u32 index, u32 len,
-			    u64 leaf_no, void *data);
 typedef int (*gfs2_dscan_t)(const struct gfs2_dirent *dent,
 			    const struct qstr *name, void *opaque);
 
-
 int gfs2_dir_get_new_buffer(struct gfs2_inode *ip, u64 block,
 			    struct buffer_head **bhp)
 {
@@ -1506,7 +1503,7 @@ struct inode *gfs2_dir_search(struct inode *dir, const struct qstr *name)
 		inode = gfs2_inode_lookup(dir->i_sb, 
 				be16_to_cpu(dent->de_type),
 				be64_to_cpu(dent->de_inum.no_addr),
-				be64_to_cpu(dent->de_inum.no_formal_ino));
+				be64_to_cpu(dent->de_inum.no_formal_ino), 0);
 		brelse(bh);
 		return inode;
 	}
@@ -1600,7 +1597,7 @@ static int dir_new_leaf(struct inode *inode, const struct qstr *name)
  */
 
 int gfs2_dir_add(struct inode *inode, const struct qstr *name,
-		 const struct gfs2_inode *nip, unsigned type)
+		 const struct gfs2_inode *nip)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct buffer_head *bh;
@@ -1616,7 +1613,7 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name,
 				return PTR_ERR(dent);
 			dent = gfs2_init_dirent(inode, dent, name, bh);
 			gfs2_inum_out(nip, dent);
-			dent->de_type = cpu_to_be16(type);
+			dent->de_type = cpu_to_be16(IF2DT(nip->i_inode.i_mode));
 			if (ip->i_diskflags & GFS2_DIF_EXHASH) {
 				leaf = (struct gfs2_leaf *)bh->b_data;
 				be16_add_cpu(&leaf->lf_entries, 1);
@@ -1628,6 +1625,8 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name,
 			gfs2_trans_add_bh(ip->i_gl, bh, 1);
 			ip->i_entries++;
 			ip->i_inode.i_mtime = ip->i_inode.i_ctime = CURRENT_TIME;
+			if (S_ISDIR(nip->i_inode.i_mode))
+				inc_nlink(&ip->i_inode);
 			gfs2_dinode_out(ip, bh->b_data);
 			brelse(bh);
 			error = 0;
@@ -1672,8 +1671,9 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name,
  * Returns: 0 on success, error code on failure
  */
 
-int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name)
+int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry)
 {
+	const struct qstr *name = &dentry->d_name;
 	struct gfs2_dirent *dent, *prev = NULL;
 	struct buffer_head *bh;
 	int error;
@@ -1714,6 +1714,8 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name)
 	gfs2_trans_add_bh(dip->i_gl, bh, 1);
 	dip->i_entries--;
 	dip->i_inode.i_mtime = dip->i_inode.i_ctime = CURRENT_TIME;
+	if (S_ISDIR(dentry->d_inode->i_mode))
+		drop_nlink(&dip->i_inode);
 	gfs2_dinode_out(dip, bh->b_data);
 	brelse(bh);
 	mark_inode_dirty(&dip->i_inode);
@@ -1768,94 +1770,20 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
 }
 
 /**
- * foreach_leaf - call a function for each leaf in a directory
- * @dip: the directory
- * @lc: the function to call for each each
- * @data: private data to pass to it
- *
- * Returns: errno
- */
-
-static int foreach_leaf(struct gfs2_inode *dip, leaf_call_t lc, void *data)
-{
-	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
-	struct buffer_head *bh;
-	struct gfs2_leaf *leaf;
-	u32 hsize, len;
-	u32 ht_offset, lp_offset, ht_offset_cur = -1;
-	u32 index = 0;
-	__be64 *lp;
-	u64 leaf_no;
-	int error = 0;
-
-	hsize = 1 << dip->i_depth;
-	if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) {
-		gfs2_consist_inode(dip);
-		return -EIO;
-	}
-
-	lp = kmalloc(sdp->sd_hash_bsize, GFP_NOFS);
-	if (!lp)
-		return -ENOMEM;
-
-	while (index < hsize) {
-		lp_offset = index & (sdp->sd_hash_ptrs - 1);
-		ht_offset = index - lp_offset;
-
-		if (ht_offset_cur != ht_offset) {
-			error = gfs2_dir_read_data(dip, (char *)lp,
-						ht_offset * sizeof(__be64),
-						sdp->sd_hash_bsize, 1);
-			if (error != sdp->sd_hash_bsize) {
-				if (error >= 0)
-					error = -EIO;
-				goto out;
-			}
-			ht_offset_cur = ht_offset;
-		}
-
-		leaf_no = be64_to_cpu(lp[lp_offset]);
-		if (leaf_no) {
-			error = get_leaf(dip, leaf_no, &bh);
-			if (error)
-				goto out;
-			leaf = (struct gfs2_leaf *)bh->b_data;
-			len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth));
-			brelse(bh);
-
-			error = lc(dip, index, len, leaf_no, data);
-			if (error)
-				goto out;
-
-			index = (index & ~(len - 1)) + len;
-		} else
-			index++;
-	}
-
-	if (index != hsize) {
-		gfs2_consist_inode(dip);
-		error = -EIO;
-	}
-
-out:
-	kfree(lp);
-
-	return error;
-}
-
-/**
  * leaf_dealloc - Deallocate a directory leaf
  * @dip: the directory
  * @index: the hash table offset in the directory
  * @len: the number of pointers to this leaf
  * @leaf_no: the leaf number
- * @data: not used
+ * @leaf_bh: buffer_head for the starting leaf
+ * last_dealloc: 1 if this is the final dealloc for the leaf, else 0
  *
  * Returns: errno
  */
 
 static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
-			u64 leaf_no, void *data)
+			u64 leaf_no, struct buffer_head *leaf_bh,
+			int last_dealloc)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
 	struct gfs2_leaf *tmp_leaf;
@@ -1887,14 +1815,18 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
 		goto out_qs;
 
 	/*  Count the number of leaves  */
+	bh = leaf_bh;
 
 	for (blk = leaf_no; blk; blk = nblk) {
-		error = get_leaf(dip, blk, &bh);
-		if (error)
-			goto out_rlist;
+		if (blk != leaf_no) {
+			error = get_leaf(dip, blk, &bh);
+			if (error)
+				goto out_rlist;
+		}
 		tmp_leaf = (struct gfs2_leaf *)bh->b_data;
 		nblk = be64_to_cpu(tmp_leaf->lf_next);
-		brelse(bh);
+		if (blk != leaf_no)
+			brelse(bh);
 
 		gfs2_rlist_add(sdp, &rlist, blk);
 		l_blocks++;
@@ -1918,13 +1850,18 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
 	if (error)
 		goto out_rg_gunlock;
 
+	bh = leaf_bh;
+
 	for (blk = leaf_no; blk; blk = nblk) {
-		error = get_leaf(dip, blk, &bh);
-		if (error)
-			goto out_end_trans;
+		if (blk != leaf_no) {
+			error = get_leaf(dip, blk, &bh);
+			if (error)
+				goto out_end_trans;
+		}
 		tmp_leaf = (struct gfs2_leaf *)bh->b_data;
 		nblk = be64_to_cpu(tmp_leaf->lf_next);
-		brelse(bh);
+		if (blk != leaf_no)
+			brelse(bh);
 
 		gfs2_free_meta(dip, blk, 1);
 		gfs2_add_inode_blocks(&dip->i_inode, -1);
@@ -1942,6 +1879,10 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
 		goto out_end_trans;
 
 	gfs2_trans_add_bh(dip->i_gl, dibh, 1);
+	/* On the last dealloc, make this a regular file in case we crash.
+	   (We don't want to free these blocks a second time.)  */
+	if (last_dealloc)
+		dip->i_inode.i_mode = S_IFREG;
 	gfs2_dinode_out(dip, dibh->b_data);
 	brelse(dibh);
 
@@ -1975,29 +1916,67 @@ int gfs2_dir_exhash_dealloc(struct gfs2_inode *dip)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
 	struct buffer_head *bh;
-	int error;
+	struct gfs2_leaf *leaf;
+	u32 hsize, len;
+	u32 ht_offset, lp_offset, ht_offset_cur = -1;
+	u32 index = 0, next_index;
+	__be64 *lp;
+	u64 leaf_no;
+	int error = 0, last;
 
-	/* Dealloc on-disk leaves to FREEMETA state */
-	error = foreach_leaf(dip, leaf_dealloc, NULL);
-	if (error)
-		return error;
+	hsize = 1 << dip->i_depth;
+	if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) {
+		gfs2_consist_inode(dip);
+		return -EIO;
+	}
 
-	/* Make this a regular file in case we crash.
-	   (We don't want to free these blocks a second time.)  */
+	lp = kmalloc(sdp->sd_hash_bsize, GFP_NOFS);
+	if (!lp)
+		return -ENOMEM;
 
-	error = gfs2_trans_begin(sdp, RES_DINODE, 0);
-	if (error)
-		return error;
+	while (index < hsize) {
+		lp_offset = index & (sdp->sd_hash_ptrs - 1);
+		ht_offset = index - lp_offset;
 
-	error = gfs2_meta_inode_buffer(dip, &bh);
-	if (!error) {
-		gfs2_trans_add_bh(dip->i_gl, bh, 1);
-		((struct gfs2_dinode *)bh->b_data)->di_mode =
-						cpu_to_be32(S_IFREG);
-		brelse(bh);
+		if (ht_offset_cur != ht_offset) {
+			error = gfs2_dir_read_data(dip, (char *)lp,
+						ht_offset * sizeof(__be64),
+						sdp->sd_hash_bsize, 1);
+			if (error != sdp->sd_hash_bsize) {
+				if (error >= 0)
+					error = -EIO;
+				goto out;
+			}
+			ht_offset_cur = ht_offset;
+		}
+
+		leaf_no = be64_to_cpu(lp[lp_offset]);
+		if (leaf_no) {
+			error = get_leaf(dip, leaf_no, &bh);
+			if (error)
+				goto out;
+			leaf = (struct gfs2_leaf *)bh->b_data;
+			len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth));
+
+			next_index = (index & ~(len - 1)) + len;
+			last = ((next_index >= hsize) ? 1 : 0);
+			error = leaf_dealloc(dip, index, len, leaf_no, bh,
+					     last);
+			brelse(bh);
+			if (error)
+				goto out;
+			index = next_index;
+		} else
+			index++;
 	}
 
-	gfs2_trans_end(sdp);
+	if (index != hsize) {
+		gfs2_consist_inode(dip);
+		error = -EIO;
+	}
+
+out:
+	kfree(lp);
 
 	return error;
 }
diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h
index a98f644bd3d..e686af11bec 100644
--- a/fs/gfs2/dir.h
+++ b/fs/gfs2/dir.h
@@ -22,8 +22,8 @@ extern struct inode *gfs2_dir_search(struct inode *dir,
 extern int gfs2_dir_check(struct inode *dir, const struct qstr *filename,
 			  const struct gfs2_inode *ip);
 extern int gfs2_dir_add(struct inode *inode, const struct qstr *filename,
-			const struct gfs2_inode *ip, unsigned int type);
-extern int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *filename);
+			const struct gfs2_inode *ip);
+extern int gfs2_dir_del(struct gfs2_inode *dip, const struct dentry *dentry);
 extern int gfs2_dir_read(struct inode *inode, u64 *offset, void *opaque,
 			 filldir_t filldir);
 extern int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c
index b5a5e60df0d..fe9945f2ff7 100644
--- a/fs/gfs2/export.c
+++ b/fs/gfs2/export.c
@@ -139,7 +139,7 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb,
 	struct gfs2_sbd *sdp = sb->s_fs_info;
 	struct inode *inode;
 
-	inode = gfs2_ilookup(sb, inum->no_addr);
+	inode = gfs2_ilookup(sb, inum->no_addr, 0);
 	if (inode) {
 		if (GFS2_I(inode)->i_no_formal_ino != inum->no_formal_ino) {
 			iput(inode);
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index b2682e073ee..a9f5cbe45cd 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -545,18 +545,10 @@ static int gfs2_close(struct inode *inode, struct file *file)
 /**
  * gfs2_fsync - sync the dirty data for a file (across the cluster)
  * @file: the file that points to the dentry (we ignore this)
- * @dentry: the dentry that points to the inode to sync
+ * @datasync: set if we can ignore timestamp changes
  *
- * The VFS will flush "normal" data for us. We only need to worry
- * about metadata here. For journaled data, we just do a log flush
- * as we can't avoid it. Otherwise we can just bale out if datasync
- * is set. For stuffed inodes we must flush the log in order to
- * ensure that all data is on disk.
- *
- * The call to write_inode_now() is there to write back metadata and
- * the inode itself. It does also try and write the data, but thats
- * (hopefully) a no-op due to the VFS having already called filemap_fdatawrite()
- * for us.
+ * The VFS will flush data for us. We only need to worry
+ * about metadata here.
  *
  * Returns: errno
  */
@@ -565,22 +557,20 @@ static int gfs2_fsync(struct file *file, int datasync)
 {
 	struct inode *inode = file->f_mapping->host;
 	int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC);
-	int ret = 0;
-
-	if (gfs2_is_jdata(GFS2_I(inode))) {
-		gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl);
-		return 0;
-	}
+	struct gfs2_inode *ip = GFS2_I(inode);
+	int ret;
 
-	if (sync_state != 0) {
-		if (!datasync)
-			ret = write_inode_now(inode, 0);
+	if (datasync)
+		sync_state &= ~I_DIRTY_SYNC;
 
-		if (gfs2_is_stuffed(GFS2_I(inode)))
-			gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl);
+	if (sync_state) {
+		ret = sync_inode_metadata(inode, 1);
+		if (ret)
+			return ret;
+		gfs2_ail_flush(ip->i_gl);
 	}
 
-	return ret;
+	return 0;
 }
 
 /**
@@ -617,18 +607,51 @@ static ssize_t gfs2_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 	return generic_file_aio_write(iocb, iov, nr_segs, pos);
 }
 
-static void empty_write_end(struct page *page, unsigned from,
-			   unsigned to)
+static int empty_write_end(struct page *page, unsigned from,
+			   unsigned to, int mode)
 {
-	struct gfs2_inode *ip = GFS2_I(page->mapping->host);
+	struct inode *inode = page->mapping->host;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct buffer_head *bh;
+	unsigned offset, blksize = 1 << inode->i_blkbits;
+	pgoff_t end_index = i_size_read(inode) >> PAGE_CACHE_SHIFT;
 
 	zero_user(page, from, to-from);
 	mark_page_accessed(page);
 
-	if (!gfs2_is_writeback(ip))
-		gfs2_page_add_databufs(ip, page, from, to);
+	if (page->index < end_index || !(mode & FALLOC_FL_KEEP_SIZE)) {
+		if (!gfs2_is_writeback(ip))
+			gfs2_page_add_databufs(ip, page, from, to);
 
-	block_commit_write(page, from, to);
+		block_commit_write(page, from, to);
+		return 0;
+	}
+
+	offset = 0;
+	bh = page_buffers(page);
+	while (offset < to) {
+		if (offset >= from) {
+			set_buffer_uptodate(bh);
+			mark_buffer_dirty(bh);
+			clear_buffer_new(bh);
+			write_dirty_buffer(bh, WRITE);
+		}
+		offset += blksize;
+		bh = bh->b_this_page;
+	}
+
+	offset = 0;
+	bh = page_buffers(page);
+	while (offset < to) {
+		if (offset >= from) {
+			wait_on_buffer(bh);
+			if (!buffer_uptodate(bh))
+				return -EIO;
+		}
+		offset += blksize;
+		bh = bh->b_this_page;
+	}
+	return 0;
 }
 
 static int needs_empty_write(sector_t block, struct inode *inode)
@@ -643,7 +666,8 @@ static int needs_empty_write(sector_t block, struct inode *inode)
 	return !buffer_mapped(&bh_map);
 }
 
-static int write_empty_blocks(struct page *page, unsigned from, unsigned to)
+static int write_empty_blocks(struct page *page, unsigned from, unsigned to,
+			      int mode)
 {
 	struct inode *inode = page->mapping->host;
 	unsigned start, end, next, blksize;
@@ -668,7 +692,9 @@ static int write_empty_blocks(struct page *page, unsigned from, unsigned to)
 							  gfs2_block_map);
 				if (unlikely(ret))
 					return ret;
-				empty_write_end(page, start, end);
+				ret = empty_write_end(page, start, end, mode);
+				if (unlikely(ret))
+					return ret;
 				end = 0;
 			}
 			start = next;
@@ -682,7 +708,9 @@ static int write_empty_blocks(struct page *page, unsigned from, unsigned to)
 		ret = __block_write_begin(page, start, end - start, gfs2_block_map);
 		if (unlikely(ret))
 			return ret;
-		empty_write_end(page, start, end);
+		ret = empty_write_end(page, start, end, mode);
+		if (unlikely(ret))
+			return ret;
 	}
 
 	return 0;
@@ -731,7 +759,7 @@ static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len,
 
 		if (curr == end)
 			to = end_offset;
-		error = write_empty_blocks(page, from, to);
+		error = write_empty_blocks(page, from, to, mode);
 		if (!error && offset + to > inode->i_size &&
 		    !(mode & FALLOC_FL_KEEP_SIZE)) {
 			i_size_write(inode, offset + to);
@@ -788,6 +816,7 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset,
 	loff_t bytes, max_bytes;
 	struct gfs2_alloc *al;
 	int error;
+	loff_t bsize_mask = ~((loff_t)sdp->sd_sb.sb_bsize - 1);
 	loff_t next = (offset + len - 1) >> sdp->sd_sb.sb_bsize_shift;
 	next = (next + 1) << sdp->sd_sb.sb_bsize_shift;
 
@@ -795,13 +824,15 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset,
 	if (mode & ~FALLOC_FL_KEEP_SIZE)
 		return -EOPNOTSUPP;
 
-	offset = (offset >> sdp->sd_sb.sb_bsize_shift) <<
-		 sdp->sd_sb.sb_bsize_shift;
+	offset &= bsize_mask;
 
 	len = next - offset;
 	bytes = sdp->sd_max_rg_data * sdp->sd_sb.sb_bsize / 2;
 	if (!bytes)
 		bytes = UINT_MAX;
+	bytes &= bsize_mask;
+	if (bytes == 0)
+		bytes = sdp->sd_sb.sb_bsize;
 
 	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &ip->i_gh);
 	error = gfs2_glock_nq(&ip->i_gh);
@@ -832,6 +863,9 @@ retry:
 		if (error) {
 			if (error == -ENOSPC && bytes > sdp->sd_sb.sb_bsize) {
 				bytes >>= 1;
+				bytes &= bsize_mask;
+				if (bytes == 0)
+					bytes = sdp->sd_sb.sb_bsize;
 				goto retry;
 			}
 			goto out_qunlock;
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index f07643e21bf..a2a6abbccc0 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -93,14 +93,12 @@ static unsigned int gl_hash(const struct gfs2_sbd *sdp,
 
 static inline void spin_lock_bucket(unsigned int hash)
 {
-	struct hlist_bl_head *bl = &gl_hash_table[hash];
-	bit_spin_lock(0, (unsigned long *)bl);
+	hlist_bl_lock(&gl_hash_table[hash]);
 }
 
 static inline void spin_unlock_bucket(unsigned int hash)
 {
-	struct hlist_bl_head *bl = &gl_hash_table[hash];
-	__bit_spin_unlock(0, (unsigned long *)bl);
+	hlist_bl_unlock(&gl_hash_table[hash]);
 }
 
 static void gfs2_glock_dealloc(struct rcu_head *rcu)
@@ -145,14 +143,9 @@ static int demote_ok(const struct gfs2_glock *gl)
 {
 	const struct gfs2_glock_operations *glops = gl->gl_ops;
 
-	/* assert_spin_locked(&gl->gl_spin); */
-
 	if (gl->gl_state == LM_ST_UNLOCKED)
 		return 0;
-	if (test_bit(GLF_LFLUSH, &gl->gl_flags))
-		return 0;
-	if ((gl->gl_name.ln_type != LM_TYPE_INODE) &&
-	    !list_empty(&gl->gl_holders))
+	if (!list_empty(&gl->gl_holders))
 		return 0;
 	if (glops->go_demote_ok)
 		return glops->go_demote_ok(gl);
@@ -160,6 +153,31 @@ static int demote_ok(const struct gfs2_glock *gl)
 }
 
 
+void gfs2_glock_add_to_lru(struct gfs2_glock *gl)
+{
+	spin_lock(&lru_lock);
+
+	if (!list_empty(&gl->gl_lru))
+		list_del_init(&gl->gl_lru);
+	else
+		atomic_inc(&lru_count);
+
+	list_add_tail(&gl->gl_lru, &lru_list);
+	set_bit(GLF_LRU, &gl->gl_flags);
+	spin_unlock(&lru_lock);
+}
+
+static void gfs2_glock_remove_from_lru(struct gfs2_glock *gl)
+{
+	spin_lock(&lru_lock);
+	if (!list_empty(&gl->gl_lru)) {
+		list_del_init(&gl->gl_lru);
+		atomic_dec(&lru_count);
+		clear_bit(GLF_LRU, &gl->gl_flags);
+	}
+	spin_unlock(&lru_lock);
+}
+
 /**
  * __gfs2_glock_schedule_for_reclaim - Add a glock to the reclaim list
  * @gl: the glock
@@ -170,24 +188,8 @@ static int demote_ok(const struct gfs2_glock *gl)
 
 static void __gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
 {
-	if (demote_ok(gl)) {
-		spin_lock(&lru_lock);
-
-		if (!list_empty(&gl->gl_lru))
-			list_del_init(&gl->gl_lru);
-		else
-			atomic_inc(&lru_count);
-
-		list_add_tail(&gl->gl_lru, &lru_list);
-		spin_unlock(&lru_lock);
-	}
-}
-
-void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
-{
-	spin_lock(&gl->gl_spin);
-	__gfs2_glock_schedule_for_reclaim(gl);
-	spin_unlock(&gl->gl_spin);
+	if (demote_ok(gl))
+		gfs2_glock_add_to_lru(gl);
 }
 
 /**
@@ -219,12 +221,7 @@ void gfs2_glock_put(struct gfs2_glock *gl)
 		spin_lock_bucket(gl->gl_hash);
 		hlist_bl_del_rcu(&gl->gl_list);
 		spin_unlock_bucket(gl->gl_hash);
-		spin_lock(&lru_lock);
-		if (!list_empty(&gl->gl_lru)) {
-			list_del_init(&gl->gl_lru);
-			atomic_dec(&lru_count);
-		}
-		spin_unlock(&lru_lock);
+		gfs2_glock_remove_from_lru(gl);
 		GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders));
 		GLOCK_BUG_ON(gl, mapping && mapping->nrpages);
 		trace_gfs2_glock_put(gl);
@@ -544,11 +541,6 @@ __acquires(&gl->gl_spin)
 	clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags);
 
 	gfs2_glock_hold(gl);
-	if (target != LM_ST_UNLOCKED && (gl->gl_state == LM_ST_SHARED ||
-	    gl->gl_state == LM_ST_DEFERRED) &&
-	    !(lck_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB)))
-		lck_flags |= LM_FLAG_TRY_1CB;
-
 	if (sdp->sd_lockstruct.ls_ops->lm_lock)	{
 		/* lock_dlm */
 		ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags);
@@ -650,7 +642,7 @@ static void delete_work_func(struct work_struct *work)
 	/* Note: Unsafe to dereference ip as we don't hold right refs/locks */
 
 	if (ip)
-		inode = gfs2_ilookup(sdp->sd_vfs, no_addr);
+		inode = gfs2_ilookup(sdp->sd_vfs, no_addr, 1);
 	else
 		inode = gfs2_lookup_by_inum(sdp, no_addr, NULL, GFS2_BLKST_UNLINKED);
 	if (inode && !IS_ERR(inode)) {
@@ -1027,6 +1019,9 @@ int gfs2_glock_nq(struct gfs2_holder *gh)
 	if (unlikely(test_bit(SDF_SHUTDOWN, &sdp->sd_flags)))
 		return -EIO;
 
+	if (test_bit(GLF_LRU, &gl->gl_flags))
+		gfs2_glock_remove_from_lru(gl);
+
 	spin_lock(&gl->gl_spin);
 	add_to_queue(gh);
 	if ((LM_FLAG_NOEXP & gh->gh_flags) &&
@@ -1084,7 +1079,8 @@ void gfs2_glock_dq(struct gfs2_holder *gh)
 		    !test_bit(GLF_DEMOTE, &gl->gl_flags))
 			fast_path = 1;
 	}
-	__gfs2_glock_schedule_for_reclaim(gl);
+	if (!test_bit(GLF_LFLUSH, &gl->gl_flags))
+		__gfs2_glock_schedule_for_reclaim(gl);
 	trace_gfs2_glock_queue(gh, 0);
 	spin_unlock(&gl->gl_spin);
 	if (likely(fast_path))
@@ -1367,6 +1363,7 @@ static int gfs2_shrink_glock_memory(struct shrinker *shrink, int nr, gfp_t gfp_m
 	while(nr && !list_empty(&lru_list)) {
 		gl = list_entry(lru_list.next, struct gfs2_glock, gl_lru);
 		list_del_init(&gl->gl_lru);
+		clear_bit(GLF_LRU, &gl->gl_flags);
 		atomic_dec(&lru_count);
 
 		/* Test for being demotable */
@@ -1389,6 +1386,7 @@ static int gfs2_shrink_glock_memory(struct shrinker *shrink, int nr, gfp_t gfp_m
 		}
 		nr_skipped++;
 		list_add(&gl->gl_lru, &skipped);
+		set_bit(GLF_LRU, &gl->gl_flags);
 	}
 	list_splice(&skipped, &lru_list);
 	atomic_add(nr_skipped, &lru_count);
@@ -1461,12 +1459,7 @@ static void thaw_glock(struct gfs2_glock *gl)
 
 static void clear_glock(struct gfs2_glock *gl)
 {
-	spin_lock(&lru_lock);
-	if (!list_empty(&gl->gl_lru)) {
-		list_del_init(&gl->gl_lru);
-		atomic_dec(&lru_count);
-	}
-	spin_unlock(&lru_lock);
+	gfs2_glock_remove_from_lru(gl);
 
 	spin_lock(&gl->gl_spin);
 	if (gl->gl_state != LM_ST_UNLOCKED)
@@ -1601,9 +1594,11 @@ static int dump_holder(struct seq_file *seq, const struct gfs2_holder *gh)
 	return 0;
 }
 
-static const char *gflags2str(char *buf, const unsigned long *gflags)
+static const char *gflags2str(char *buf, const struct gfs2_glock *gl)
 {
+	const unsigned long *gflags = &gl->gl_flags;
 	char *p = buf;
+
 	if (test_bit(GLF_LOCK, gflags))
 		*p++ = 'l';
 	if (test_bit(GLF_DEMOTE, gflags))
@@ -1626,6 +1621,10 @@ static const char *gflags2str(char *buf, const unsigned long *gflags)
 		*p++ = 'F';
 	if (test_bit(GLF_QUEUED, gflags))
 		*p++ = 'q';
+	if (test_bit(GLF_LRU, gflags))
+		*p++ = 'L';
+	if (gl->gl_object)
+		*p++ = 'o';
 	*p = 0;
 	return buf;
 }
@@ -1660,14 +1659,15 @@ static int __dump_glock(struct seq_file *seq, const struct gfs2_glock *gl)
 	dtime *= 1000000/HZ; /* demote time in uSec */
 	if (!test_bit(GLF_DEMOTE, &gl->gl_flags))
 		dtime = 0;
-	gfs2_print_dbg(seq, "G:  s:%s n:%u/%llx f:%s t:%s d:%s/%llu a:%d r:%d\n",
+	gfs2_print_dbg(seq, "G:  s:%s n:%u/%llx f:%s t:%s d:%s/%llu a:%d v:%d r:%d\n",
 		  state2str(gl->gl_state),
 		  gl->gl_name.ln_type,
 		  (unsigned long long)gl->gl_name.ln_number,
-		  gflags2str(gflags_buf, &gl->gl_flags),
+		  gflags2str(gflags_buf, gl),
 		  state2str(gl->gl_target),
 		  state2str(gl->gl_demote_state), dtime,
 		  atomic_read(&gl->gl_ail_count),
+		  atomic_read(&gl->gl_revokes),
 		  atomic_read(&gl->gl_ref));
 
 	list_for_each_entry(gh, &gl->gl_holders, gh_list) {
diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h
index aea160690e9..6b2f757b928 100644
--- a/fs/gfs2/glock.h
+++ b/fs/gfs2/glock.h
@@ -225,11 +225,10 @@ static inline int gfs2_glock_nq_init(struct gfs2_glock *gl,
 
 extern void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state);
 extern void gfs2_glock_complete(struct gfs2_glock *gl, int ret);
-extern void gfs2_reclaim_glock(struct gfs2_sbd *sdp);
 extern void gfs2_gl_hash_clear(struct gfs2_sbd *sdp);
 extern void gfs2_glock_finish_truncate(struct gfs2_inode *ip);
 extern void gfs2_glock_thaw(struct gfs2_sbd *sdp);
-extern void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl);
+extern void gfs2_glock_add_to_lru(struct gfs2_glock *gl);
 extern void gfs2_glock_free(struct gfs2_glock *gl);
 
 extern int __init gfs2_glock_init(void);
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 3754e3cbf02..8ef70f46473 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -28,33 +28,18 @@
 #include "trans.h"
 
 /**
- * ail_empty_gl - remove all buffers for a given lock from the AIL
+ * __gfs2_ail_flush - remove all buffers for a given lock from the AIL
  * @gl: the glock
  *
  * None of the buffers should be dirty, locked, or pinned.
  */
 
-static void gfs2_ail_empty_gl(struct gfs2_glock *gl)
+static void __gfs2_ail_flush(struct gfs2_glock *gl)
 {
 	struct gfs2_sbd *sdp = gl->gl_sbd;
 	struct list_head *head = &gl->gl_ail_list;
 	struct gfs2_bufdata *bd;
 	struct buffer_head *bh;
-	struct gfs2_trans tr;
-
-	memset(&tr, 0, sizeof(tr));
-	tr.tr_revokes = atomic_read(&gl->gl_ail_count);
-
-	if (!tr.tr_revokes)
-		return;
-
-	/* A shortened, inline version of gfs2_trans_begin() */
-	tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64));
-	tr.tr_ip = (unsigned long)__builtin_return_address(0);
-	INIT_LIST_HEAD(&tr.tr_list_buf);
-	gfs2_log_reserve(sdp, tr.tr_reserved);
-	BUG_ON(current->journal_info);
-	current->journal_info = &tr;
 
 	spin_lock(&sdp->sd_ail_lock);
 	while (!list_empty(head)) {
@@ -76,11 +61,51 @@ static void gfs2_ail_empty_gl(struct gfs2_glock *gl)
 	}
 	gfs2_assert_withdraw(sdp, !atomic_read(&gl->gl_ail_count));
 	spin_unlock(&sdp->sd_ail_lock);
+}
+
+
+static void gfs2_ail_empty_gl(struct gfs2_glock *gl)
+{
+	struct gfs2_sbd *sdp = gl->gl_sbd;
+	struct gfs2_trans tr;
+
+	memset(&tr, 0, sizeof(tr));
+	tr.tr_revokes = atomic_read(&gl->gl_ail_count);
+
+	if (!tr.tr_revokes)
+		return;
+
+	/* A shortened, inline version of gfs2_trans_begin() */
+	tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes, sizeof(u64));
+	tr.tr_ip = (unsigned long)__builtin_return_address(0);
+	INIT_LIST_HEAD(&tr.tr_list_buf);
+	gfs2_log_reserve(sdp, tr.tr_reserved);
+	BUG_ON(current->journal_info);
+	current->journal_info = &tr;
+
+	__gfs2_ail_flush(gl);
 
 	gfs2_trans_end(sdp);
 	gfs2_log_flush(sdp, NULL);
 }
 
+void gfs2_ail_flush(struct gfs2_glock *gl)
+{
+	struct gfs2_sbd *sdp = gl->gl_sbd;
+	unsigned int revokes = atomic_read(&gl->gl_ail_count);
+	int ret;
+
+	if (!revokes)
+		return;
+
+	ret = gfs2_trans_begin(sdp, 0, revokes);
+	if (ret)
+		return;
+	__gfs2_ail_flush(gl);
+	gfs2_trans_end(sdp);
+	gfs2_log_flush(sdp, NULL);
+}
+
 /**
  * rgrp_go_sync - sync out the metadata for this glock
  * @gl: the glock
@@ -227,6 +252,119 @@ static int inode_go_demote_ok(const struct gfs2_glock *gl)
 }
 
 /**
+ * gfs2_set_nlink - Set the inode's link count based on on-disk info
+ * @inode: The inode in question
+ * @nlink: The link count
+ *
+ * If the link count has hit zero, it must never be raised, whatever the
+ * on-disk inode might say. When new struct inodes are created the link
+ * count is set to 1, so that we can safely use this test even when reading
+ * in on disk information for the first time.
+ */
+
+static void gfs2_set_nlink(struct inode *inode, u32 nlink)
+{
+	/*
+	 * We will need to review setting the nlink count here in the
+	 * light of the forthcoming ro bind mount work. This is a reminder
+	 * to do that.
+	 */
+	if ((inode->i_nlink != nlink) && (inode->i_nlink != 0)) {
+		if (nlink == 0)
+			clear_nlink(inode);
+		else
+			inode->i_nlink = nlink;
+	}
+}
+
+static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf)
+{
+	const struct gfs2_dinode *str = buf;
+	struct timespec atime;
+	u16 height, depth;
+
+	if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr)))
+		goto corrupt;
+	ip->i_no_formal_ino = be64_to_cpu(str->di_num.no_formal_ino);
+	ip->i_inode.i_mode = be32_to_cpu(str->di_mode);
+	ip->i_inode.i_rdev = 0;
+	switch (ip->i_inode.i_mode & S_IFMT) {
+	case S_IFBLK:
+	case S_IFCHR:
+		ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major),
+					   be32_to_cpu(str->di_minor));
+		break;
+	};
+
+	ip->i_inode.i_uid = be32_to_cpu(str->di_uid);
+	ip->i_inode.i_gid = be32_to_cpu(str->di_gid);
+	gfs2_set_nlink(&ip->i_inode, be32_to_cpu(str->di_nlink));
+	i_size_write(&ip->i_inode, be64_to_cpu(str->di_size));
+	gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks));
+	atime.tv_sec = be64_to_cpu(str->di_atime);
+	atime.tv_nsec = be32_to_cpu(str->di_atime_nsec);
+	if (timespec_compare(&ip->i_inode.i_atime, &atime) < 0)
+		ip->i_inode.i_atime = atime;
+	ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime);
+	ip->i_inode.i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec);
+	ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime);
+	ip->i_inode.i_ctime.tv_nsec = be32_to_cpu(str->di_ctime_nsec);
+
+	ip->i_goal = be64_to_cpu(str->di_goal_meta);
+	ip->i_generation = be64_to_cpu(str->di_generation);
+
+	ip->i_diskflags = be32_to_cpu(str->di_flags);
+	gfs2_set_inode_flags(&ip->i_inode);
+	height = be16_to_cpu(str->di_height);
+	if (unlikely(height > GFS2_MAX_META_HEIGHT))
+		goto corrupt;
+	ip->i_height = (u8)height;
+
+	depth = be16_to_cpu(str->di_depth);
+	if (unlikely(depth > GFS2_DIR_MAX_DEPTH))
+		goto corrupt;
+	ip->i_depth = (u8)depth;
+	ip->i_entries = be32_to_cpu(str->di_entries);
+
+	ip->i_eattr = be64_to_cpu(str->di_eattr);
+	if (S_ISREG(ip->i_inode.i_mode))
+		gfs2_set_aops(&ip->i_inode);
+
+	return 0;
+corrupt:
+	gfs2_consist_inode(ip);
+	return -EIO;
+}
+
+/**
+ * gfs2_inode_refresh - Refresh the incore copy of the dinode
+ * @ip: The GFS2 inode
+ *
+ * Returns: errno
+ */
+
+int gfs2_inode_refresh(struct gfs2_inode *ip)
+{
+	struct buffer_head *dibh;
+	int error;
+
+	error = gfs2_meta_inode_buffer(ip, &dibh);
+	if (error)
+		return error;
+
+	if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), dibh, GFS2_METATYPE_DI)) {
+		brelse(dibh);
+		return -EIO;
+	}
+
+	error = gfs2_dinode_in(ip, dibh->b_data);
+	brelse(dibh);
+	clear_bit(GIF_INVALID, &ip->i_flags);
+
+	return error;
+}
+
+/**
  * inode_go_lock - operation done after an inode lock is locked by a process
  * @gl: the glock
  * @flags:
@@ -385,6 +523,10 @@ static int trans_go_demote_ok(const struct gfs2_glock *gl)
 static void iopen_go_callback(struct gfs2_glock *gl)
 {
 	struct gfs2_inode *ip = (struct gfs2_inode *)gl->gl_object;
+	struct gfs2_sbd *sdp = gl->gl_sbd;
+
+	if (sdp->sd_vfs->s_flags & MS_RDONLY)
+		return;
 
 	if (gl->gl_demote_state == LM_ST_UNLOCKED &&
 	    gl->gl_state == LM_ST_SHARED && ip) {
diff --git a/fs/gfs2/glops.h b/fs/gfs2/glops.h
index b3aa2e3210f..6fce409b5a5 100644
--- a/fs/gfs2/glops.h
+++ b/fs/gfs2/glops.h
@@ -23,4 +23,6 @@ extern const struct gfs2_glock_operations gfs2_quota_glops;
 extern const struct gfs2_glock_operations gfs2_journal_glops;
 extern const struct gfs2_glock_operations *gfs2_glops_list[];
 
+extern void gfs2_ail_flush(struct gfs2_glock *gl);
+
 #endif /* __GLOPS_DOT_H__ */
diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h
index 870a89d6d4d..0a064e91ac7 100644
--- a/fs/gfs2/incore.h
+++ b/fs/gfs2/incore.h
@@ -20,7 +20,6 @@
 
 #define DIO_WAIT	0x00000010
 #define DIO_METADATA	0x00000020
-#define DIO_ALL		0x00000100
 
 struct gfs2_log_operations;
 struct gfs2_log_element;
@@ -200,6 +199,8 @@ enum {
 	GLF_INITIAL			= 10,
 	GLF_FROZEN			= 11,
 	GLF_QUEUED			= 12,
+	GLF_LRU				= 13,
+	GLF_OBJECT			= 14, /* Used only for tracing */
 };
 
 struct gfs2_glock {
@@ -234,6 +235,7 @@ struct gfs2_glock {
 
 	struct list_head gl_ail_list;
 	atomic_t gl_ail_count;
+	atomic_t gl_revokes;
 	struct delayed_work gl_work;
 	struct work_struct gl_delete;
 	struct rcu_head gl_rcu;
@@ -374,8 +376,6 @@ struct gfs2_ail {
 	unsigned int ai_first;
 	struct list_head ai_ail1_list;
 	struct list_head ai_ail2_list;
-
-	u64 ai_sync_gen;
 };
 
 struct gfs2_journal_extent {
@@ -488,7 +488,6 @@ struct gfs2_sb_host {
 
 	char sb_lockproto[GFS2_LOCKNAME_LEN];
 	char sb_locktable[GFS2_LOCKNAME_LEN];
-	u8 sb_uuid[16];
 };
 
 /*
@@ -654,7 +653,6 @@ struct gfs2_sbd {
 	spinlock_t sd_ail_lock;
 	struct list_head sd_ail1_list;
 	struct list_head sd_ail2_list;
-	u64 sd_ail_sync_gen;
 
 	/* Replay stuff */
 
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 97d54a28776..03e0c529063 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -1,23 +1,25 @@
 /*
  * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
- * Copyright (C) 2004-2008 Red Hat, Inc.  All rights reserved.
+ * Copyright (C) 2004-2011 Red Hat, Inc.  All rights reserved.
  *
  * This copyrighted material is made available to anyone wishing to use,
  * modify, copy, or redistribute it subject to the terms and conditions
  * of the GNU General Public License version 2.
  */
 
-#include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 #include <linux/completion.h>
 #include <linux/buffer_head.h>
+#include <linux/namei.h>
+#include <linux/mm.h>
+#include <linux/xattr.h>
 #include <linux/posix_acl.h>
-#include <linux/sort.h>
 #include <linux/gfs2_ondisk.h>
 #include <linux/crc32.h>
+#include <linux/fiemap.h>
 #include <linux/security.h>
-#include <linux/time.h>
+#include <asm/uaccess.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -26,51 +28,70 @@
 #include "dir.h"
 #include "xattr.h"
 #include "glock.h"
-#include "glops.h"
 #include "inode.h"
-#include "log.h"
 #include "meta_io.h"
 #include "quota.h"
 #include "rgrp.h"
 #include "trans.h"
 #include "util.h"
+#include "super.h"
+#include "glops.h"
 
-struct gfs2_inum_range_host {
-	u64 ir_start;
-	u64 ir_length;
+struct gfs2_skip_data {
+	u64 no_addr;
+	int skipped;
+	int non_block;
 };
 
 static int iget_test(struct inode *inode, void *opaque)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
-	u64 *no_addr = opaque;
+	struct gfs2_skip_data *data = opaque;
 
-	if (ip->i_no_addr == *no_addr)
+	if (ip->i_no_addr == data->no_addr) {
+		if (data->non_block &&
+		    inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) {
+			data->skipped = 1;
+			return 0;
+		}
 		return 1;
-
+	}
 	return 0;
 }
 
 static int iget_set(struct inode *inode, void *opaque)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
-	u64 *no_addr = opaque;
+	struct gfs2_skip_data *data = opaque;
 
-	inode->i_ino = (unsigned long)*no_addr;
-	ip->i_no_addr = *no_addr;
+	if (data->skipped)
+		return -ENOENT;
+	inode->i_ino = (unsigned long)(data->no_addr);
+	ip->i_no_addr = data->no_addr;
 	return 0;
 }
 
-struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr)
+struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr, int non_block)
 {
 	unsigned long hash = (unsigned long)no_addr;
-	return ilookup5(sb, hash, iget_test, &no_addr);
+	struct gfs2_skip_data data;
+
+	data.no_addr = no_addr;
+	data.skipped = 0;
+	data.non_block = non_block;
+	return ilookup5(sb, hash, iget_test, &data);
 }
 
-static struct inode *gfs2_iget(struct super_block *sb, u64 no_addr)
+static struct inode *gfs2_iget(struct super_block *sb, u64 no_addr,
+			       int non_block)
 {
+	struct gfs2_skip_data data;
 	unsigned long hash = (unsigned long)no_addr;
-	return iget5_locked(sb, hash, iget_test, iget_set, &no_addr);
+
+	data.no_addr = no_addr;
+	data.skipped = 0;
+	data.non_block = non_block;
+	return iget5_locked(sb, hash, iget_test, iget_set, &data);
 }
 
 /**
@@ -111,19 +132,20 @@ static void gfs2_set_iop(struct inode *inode)
  * @sb: The super block
  * @no_addr: The inode number
  * @type: The type of the inode
+ * non_block: Can we block on inodes that are being freed?
  *
  * Returns: A VFS inode, or an error
  */
 
 struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type,
-				u64 no_addr, u64 no_formal_ino)
+				u64 no_addr, u64 no_formal_ino, int non_block)
 {
 	struct inode *inode;
 	struct gfs2_inode *ip;
 	struct gfs2_glock *io_gl = NULL;
 	int error;
 
-	inode = gfs2_iget(sb, no_addr);
+	inode = gfs2_iget(sb, no_addr, non_block);
 	ip = GFS2_I(inode);
 
 	if (!inode)
@@ -185,11 +207,12 @@ struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
 {
 	struct super_block *sb = sdp->sd_vfs;
 	struct gfs2_holder i_gh;
-	struct inode *inode;
+	struct inode *inode = NULL;
 	int error;
 
+	/* Must not read in block until block type is verified */
 	error = gfs2_glock_nq_num(sdp, no_addr, &gfs2_inode_glops,
-				  LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
+				  LM_ST_EXCLUSIVE, GL_SKIP, &i_gh);
 	if (error)
 		return ERR_PTR(error);
 
@@ -197,7 +220,7 @@ struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
 	if (error)
 		goto fail;
 
-	inode = gfs2_inode_lookup(sb, DT_UNKNOWN, no_addr, 0);
+	inode = gfs2_inode_lookup(sb, DT_UNKNOWN, no_addr, 0, 1);
 	if (IS_ERR(inode))
 		goto fail;
 
@@ -222,203 +245,6 @@ fail_iput:
 	goto fail;
 }
 
-static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf)
-{
-	const struct gfs2_dinode *str = buf;
-	struct timespec atime;
-	u16 height, depth;
-
-	if (unlikely(ip->i_no_addr != be64_to_cpu(str->di_num.no_addr)))
-		goto corrupt;
-	ip->i_no_formal_ino = be64_to_cpu(str->di_num.no_formal_ino);
-	ip->i_inode.i_mode = be32_to_cpu(str->di_mode);
-	ip->i_inode.i_rdev = 0;
-	switch (ip->i_inode.i_mode & S_IFMT) {
-	case S_IFBLK:
-	case S_IFCHR:
-		ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major),
-					   be32_to_cpu(str->di_minor));
-		break;
-	};
-
-	ip->i_inode.i_uid = be32_to_cpu(str->di_uid);
-	ip->i_inode.i_gid = be32_to_cpu(str->di_gid);
-	/*
-	 * We will need to review setting the nlink count here in the
-	 * light of the forthcoming ro bind mount work. This is a reminder
-	 * to do that.
-	 */
-	ip->i_inode.i_nlink = be32_to_cpu(str->di_nlink);
-	i_size_write(&ip->i_inode, be64_to_cpu(str->di_size));
-	gfs2_set_inode_blocks(&ip->i_inode, be64_to_cpu(str->di_blocks));
-	atime.tv_sec = be64_to_cpu(str->di_atime);
-	atime.tv_nsec = be32_to_cpu(str->di_atime_nsec);
-	if (timespec_compare(&ip->i_inode.i_atime, &atime) < 0)
-		ip->i_inode.i_atime = atime;
-	ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime);
-	ip->i_inode.i_mtime.tv_nsec = be32_to_cpu(str->di_mtime_nsec);
-	ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime);
-	ip->i_inode.i_ctime.tv_nsec = be32_to_cpu(str->di_ctime_nsec);
-
-	ip->i_goal = be64_to_cpu(str->di_goal_meta);
-	ip->i_generation = be64_to_cpu(str->di_generation);
-
-	ip->i_diskflags = be32_to_cpu(str->di_flags);
-	gfs2_set_inode_flags(&ip->i_inode);
-	height = be16_to_cpu(str->di_height);
-	if (unlikely(height > GFS2_MAX_META_HEIGHT))
-		goto corrupt;
-	ip->i_height = (u8)height;
-
-	depth = be16_to_cpu(str->di_depth);
-	if (unlikely(depth > GFS2_DIR_MAX_DEPTH))
-		goto corrupt;
-	ip->i_depth = (u8)depth;
-	ip->i_entries = be32_to_cpu(str->di_entries);
-
-	ip->i_eattr = be64_to_cpu(str->di_eattr);
-	if (S_ISREG(ip->i_inode.i_mode))
-		gfs2_set_aops(&ip->i_inode);
-
-	return 0;
-corrupt:
-	if (gfs2_consist_inode(ip))
-		gfs2_dinode_print(ip);
-	return -EIO;
-}
-
-/**
- * gfs2_inode_refresh - Refresh the incore copy of the dinode
- * @ip: The GFS2 inode
- *
- * Returns: errno
- */
-
-int gfs2_inode_refresh(struct gfs2_inode *ip)
-{
-	struct buffer_head *dibh;
-	int error;
-
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-	if (error)
-		return error;
-
-	if (gfs2_metatype_check(GFS2_SB(&ip->i_inode), dibh, GFS2_METATYPE_DI)) {
-		brelse(dibh);
-		return -EIO;
-	}
-
-	error = gfs2_dinode_in(ip, dibh->b_data);
-	brelse(dibh);
-	clear_bit(GIF_INVALID, &ip->i_flags);
-
-	return error;
-}
-
-int gfs2_dinode_dealloc(struct gfs2_inode *ip)
-{
-	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
-	struct gfs2_alloc *al;
-	struct gfs2_rgrpd *rgd;
-	int error;
-
-	if (gfs2_get_inode_blocks(&ip->i_inode) != 1) {
-		if (gfs2_consist_inode(ip))
-			gfs2_dinode_print(ip);
-		return -EIO;
-	}
-
-	al = gfs2_alloc_get(ip);
-	if (!al)
-		return -ENOMEM;
-
-	error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE);
-	if (error)
-		goto out;
-
-	error = gfs2_rindex_hold(sdp, &al->al_ri_gh);
-	if (error)
-		goto out_qs;
-
-	rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr);
-	if (!rgd) {
-		gfs2_consist_inode(ip);
-		error = -EIO;
-		goto out_rindex_relse;
-	}
-
-	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0,
-				   &al->al_rgd_gh);
-	if (error)
-		goto out_rindex_relse;
-
-	error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_STATFS + RES_QUOTA, 1);
-	if (error)
-		goto out_rg_gunlock;
-
-	set_bit(GLF_DIRTY, &ip->i_gl->gl_flags);
-	set_bit(GLF_LFLUSH, &ip->i_gl->gl_flags);
-
-	gfs2_free_di(rgd, ip);
-
-	gfs2_trans_end(sdp);
-
-out_rg_gunlock:
-	gfs2_glock_dq_uninit(&al->al_rgd_gh);
-out_rindex_relse:
-	gfs2_glock_dq_uninit(&al->al_ri_gh);
-out_qs:
-	gfs2_quota_unhold(ip);
-out:
-	gfs2_alloc_put(ip);
-	return error;
-}
-
-/**
- * gfs2_change_nlink - Change nlink count on inode
- * @ip: The GFS2 inode
- * @diff: The change in the nlink count required
- *
- * Returns: errno
- */
-int gfs2_change_nlink(struct gfs2_inode *ip, int diff)
-{
-	struct buffer_head *dibh;
-	u32 nlink;
-	int error;
-
-	BUG_ON(diff != 1 && diff != -1);
-	nlink = ip->i_inode.i_nlink + diff;
-
-	/* If we are reducing the nlink count, but the new value ends up being
-	   bigger than the old one, we must have underflowed. */
-	if (diff < 0 && nlink > ip->i_inode.i_nlink) {
-		if (gfs2_consist_inode(ip))
-			gfs2_dinode_print(ip);
-		return -EIO;
-	}
-
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-	if (error)
-		return error;
-
-	if (diff > 0)
-		inc_nlink(&ip->i_inode);
-	else
-		drop_nlink(&ip->i_inode);
-
-	ip->i_inode.i_ctime = CURRENT_TIME;
-
-	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
-	gfs2_dinode_out(ip, dibh->b_data);
-	brelse(dibh);
-	mark_inode_dirty(&ip->i_inode);
-
-	if (ip->i_inode.i_nlink == 0)
-		gfs2_unlink_di(&ip->i_inode); /* mark inode unlinked */
-
-	return error;
-}
 
 struct inode *gfs2_lookup_simple(struct inode *dip, const char *name)
 {
@@ -517,7 +343,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name,
 
 	/*  Don't create entries in an unlinked directory  */
 	if (!dip->i_inode.i_nlink)
-		return -EPERM;
+		return -ENOENT;
 
 	error = gfs2_dir_check(&dip->i_inode, name, NULL);
 	switch (error) {
@@ -587,21 +413,44 @@ out:
 	return error;
 }
 
+static void gfs2_init_dir(struct buffer_head *dibh,
+			  const struct gfs2_inode *parent)
+{
+	struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data;
+	struct gfs2_dirent *dent = (struct gfs2_dirent *)(di+1);
+
+	gfs2_qstr2dirent(&gfs2_qdot, GFS2_DIRENT_SIZE(gfs2_qdot.len), dent);
+	dent->de_inum = di->di_num; /* already GFS2 endian */
+	dent->de_type = cpu_to_be16(DT_DIR);
+
+	dent = (struct gfs2_dirent *)((char*)dent + GFS2_DIRENT_SIZE(1));
+	gfs2_qstr2dirent(&gfs2_qdotdot, dibh->b_size - GFS2_DIRENT_SIZE(1) - sizeof(struct gfs2_dinode), dent);
+	gfs2_inum_out(parent, dent);
+	dent->de_type = cpu_to_be16(DT_DIR);
+	
+}
+
 /**
  * init_dinode - Fill in a new dinode structure
- * @dip: the directory this inode is being created in
+ * @dip: The directory this inode is being created in
  * @gl: The glock covering the new inode
- * @inum: the inode number
- * @mode: the file permissions
- * @uid:
- * @gid:
+ * @inum: The inode number
+ * @mode: The file permissions
+ * @uid: The uid of the new inode
+ * @gid: The gid of the new inode
+ * @generation: The generation number of the new inode
+ * @dev: The device number (if a device node)
+ * @symname: The symlink destination (if a symlink)
+ * @size: The inode size (ignored for directories)
+ * @bhp: The buffer head (returned to caller)
  *
  */
 
 static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl,
 			const struct gfs2_inum_host *inum, unsigned int mode,
 			unsigned int uid, unsigned int gid,
-			const u64 *generation, dev_t dev, struct buffer_head **bhp)
+			const u64 *generation, dev_t dev, const char *symname,
+			unsigned size, struct buffer_head **bhp)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
 	struct gfs2_dinode *di;
@@ -620,7 +469,7 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl,
 	di->di_uid = cpu_to_be32(uid);
 	di->di_gid = cpu_to_be32(gid);
 	di->di_nlink = 0;
-	di->di_size = 0;
+	di->di_size = cpu_to_be64(size);
 	di->di_blocks = cpu_to_be64(1);
 	di->di_atime = di->di_mtime = di->di_ctime = cpu_to_be64(tv.tv_sec);
 	di->di_major = cpu_to_be32(MAJOR(dev));
@@ -628,16 +477,6 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl,
 	di->di_goal_meta = di->di_goal_data = cpu_to_be64(inum->no_addr);
 	di->di_generation = cpu_to_be64(*generation);
 	di->di_flags = 0;
-
-	if (S_ISREG(mode)) {
-		if ((dip->i_diskflags & GFS2_DIF_INHERIT_JDATA) ||
-		    gfs2_tune_get(sdp, gt_new_files_jdata))
-			di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA);
-	} else if (S_ISDIR(mode)) {
-		di->di_flags |= cpu_to_be32(dip->i_diskflags &
-					    GFS2_DIF_INHERIT_JDATA);
-	}
-
 	di->__pad1 = 0;
 	di->di_payload_format = cpu_to_be32(S_ISDIR(mode) ? GFS2_FORMAT_DE : 0);
 	di->di_height = 0;
@@ -651,7 +490,26 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl,
 	di->di_mtime_nsec = cpu_to_be32(tv.tv_nsec);
 	di->di_ctime_nsec = cpu_to_be32(tv.tv_nsec);
 	memset(&di->di_reserved, 0, sizeof(di->di_reserved));
-	
+
+	switch(mode & S_IFMT) {	
+	case S_IFREG:
+		if ((dip->i_diskflags & GFS2_DIF_INHERIT_JDATA) ||
+		    gfs2_tune_get(sdp, gt_new_files_jdata))
+			di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA);
+		break;
+	case S_IFDIR:
+		di->di_flags |= cpu_to_be32(dip->i_diskflags &
+					    GFS2_DIF_INHERIT_JDATA);
+		di->di_flags |= cpu_to_be32(GFS2_DIF_JDATA);
+		di->di_size = cpu_to_be64(sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode));
+		di->di_entries = cpu_to_be32(2);
+		gfs2_init_dir(dibh, dip);
+		break;
+	case S_IFLNK:
+		memcpy(dibh->b_data + sizeof(struct gfs2_dinode), symname, size);
+		break;
+	}
+
 	set_buffer_uptodate(dibh);
 
 	*bhp = dibh;
@@ -659,7 +517,8 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl,
 
 static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl,
 		       unsigned int mode, const struct gfs2_inum_host *inum,
-		       const u64 *generation, dev_t dev, struct buffer_head **bhp)
+		       const u64 *generation, dev_t dev, const char *symname,
+		       unsigned int size, struct buffer_head **bhp)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
 	unsigned int uid, gid;
@@ -681,7 +540,7 @@ static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl,
 	if (error)
 		goto out_quota;
 
-	init_dinode(dip, gl, inum, mode, uid, gid, generation, dev, bhp);
+	init_dinode(dip, gl, inum, mode, uid, gid, generation, dev, symname, size, bhp);
 	gfs2_quota_change(dip, +1, uid, gid);
 	gfs2_trans_end(sdp);
 
@@ -735,14 +594,16 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name,
 			goto fail_quota_locks;
 	}
 
-	error = gfs2_dir_add(&dip->i_inode, name, ip, IF2DT(ip->i_inode.i_mode));
+	error = gfs2_dir_add(&dip->i_inode, name, ip);
 	if (error)
 		goto fail_end_trans;
 
 	error = gfs2_meta_inode_buffer(ip, &dibh);
 	if (error)
 		goto fail_end_trans;
-	ip->i_inode.i_nlink = 1;
+	inc_nlink(&ip->i_inode);
+	if (S_ISDIR(ip->i_inode.i_mode))
+		inc_nlink(&ip->i_inode);
 	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
 	gfs2_dinode_out(ip, dibh->b_data);
 	brelse(dibh);
@@ -789,27 +650,25 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip,
 }
 
 /**
- * gfs2_createi - Create a new inode
- * @ghs: An array of two holders
- * @name: The name of the new file
- * @mode: the permissions on the new inode
+ * gfs2_create_inode - Create a new inode
+ * @dir: The parent directory
+ * @dentry: The new dentry
+ * @mode: The permissions on the new inode
+ * @dev: For device nodes, this is the device number
+ * @symname: For symlinks, this is the link destination
+ * @size: The initial size of the inode (ignored for directories)
  *
- * @ghs[0] is an initialized holder for the directory
- * @ghs[1] is the holder for the inode lock
- *
- * If the return value is not NULL, the glocks on both the directory and the new
- * file are held.  A transaction has been started and an inplace reservation
- * is held, as well.
- *
- * Returns: An inode
+ * Returns: 0 on success, or error code
  */
 
-struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
-			   unsigned int mode, dev_t dev)
+static int gfs2_create_inode(struct inode *dir, struct dentry *dentry,
+			     unsigned int mode, dev_t dev, const char *symname,
+			     unsigned int size)
 {
+	const struct qstr *name = &dentry->d_name;
+	struct gfs2_holder ghs[2];
 	struct inode *inode = NULL;
-	struct gfs2_inode *dip = ghs->gh_gl->gl_object;
-	struct inode *dir = &dip->i_inode;
+	struct gfs2_inode *dip = GFS2_I(dir);
 	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
 	struct gfs2_inum_host inum = { .no_addr = 0, .no_formal_ino = 0 };
 	int error;
@@ -817,10 +676,9 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
 	struct buffer_head *bh = NULL;
 
 	if (!name->len || name->len > GFS2_FNAMESIZE)
-		return ERR_PTR(-ENAMETOOLONG);
+		return -ENAMETOOLONG;
 
-	gfs2_holder_reinit(LM_ST_EXCLUSIVE, 0, ghs);
-	error = gfs2_glock_nq(ghs);
+	error = gfs2_glock_nq_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
 	if (error)
 		goto fail;
 
@@ -838,12 +696,12 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
 	if (error)
 		goto fail_gunlock;
 
-	error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation, dev, &bh);
+	error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation, dev, symname, size, &bh);
 	if (error)
 		goto fail_gunlock2;
 
 	inode = gfs2_inode_lookup(dir->i_sb, IF2DT(mode), inum.no_addr,
-				  inum.no_formal_ino);
+				  inum.no_formal_ino, 0);
 	if (IS_ERR(inode))
 		goto fail_gunlock2;
 
@@ -865,18 +723,852 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
 
 	if (bh)
 		brelse(bh);
-	return inode;
+
+	gfs2_trans_end(sdp);
+	if (dip->i_alloc->al_rgd)
+		gfs2_inplace_release(dip);
+	gfs2_quota_unlock(dip);
+	gfs2_alloc_put(dip);
+	gfs2_glock_dq_uninit_m(2, ghs);
+	mark_inode_dirty(inode);
+	d_instantiate(dentry, inode);
+	return 0;
 
 fail_gunlock2:
 	gfs2_glock_dq_uninit(ghs + 1);
 	if (inode && !IS_ERR(inode))
 		iput(inode);
 fail_gunlock:
-	gfs2_glock_dq(ghs);
+	gfs2_glock_dq_uninit(ghs);
 fail:
 	if (bh)
 		brelse(bh);
-	return ERR_PTR(error);
+	return error;
+}
+
+/**
+ * gfs2_create - Create a file
+ * @dir: The directory in which to create the file
+ * @dentry: The dentry of the new file
+ * @mode: The mode of the new file
+ *
+ * Returns: errno
+ */
+
+static int gfs2_create(struct inode *dir, struct dentry *dentry,
+		       int mode, struct nameidata *nd)
+{
+	struct inode *inode;
+	int ret;
+
+	for (;;) {
+		ret = gfs2_create_inode(dir, dentry, S_IFREG | mode, 0, NULL, 0);
+		if (ret != -EEXIST || (nd && (nd->flags & LOOKUP_EXCL)))
+			return ret;
+
+		inode = gfs2_lookupi(dir, &dentry->d_name, 0);
+		if (inode) {
+			if (!IS_ERR(inode))
+				break;
+			return PTR_ERR(inode);
+		}
+	}
+
+	d_instantiate(dentry, inode);
+	return 0;
+}
+
+/**
+ * gfs2_lookup - Look up a filename in a directory and return its inode
+ * @dir: The directory inode
+ * @dentry: The dentry of the new inode
+ * @nd: passed from Linux VFS, ignored by us
+ *
+ * Called by the VFS layer. Lock dir and call gfs2_lookupi()
+ *
+ * Returns: errno
+ */
+
+static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry,
+				  struct nameidata *nd)
+{
+	struct inode *inode = NULL;
+
+	inode = gfs2_lookupi(dir, &dentry->d_name, 0);
+	if (inode && IS_ERR(inode))
+		return ERR_CAST(inode);
+
+	if (inode) {
+		struct gfs2_glock *gl = GFS2_I(inode)->i_gl;
+		struct gfs2_holder gh;
+		int error;
+		error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
+		if (error) {
+			iput(inode);
+			return ERR_PTR(error);
+		}
+		gfs2_glock_dq_uninit(&gh);
+		return d_splice_alias(inode, dentry);
+	}
+	d_add(dentry, inode);
+
+	return NULL;
+}
+
+/**
+ * gfs2_link - Link to a file
+ * @old_dentry: The inode to link
+ * @dir: Add link to this directory
+ * @dentry: The name of the link
+ *
+ * Link the inode in "old_dentry" into the directory "dir" with the
+ * name in "dentry".
+ *
+ * Returns: errno
+ */
+
+static int gfs2_link(struct dentry *old_dentry, struct inode *dir,
+		     struct dentry *dentry)
+{
+	struct gfs2_inode *dip = GFS2_I(dir);
+	struct gfs2_sbd *sdp = GFS2_SB(dir);
+	struct inode *inode = old_dentry->d_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder ghs[2];
+	struct buffer_head *dibh;
+	int alloc_required;
+	int error;
+
+	if (S_ISDIR(inode->i_mode))
+		return -EPERM;
+
+	gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
+	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1);
+
+	error = gfs2_glock_nq(ghs); /* parent */
+	if (error)
+		goto out_parent;
+
+	error = gfs2_glock_nq(ghs + 1); /* child */
+	if (error)
+		goto out_child;
+
+	error = -ENOENT;
+	if (inode->i_nlink == 0)
+		goto out_gunlock;
+
+	error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC, 0);
+	if (error)
+		goto out_gunlock;
+
+	error = gfs2_dir_check(dir, &dentry->d_name, NULL);
+	switch (error) {
+	case -ENOENT:
+		break;
+	case 0:
+		error = -EEXIST;
+	default:
+		goto out_gunlock;
+	}
+
+	error = -EINVAL;
+	if (!dip->i_inode.i_nlink)
+		goto out_gunlock;
+	error = -EFBIG;
+	if (dip->i_entries == (u32)-1)
+		goto out_gunlock;
+	error = -EPERM;
+	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
+		goto out_gunlock;
+	error = -EINVAL;
+	if (!ip->i_inode.i_nlink)
+		goto out_gunlock;
+	error = -EMLINK;
+	if (ip->i_inode.i_nlink == (u32)-1)
+		goto out_gunlock;
+
+	alloc_required = error = gfs2_diradd_alloc_required(dir, &dentry->d_name);
+	if (error < 0)
+		goto out_gunlock;
+	error = 0;
+
+	if (alloc_required) {
+		struct gfs2_alloc *al = gfs2_alloc_get(dip);
+		if (!al) {
+			error = -ENOMEM;
+			goto out_gunlock;
+		}
+
+		error = gfs2_quota_lock_check(dip);
+		if (error)
+			goto out_alloc;
+
+		al->al_requested = sdp->sd_max_dirres;
+
+		error = gfs2_inplace_reserve(dip);
+		if (error)
+			goto out_gunlock_q;
+
+		error = gfs2_trans_begin(sdp, sdp->sd_max_dirres +
+					 gfs2_rg_blocks(al) +
+					 2 * RES_DINODE + RES_STATFS +
+					 RES_QUOTA, 0);
+		if (error)
+			goto out_ipres;
+	} else {
+		error = gfs2_trans_begin(sdp, 2 * RES_DINODE + RES_LEAF, 0);
+		if (error)
+			goto out_ipres;
+	}
+
+	error = gfs2_meta_inode_buffer(ip, &dibh);
+	if (error)
+		goto out_end_trans;
+
+	error = gfs2_dir_add(dir, &dentry->d_name, ip);
+	if (error)
+		goto out_brelse;
+
+	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+	inc_nlink(&ip->i_inode);
+	ip->i_inode.i_ctime = CURRENT_TIME;
+	gfs2_dinode_out(ip, dibh->b_data);
+	mark_inode_dirty(&ip->i_inode);
+
+out_brelse:
+	brelse(dibh);
+out_end_trans:
+	gfs2_trans_end(sdp);
+out_ipres:
+	if (alloc_required)
+		gfs2_inplace_release(dip);
+out_gunlock_q:
+	if (alloc_required)
+		gfs2_quota_unlock(dip);
+out_alloc:
+	if (alloc_required)
+		gfs2_alloc_put(dip);
+out_gunlock:
+	gfs2_glock_dq(ghs + 1);
+out_child:
+	gfs2_glock_dq(ghs);
+out_parent:
+	gfs2_holder_uninit(ghs);
+	gfs2_holder_uninit(ghs + 1);
+	if (!error) {
+		ihold(inode);
+		d_instantiate(dentry, inode);
+		mark_inode_dirty(inode);
+	}
+	return error;
+}
+
+/*
+ * gfs2_unlink_ok - check to see that a inode is still in a directory
+ * @dip: the directory
+ * @name: the name of the file
+ * @ip: the inode
+ *
+ * Assumes that the lock on (at least) @dip is held.
+ *
+ * Returns: 0 if the parent/child relationship is correct, errno if it isn't
+ */
+
+static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name,
+			  const struct gfs2_inode *ip)
+{
+	int error;
+
+	if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode))
+		return -EPERM;
+
+	if ((dip->i_inode.i_mode & S_ISVTX) &&
+	    dip->i_inode.i_uid != current_fsuid() &&
+	    ip->i_inode.i_uid != current_fsuid() && !capable(CAP_FOWNER))
+		return -EPERM;
+
+	if (IS_APPEND(&dip->i_inode))
+		return -EPERM;
+
+	error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0);
+	if (error)
+		return error;
+
+	error = gfs2_dir_check(&dip->i_inode, name, ip);
+	if (error)
+		return error;
+
+	return 0;
+}
+
+/**
+ * gfs2_unlink_inode - Removes an inode from its parent dir and unlinks it
+ * @dip: The parent directory
+ * @name: The name of the entry in the parent directory
+ * @bh: The inode buffer for the inode to be removed
+ * @inode: The inode to be removed
+ *
+ * Called with all the locks and in a transaction. This will only be
+ * called for a directory after it has been checked to ensure it is empty.
+ *
+ * Returns: 0 on success, or an error
+ */
+
+static int gfs2_unlink_inode(struct gfs2_inode *dip,
+			     const struct dentry *dentry,
+			     struct buffer_head *bh)
+{
+	struct inode *inode = dentry->d_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	int error;
+
+	error = gfs2_dir_del(dip, dentry);
+	if (error)
+		return error;
+
+	ip->i_entries = 0;
+	inode->i_ctime = CURRENT_TIME;
+	if (S_ISDIR(inode->i_mode))
+		clear_nlink(inode);
+	else
+		drop_nlink(inode);
+	gfs2_trans_add_bh(ip->i_gl, bh, 1);
+	gfs2_dinode_out(ip, bh->b_data);
+	mark_inode_dirty(inode);
+	if (inode->i_nlink == 0)
+		gfs2_unlink_di(inode);
+	return 0;
+}
+
+
+/**
+ * gfs2_unlink - Unlink an inode (this does rmdir as well)
+ * @dir: The inode of the directory containing the inode to unlink
+ * @dentry: The file itself
+ *
+ * This routine uses the type of the inode as a flag to figure out
+ * whether this is an unlink or an rmdir.
+ *
+ * Returns: errno
+ */
+
+static int gfs2_unlink(struct inode *dir, struct dentry *dentry)
+{
+	struct gfs2_inode *dip = GFS2_I(dir);
+	struct gfs2_sbd *sdp = GFS2_SB(dir);
+	struct inode *inode = dentry->d_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct buffer_head *bh;
+	struct gfs2_holder ghs[3];
+	struct gfs2_rgrpd *rgd;
+	struct gfs2_holder ri_gh;
+	int error;
+
+	error = gfs2_rindex_hold(sdp, &ri_gh);
+	if (error)
+		return error;
+
+	gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
+	gfs2_holder_init(ip->i_gl,  LM_ST_EXCLUSIVE, 0, ghs + 1);
+
+	rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr);
+	gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2);
+
+
+	error = gfs2_glock_nq(ghs); /* parent */
+	if (error)
+		goto out_parent;
+
+	error = gfs2_glock_nq(ghs + 1); /* child */
+	if (error)
+		goto out_child;
+
+	error = -ENOENT;
+	if (inode->i_nlink == 0)
+		goto out_rgrp;
+
+	if (S_ISDIR(inode->i_mode)) {
+		error = -ENOTEMPTY;
+		if (ip->i_entries > 2 || inode->i_nlink > 2)
+			goto out_rgrp;
+	}
+
+	error = gfs2_glock_nq(ghs + 2); /* rgrp */
+	if (error)
+		goto out_rgrp;
+
+	error = gfs2_unlink_ok(dip, &dentry->d_name, ip);
+	if (error)
+		goto out_gunlock;
+
+	error = gfs2_trans_begin(sdp, 2*RES_DINODE + 3*RES_LEAF + RES_RG_BIT, 0);
+	if (error)
+		goto out_gunlock;
+
+	error = gfs2_meta_inode_buffer(ip, &bh);
+	if (error)
+		goto out_end_trans;
+
+	error = gfs2_unlink_inode(dip, dentry, bh);
+	brelse(bh);
+
+out_end_trans:
+	gfs2_trans_end(sdp);
+out_gunlock:
+	gfs2_glock_dq(ghs + 2);
+out_rgrp:
+	gfs2_holder_uninit(ghs + 2);
+	gfs2_glock_dq(ghs + 1);
+out_child:
+	gfs2_holder_uninit(ghs + 1);
+	gfs2_glock_dq(ghs);
+out_parent:
+	gfs2_holder_uninit(ghs);
+	gfs2_glock_dq_uninit(&ri_gh);
+	return error;
+}
+
+/**
+ * gfs2_symlink - Create a symlink
+ * @dir: The directory to create the symlink in
+ * @dentry: The dentry to put the symlink in
+ * @symname: The thing which the link points to
+ *
+ * Returns: errno
+ */
+
+static int gfs2_symlink(struct inode *dir, struct dentry *dentry,
+			const char *symname)
+{
+	struct gfs2_sbd *sdp = GFS2_SB(dir);
+	unsigned int size;
+
+	size = strlen(symname);
+	if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode) - 1)
+		return -ENAMETOOLONG;
+
+	return gfs2_create_inode(dir, dentry, S_IFLNK | S_IRWXUGO, 0, symname, size);
+}
+
+/**
+ * gfs2_mkdir - Make a directory
+ * @dir: The parent directory of the new one
+ * @dentry: The dentry of the new directory
+ * @mode: The mode of the new directory
+ *
+ * Returns: errno
+ */
+
+static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+{
+	return gfs2_create_inode(dir, dentry, S_IFDIR | mode, 0, NULL, 0);
+}
+
+/**
+ * gfs2_mknod - Make a special file
+ * @dir: The directory in which the special file will reside
+ * @dentry: The dentry of the special file
+ * @mode: The mode of the special file
+ * @dev: The device specification of the special file
+ *
+ */
+
+static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode,
+		      dev_t dev)
+{
+	return gfs2_create_inode(dir, dentry, mode, dev, NULL, 0);
+}
+
+/*
+ * gfs2_ok_to_move - check if it's ok to move a directory to another directory
+ * @this: move this
+ * @to: to here
+ *
+ * Follow @to back to the root and make sure we don't encounter @this
+ * Assumes we already hold the rename lock.
+ *
+ * Returns: errno
+ */
+
+static int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to)
+{
+	struct inode *dir = &to->i_inode;
+	struct super_block *sb = dir->i_sb;
+	struct inode *tmp;
+	int error = 0;
+
+	igrab(dir);
+
+	for (;;) {
+		if (dir == &this->i_inode) {
+			error = -EINVAL;
+			break;
+		}
+		if (dir == sb->s_root->d_inode) {
+			error = 0;
+			break;
+		}
+
+		tmp = gfs2_lookupi(dir, &gfs2_qdotdot, 1);
+		if (IS_ERR(tmp)) {
+			error = PTR_ERR(tmp);
+			break;
+		}
+
+		iput(dir);
+		dir = tmp;
+	}
+
+	iput(dir);
+
+	return error;
+}
+
+/**
+ * gfs2_rename - Rename a file
+ * @odir: Parent directory of old file name
+ * @odentry: The old dentry of the file
+ * @ndir: Parent directory of new file name
+ * @ndentry: The new dentry of the file
+ *
+ * Returns: errno
+ */
+
+static int gfs2_rename(struct inode *odir, struct dentry *odentry,
+		       struct inode *ndir, struct dentry *ndentry)
+{
+	struct gfs2_inode *odip = GFS2_I(odir);
+	struct gfs2_inode *ndip = GFS2_I(ndir);
+	struct gfs2_inode *ip = GFS2_I(odentry->d_inode);
+	struct gfs2_inode *nip = NULL;
+	struct gfs2_sbd *sdp = GFS2_SB(odir);
+	struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }, ri_gh;
+	struct gfs2_rgrpd *nrgd;
+	unsigned int num_gh;
+	int dir_rename = 0;
+	int alloc_required = 0;
+	unsigned int x;
+	int error;
+
+	if (ndentry->d_inode) {
+		nip = GFS2_I(ndentry->d_inode);
+		if (ip == nip)
+			return 0;
+	}
+
+	error = gfs2_rindex_hold(sdp, &ri_gh);
+	if (error)
+		return error;
+
+	if (odip != ndip) {
+		error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE,
+					   0, &r_gh);
+		if (error)
+			goto out;
+
+		if (S_ISDIR(ip->i_inode.i_mode)) {
+			dir_rename = 1;
+			/* don't move a dirctory into it's subdir */
+			error = gfs2_ok_to_move(ip, ndip);
+			if (error)
+				goto out_gunlock_r;
+		}
+	}
+
+	num_gh = 1;
+	gfs2_holder_init(odip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
+	if (odip != ndip) {
+		gfs2_holder_init(ndip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh);
+		num_gh++;
+	}
+	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh);
+	num_gh++;
+
+	if (nip) {
+		gfs2_holder_init(nip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh);
+		num_gh++;
+		/* grab the resource lock for unlink flag twiddling 
+		 * this is the case of the target file already existing
+		 * so we unlink before doing the rename
+		 */
+		nrgd = gfs2_blk2rgrpd(sdp, nip->i_no_addr);
+		if (nrgd)
+			gfs2_holder_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh++);
+	}
+
+	for (x = 0; x < num_gh; x++) {
+		error = gfs2_glock_nq(ghs + x);
+		if (error)
+			goto out_gunlock;
+	}
+
+	error = -ENOENT;
+	if (ip->i_inode.i_nlink == 0)
+		goto out_gunlock;
+
+	/* Check out the old directory */
+
+	error = gfs2_unlink_ok(odip, &odentry->d_name, ip);
+	if (error)
+		goto out_gunlock;
+
+	/* Check out the new directory */
+
+	if (nip) {
+		error = gfs2_unlink_ok(ndip, &ndentry->d_name, nip);
+		if (error)
+			goto out_gunlock;
+
+		if (nip->i_inode.i_nlink == 0) {
+			error = -EAGAIN;
+			goto out_gunlock;
+		}
+
+		if (S_ISDIR(nip->i_inode.i_mode)) {
+			if (nip->i_entries < 2) {
+				gfs2_consist_inode(nip);
+				error = -EIO;
+				goto out_gunlock;
+			}
+			if (nip->i_entries > 2) {
+				error = -ENOTEMPTY;
+				goto out_gunlock;
+			}
+		}
+	} else {
+		error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC, 0);
+		if (error)
+			goto out_gunlock;
+
+		error = gfs2_dir_check(ndir, &ndentry->d_name, NULL);
+		switch (error) {
+		case -ENOENT:
+			error = 0;
+			break;
+		case 0:
+			error = -EEXIST;
+		default:
+			goto out_gunlock;
+		};
+
+		if (odip != ndip) {
+			if (!ndip->i_inode.i_nlink) {
+				error = -ENOENT;
+				goto out_gunlock;
+			}
+			if (ndip->i_entries == (u32)-1) {
+				error = -EFBIG;
+				goto out_gunlock;
+			}
+			if (S_ISDIR(ip->i_inode.i_mode) &&
+			    ndip->i_inode.i_nlink == (u32)-1) {
+				error = -EMLINK;
+				goto out_gunlock;
+			}
+		}
+	}
+
+	/* Check out the dir to be renamed */
+
+	if (dir_rename) {
+		error = gfs2_permission(odentry->d_inode, MAY_WRITE, 0);
+		if (error)
+			goto out_gunlock;
+	}
+
+	if (nip == NULL)
+		alloc_required = gfs2_diradd_alloc_required(ndir, &ndentry->d_name);
+	error = alloc_required;
+	if (error < 0)
+		goto out_gunlock;
+	error = 0;
+
+	if (alloc_required) {
+		struct gfs2_alloc *al = gfs2_alloc_get(ndip);
+		if (!al) {
+			error = -ENOMEM;
+			goto out_gunlock;
+		}
+
+		error = gfs2_quota_lock_check(ndip);
+		if (error)
+			goto out_alloc;
+
+		al->al_requested = sdp->sd_max_dirres;
+
+		error = gfs2_inplace_reserve_ri(ndip);
+		if (error)
+			goto out_gunlock_q;
+
+		error = gfs2_trans_begin(sdp, sdp->sd_max_dirres +
+					 gfs2_rg_blocks(al) +
+					 4 * RES_DINODE + 4 * RES_LEAF +
+					 RES_STATFS + RES_QUOTA + 4, 0);
+		if (error)
+			goto out_ipreserv;
+	} else {
+		error = gfs2_trans_begin(sdp, 4 * RES_DINODE +
+					 5 * RES_LEAF + 4, 0);
+		if (error)
+			goto out_gunlock;
+	}
+
+	/* Remove the target file, if it exists */
+
+	if (nip) {
+		struct buffer_head *bh;
+		error = gfs2_meta_inode_buffer(nip, &bh);
+		if (error)
+			goto out_end_trans;
+		error = gfs2_unlink_inode(ndip, ndentry, bh);
+		brelse(bh);
+	}
+
+	if (dir_rename) {
+		error = gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR);
+		if (error)
+			goto out_end_trans;
+	} else {
+		struct buffer_head *dibh;
+		error = gfs2_meta_inode_buffer(ip, &dibh);
+		if (error)
+			goto out_end_trans;
+		ip->i_inode.i_ctime = CURRENT_TIME;
+		gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+		gfs2_dinode_out(ip, dibh->b_data);
+		brelse(dibh);
+	}
+
+	error = gfs2_dir_del(odip, odentry);
+	if (error)
+		goto out_end_trans;
+
+	error = gfs2_dir_add(ndir, &ndentry->d_name, ip);
+	if (error)
+		goto out_end_trans;
+
+out_end_trans:
+	gfs2_trans_end(sdp);
+out_ipreserv:
+	if (alloc_required)
+		gfs2_inplace_release(ndip);
+out_gunlock_q:
+	if (alloc_required)
+		gfs2_quota_unlock(ndip);
+out_alloc:
+	if (alloc_required)
+		gfs2_alloc_put(ndip);
+out_gunlock:
+	while (x--) {
+		gfs2_glock_dq(ghs + x);
+		gfs2_holder_uninit(ghs + x);
+	}
+out_gunlock_r:
+	if (r_gh.gh_gl)
+		gfs2_glock_dq_uninit(&r_gh);
+out:
+	gfs2_glock_dq_uninit(&ri_gh);
+	return error;
+}
+
+/**
+ * gfs2_follow_link - Follow a symbolic link
+ * @dentry: The dentry of the link
+ * @nd: Data that we pass to vfs_follow_link()
+ *
+ * This can handle symlinks of any size.
+ *
+ * Returns: 0 on success or error code
+ */
+
+static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd)
+{
+	struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
+	struct gfs2_holder i_gh;
+	struct buffer_head *dibh;
+	unsigned int size;
+	char *buf;
+	int error;
+
+	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh);
+	error = gfs2_glock_nq(&i_gh);
+	if (error) {
+		gfs2_holder_uninit(&i_gh);
+		nd_set_link(nd, ERR_PTR(error));
+		return NULL;
+	}
+
+	size = (unsigned int)i_size_read(&ip->i_inode);
+	if (size == 0) {
+		gfs2_consist_inode(ip);
+		buf = ERR_PTR(-EIO);
+		goto out;
+	}
+
+	error = gfs2_meta_inode_buffer(ip, &dibh);
+	if (error) {
+		buf = ERR_PTR(error);
+		goto out;
+	}
+
+	buf = kzalloc(size + 1, GFP_NOFS);
+	if (!buf)
+		buf = ERR_PTR(-ENOMEM);
+	else
+		memcpy(buf, dibh->b_data + sizeof(struct gfs2_dinode), size);
+	brelse(dibh);
+out:
+	gfs2_glock_dq_uninit(&i_gh);
+	nd_set_link(nd, buf);
+	return NULL;
+}
+
+static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p)
+{
+	char *s = nd_get_link(nd);
+	if (!IS_ERR(s))
+		kfree(s);
+}
+
+/**
+ * gfs2_permission -
+ * @inode: The inode
+ * @mask: The mask to be tested
+ * @flags: Indicates whether this is an RCU path walk or not
+ *
+ * This may be called from the VFS directly, or from within GFS2 with the
+ * inode locked, so we look to see if the glock is already locked and only
+ * lock the glock if its not already been done.
+ *
+ * Returns: errno
+ */
+
+int gfs2_permission(struct inode *inode, int mask, unsigned int flags)
+{
+	struct gfs2_inode *ip;
+	struct gfs2_holder i_gh;
+	int error;
+	int unlock = 0;
+
+
+	ip = GFS2_I(inode);
+	if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) {
+		if (flags & IPERM_FLAG_RCU)
+			return -ECHILD;
+		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
+		if (error)
+			return error;
+		unlock = 1;
+	}
+
+	if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode))
+		error = -EACCES;
+	else
+		error = generic_permission(inode, mask, flags, gfs2_check_acl);
+	if (unlock)
+		gfs2_glock_dq_uninit(&i_gh);
+
+	return error;
 }
 
 static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr)
@@ -902,8 +1594,6 @@ static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr)
  * @ip:
  * @attr:
  *
- * Called with a reference on the vnode.
- *
  * Returns: errno
  */
 
@@ -923,60 +1613,280 @@ int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr)
 	return error;
 }
 
-void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf)
-{
-	struct gfs2_dinode *str = buf;
-
-	str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
-	str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI);
-	str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI);
-	str->di_num.no_addr = cpu_to_be64(ip->i_no_addr);
-	str->di_num.no_formal_ino = cpu_to_be64(ip->i_no_formal_ino);
-	str->di_mode = cpu_to_be32(ip->i_inode.i_mode);
-	str->di_uid = cpu_to_be32(ip->i_inode.i_uid);
-	str->di_gid = cpu_to_be32(ip->i_inode.i_gid);
-	str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink);
-	str->di_size = cpu_to_be64(i_size_read(&ip->i_inode));
-	str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode));
-	str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec);
-	str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec);
-	str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec);
-
-	str->di_goal_meta = cpu_to_be64(ip->i_goal);
-	str->di_goal_data = cpu_to_be64(ip->i_goal);
-	str->di_generation = cpu_to_be64(ip->i_generation);
-
-	str->di_flags = cpu_to_be32(ip->i_diskflags);
-	str->di_height = cpu_to_be16(ip->i_height);
-	str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) &&
-					     !(ip->i_diskflags & GFS2_DIF_EXHASH) ?
-					     GFS2_FORMAT_DE : 0);
-	str->di_depth = cpu_to_be16(ip->i_depth);
-	str->di_entries = cpu_to_be32(ip->i_entries);
-
-	str->di_eattr = cpu_to_be64(ip->i_eattr);
-	str->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec);
-	str->di_mtime_nsec = cpu_to_be32(ip->i_inode.i_mtime.tv_nsec);
-	str->di_ctime_nsec = cpu_to_be32(ip->i_inode.i_ctime.tv_nsec);
-}
-
-void gfs2_dinode_print(const struct gfs2_inode *ip)
-{
-	printk(KERN_INFO "  no_formal_ino = %llu\n",
-	       (unsigned long long)ip->i_no_formal_ino);
-	printk(KERN_INFO "  no_addr = %llu\n",
-	       (unsigned long long)ip->i_no_addr);
-	printk(KERN_INFO "  i_size = %llu\n",
-	       (unsigned long long)i_size_read(&ip->i_inode));
-	printk(KERN_INFO "  blocks = %llu\n",
-	       (unsigned long long)gfs2_get_inode_blocks(&ip->i_inode));
-	printk(KERN_INFO "  i_goal = %llu\n",
-	       (unsigned long long)ip->i_goal);
-	printk(KERN_INFO "  i_diskflags = 0x%.8X\n", ip->i_diskflags);
-	printk(KERN_INFO "  i_height = %u\n", ip->i_height);
-	printk(KERN_INFO "  i_depth = %u\n", ip->i_depth);
-	printk(KERN_INFO "  i_entries = %u\n", ip->i_entries);
-	printk(KERN_INFO "  i_eattr = %llu\n",
-	       (unsigned long long)ip->i_eattr);
+static int setattr_chown(struct inode *inode, struct iattr *attr)
+{
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
+	u32 ouid, ogid, nuid, ngid;
+	int error;
+
+	ouid = inode->i_uid;
+	ogid = inode->i_gid;
+	nuid = attr->ia_uid;
+	ngid = attr->ia_gid;
+
+	if (!(attr->ia_valid & ATTR_UID) || ouid == nuid)
+		ouid = nuid = NO_QUOTA_CHANGE;
+	if (!(attr->ia_valid & ATTR_GID) || ogid == ngid)
+		ogid = ngid = NO_QUOTA_CHANGE;
+
+	if (!gfs2_alloc_get(ip))
+		return -ENOMEM;
+
+	error = gfs2_quota_lock(ip, nuid, ngid);
+	if (error)
+		goto out_alloc;
+
+	if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) {
+		error = gfs2_quota_check(ip, nuid, ngid);
+		if (error)
+			goto out_gunlock_q;
+	}
+
+	error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_QUOTA, 0);
+	if (error)
+		goto out_gunlock_q;
+
+	error = gfs2_setattr_simple(ip, attr);
+	if (error)
+		goto out_end_trans;
+
+	if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) {
+		u64 blocks = gfs2_get_inode_blocks(&ip->i_inode);
+		gfs2_quota_change(ip, -blocks, ouid, ogid);
+		gfs2_quota_change(ip, blocks, nuid, ngid);
+	}
+
+out_end_trans:
+	gfs2_trans_end(sdp);
+out_gunlock_q:
+	gfs2_quota_unlock(ip);
+out_alloc:
+	gfs2_alloc_put(ip);
+	return error;
 }
 
+/**
+ * gfs2_setattr - Change attributes on an inode
+ * @dentry: The dentry which is changing
+ * @attr: The structure describing the change
+ *
+ * The VFS layer wants to change one or more of an inodes attributes.  Write
+ * that change out to disk.
+ *
+ * Returns: errno
+ */
+
+static int gfs2_setattr(struct dentry *dentry, struct iattr *attr)
+{
+	struct inode *inode = dentry->d_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder i_gh;
+	int error;
+
+	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh);
+	if (error)
+		return error;
+
+	error = -EPERM;
+	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
+		goto out;
+
+	error = inode_change_ok(inode, attr);
+	if (error)
+		goto out;
+
+	if (attr->ia_valid & ATTR_SIZE)
+		error = gfs2_setattr_size(inode, attr->ia_size);
+	else if (attr->ia_valid & (ATTR_UID | ATTR_GID))
+		error = setattr_chown(inode, attr);
+	else if ((attr->ia_valid & ATTR_MODE) && IS_POSIXACL(inode))
+		error = gfs2_acl_chmod(ip, attr);
+	else
+		error = gfs2_setattr_simple(ip, attr);
+
+out:
+	gfs2_glock_dq_uninit(&i_gh);
+	if (!error)
+		mark_inode_dirty(inode);
+	return error;
+}
+
+/**
+ * gfs2_getattr - Read out an inode's attributes
+ * @mnt: The vfsmount the inode is being accessed from
+ * @dentry: The dentry to stat
+ * @stat: The inode's stats
+ *
+ * This may be called from the VFS directly, or from within GFS2 with the
+ * inode locked, so we look to see if the glock is already locked and only
+ * lock the glock if its not already been done. Note that its the NFS
+ * readdirplus operation which causes this to be called (from filldir)
+ * with the glock already held.
+ *
+ * Returns: errno
+ */
+
+static int gfs2_getattr(struct vfsmount *mnt, struct dentry *dentry,
+			struct kstat *stat)
+{
+	struct inode *inode = dentry->d_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	int error;
+	int unlock = 0;
+
+	if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) {
+		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
+		if (error)
+			return error;
+		unlock = 1;
+	}
+
+	generic_fillattr(inode, stat);
+	if (unlock)
+		gfs2_glock_dq_uninit(&gh);
+
+	return 0;
+}
+
+static int gfs2_setxattr(struct dentry *dentry, const char *name,
+			 const void *data, size_t size, int flags)
+{
+	struct inode *inode = dentry->d_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	int ret;
+
+	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+	ret = gfs2_glock_nq(&gh);
+	if (ret == 0) {
+		ret = generic_setxattr(dentry, name, data, size, flags);
+		gfs2_glock_dq(&gh);
+	}
+	gfs2_holder_uninit(&gh);
+	return ret;
+}
+
+static ssize_t gfs2_getxattr(struct dentry *dentry, const char *name,
+			     void *data, size_t size)
+{
+	struct inode *inode = dentry->d_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	int ret;
+
+	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
+	ret = gfs2_glock_nq(&gh);
+	if (ret == 0) {
+		ret = generic_getxattr(dentry, name, data, size);
+		gfs2_glock_dq(&gh);
+	}
+	gfs2_holder_uninit(&gh);
+	return ret;
+}
+
+static int gfs2_removexattr(struct dentry *dentry, const char *name)
+{
+	struct inode *inode = dentry->d_inode;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	int ret;
+
+	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+	ret = gfs2_glock_nq(&gh);
+	if (ret == 0) {
+		ret = generic_removexattr(dentry, name);
+		gfs2_glock_dq(&gh);
+	}
+	gfs2_holder_uninit(&gh);
+	return ret;
+}
+
+static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+		       u64 start, u64 len)
+{
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_holder gh;
+	int ret;
+
+	ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC);
+	if (ret)
+		return ret;
+
+	mutex_lock(&inode->i_mutex);
+
+	ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
+	if (ret)
+		goto out;
+
+	if (gfs2_is_stuffed(ip)) {
+		u64 phys = ip->i_no_addr << inode->i_blkbits;
+		u64 size = i_size_read(inode);
+		u32 flags = FIEMAP_EXTENT_LAST|FIEMAP_EXTENT_NOT_ALIGNED|
+			    FIEMAP_EXTENT_DATA_INLINE;
+		phys += sizeof(struct gfs2_dinode);
+		phys += start;
+		if (start + len > size)
+			len = size - start;
+		if (start < size)
+			ret = fiemap_fill_next_extent(fieinfo, start, phys,
+						      len, flags);
+		if (ret == 1)
+			ret = 0;
+	} else {
+		ret = __generic_block_fiemap(inode, fieinfo, start, len,
+					     gfs2_block_map);
+	}
+
+	gfs2_glock_dq_uninit(&gh);
+out:
+	mutex_unlock(&inode->i_mutex);
+	return ret;
+}
+
+const struct inode_operations gfs2_file_iops = {
+	.permission = gfs2_permission,
+	.setattr = gfs2_setattr,
+	.getattr = gfs2_getattr,
+	.setxattr = gfs2_setxattr,
+	.getxattr = gfs2_getxattr,
+	.listxattr = gfs2_listxattr,
+	.removexattr = gfs2_removexattr,
+	.fiemap = gfs2_fiemap,
+};
+
+const struct inode_operations gfs2_dir_iops = {
+	.create = gfs2_create,
+	.lookup = gfs2_lookup,
+	.link = gfs2_link,
+	.unlink = gfs2_unlink,
+	.symlink = gfs2_symlink,
+	.mkdir = gfs2_mkdir,
+	.rmdir = gfs2_unlink,
+	.mknod = gfs2_mknod,
+	.rename = gfs2_rename,
+	.permission = gfs2_permission,
+	.setattr = gfs2_setattr,
+	.getattr = gfs2_getattr,
+	.setxattr = gfs2_setxattr,
+	.getxattr = gfs2_getxattr,
+	.listxattr = gfs2_listxattr,
+	.removexattr = gfs2_removexattr,
+	.fiemap = gfs2_fiemap,
+};
+
+const struct inode_operations gfs2_symlink_iops = {
+	.readlink = generic_readlink,
+	.follow_link = gfs2_follow_link,
+	.put_link = gfs2_put_link,
+	.permission = gfs2_permission,
+	.setattr = gfs2_setattr,
+	.getattr = gfs2_getattr,
+	.setxattr = gfs2_setxattr,
+	.getxattr = gfs2_getxattr,
+	.listxattr = gfs2_listxattr,
+	.removexattr = gfs2_removexattr,
+	.fiemap = gfs2_fiemap,
+};
+
diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h
index 3e00a66e7cb..31606076f70 100644
--- a/fs/gfs2/inode.h
+++ b/fs/gfs2/inode.h
@@ -97,26 +97,21 @@ err:
 }
 
 extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, 
-				       u64 no_addr, u64 no_formal_ino);
+				       u64 no_addr, u64 no_formal_ino,
+				       int non_block);
 extern struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
 					 u64 *no_formal_ino,
 					 unsigned int blktype);
-extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr);
+extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr, int nonblock);
 
 extern int gfs2_inode_refresh(struct gfs2_inode *ip);
 
-extern int gfs2_dinode_dealloc(struct gfs2_inode *inode);
-extern int gfs2_change_nlink(struct gfs2_inode *ip, int diff);
 extern struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name,
 				  int is_root);
-extern struct inode *gfs2_createi(struct gfs2_holder *ghs,
-				  const struct qstr *name,
-				  unsigned int mode, dev_t dev);
 extern int gfs2_permission(struct inode *inode, int mask, unsigned int flags);
 extern int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr);
 extern struct inode *gfs2_lookup_simple(struct inode *dip, const char *name);
 extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf);
-extern void gfs2_dinode_print(const struct gfs2_inode *ip);
 
 extern const struct inode_operations gfs2_file_iops;
 extern const struct inode_operations gfs2_dir_iops;
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index 5b102c1887f..cec26c00b50 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -18,6 +18,7 @@
 #include <linux/kthread.h>
 #include <linux/freezer.h>
 #include <linux/bio.h>
+#include <linux/writeback.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -83,55 +84,97 @@ void gfs2_remove_from_ail(struct gfs2_bufdata *bd)
 /**
  * gfs2_ail1_start_one - Start I/O on a part of the AIL
  * @sdp: the filesystem
- * @tr: the part of the AIL
+ * @wbc: The writeback control structure
+ * @ai: The ail structure
  *
  */
 
-static void gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
+static int gfs2_ail1_start_one(struct gfs2_sbd *sdp,
+			       struct writeback_control *wbc,
+			       struct gfs2_ail *ai)
 __releases(&sdp->sd_ail_lock)
 __acquires(&sdp->sd_ail_lock)
 {
+	struct gfs2_glock *gl = NULL;
+	struct address_space *mapping;
 	struct gfs2_bufdata *bd, *s;
 	struct buffer_head *bh;
-	int retry;
 
-	do {
-		retry = 0;
+	list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list, bd_ail_st_list) {
+		bh = bd->bd_bh;
 
-		list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list,
-						 bd_ail_st_list) {
-			bh = bd->bd_bh;
+		gfs2_assert(sdp, bd->bd_ail == ai);
 
-			gfs2_assert(sdp, bd->bd_ail == ai);
+		if (!buffer_busy(bh)) {
+			if (!buffer_uptodate(bh))
+				gfs2_io_error_bh(sdp, bh);
+			list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list);
+			continue;
+		}
 
-			if (!buffer_busy(bh)) {
-				if (!buffer_uptodate(bh))
-					gfs2_io_error_bh(sdp, bh);
-				list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list);
-				continue;
-			}
+		if (!buffer_dirty(bh))
+			continue;
+		if (gl == bd->bd_gl)
+			continue;
+		gl = bd->bd_gl;
+		list_move(&bd->bd_ail_st_list, &ai->ai_ail1_list);
+		mapping = bh->b_page->mapping;
+		if (!mapping)
+			continue;
+		spin_unlock(&sdp->sd_ail_lock);
+		generic_writepages(mapping, wbc);
+		spin_lock(&sdp->sd_ail_lock);
+		if (wbc->nr_to_write <= 0)
+			break;
+		return 1;
+	}
 
-			if (!buffer_dirty(bh))
-				continue;
+	return 0;
+}
 
-			list_move(&bd->bd_ail_st_list, &ai->ai_ail1_list);
 
-			get_bh(bh);
-			spin_unlock(&sdp->sd_ail_lock);
-			lock_buffer(bh);
-			if (test_clear_buffer_dirty(bh)) {
-				bh->b_end_io = end_buffer_write_sync;
-				submit_bh(WRITE_SYNC, bh);
-			} else {
-				unlock_buffer(bh);
-				brelse(bh);
-			}
-			spin_lock(&sdp->sd_ail_lock);
-
-			retry = 1;
+/**
+ * gfs2_ail1_flush - start writeback of some ail1 entries 
+ * @sdp: The super block
+ * @wbc: The writeback control structure
+ *
+ * Writes back some ail1 entries, according to the limits in the
+ * writeback control structure
+ */
+
+void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc)
+{
+	struct list_head *head = &sdp->sd_ail1_list;
+	struct gfs2_ail *ai;
+
+	trace_gfs2_ail_flush(sdp, wbc, 1);
+	spin_lock(&sdp->sd_ail_lock);
+restart:
+	list_for_each_entry_reverse(ai, head, ai_list) {
+		if (wbc->nr_to_write <= 0)
 			break;
-		}
-	} while (retry);
+		if (gfs2_ail1_start_one(sdp, wbc, ai))
+			goto restart;
+	}
+	spin_unlock(&sdp->sd_ail_lock);
+	trace_gfs2_ail_flush(sdp, wbc, 0);
+}
+
+/**
+ * gfs2_ail1_start - start writeback of all ail1 entries
+ * @sdp: The superblock
+ */
+
+static void gfs2_ail1_start(struct gfs2_sbd *sdp)
+{
+	struct writeback_control wbc = {
+		.sync_mode = WB_SYNC_NONE,
+		.nr_to_write = LONG_MAX,
+		.range_start = 0,
+		.range_end = LLONG_MAX,
+	};
+
+	return gfs2_ail1_flush(sdp, &wbc);
 }
 
 /**
@@ -141,7 +184,7 @@ __acquires(&sdp->sd_ail_lock)
  *
  */
 
-static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int flags)
+static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
 {
 	struct gfs2_bufdata *bd, *s;
 	struct buffer_head *bh;
@@ -149,71 +192,37 @@ static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int fl
 	list_for_each_entry_safe_reverse(bd, s, &ai->ai_ail1_list,
 					 bd_ail_st_list) {
 		bh = bd->bd_bh;
-
 		gfs2_assert(sdp, bd->bd_ail == ai);
-
-		if (buffer_busy(bh)) {
-			if (flags & DIO_ALL)
-				continue;
-			else
-				break;
-		}
-
+		if (buffer_busy(bh))
+			continue;
 		if (!buffer_uptodate(bh))
 			gfs2_io_error_bh(sdp, bh);
-
 		list_move(&bd->bd_ail_st_list, &ai->ai_ail2_list);
 	}
 
-	return list_empty(&ai->ai_ail1_list);
 }
 
-static void gfs2_ail1_start(struct gfs2_sbd *sdp)
-{
-	struct list_head *head;
-	u64 sync_gen;
-	struct gfs2_ail *ai;
-	int done = 0;
-
-	spin_lock(&sdp->sd_ail_lock);
-	head = &sdp->sd_ail1_list;
-	if (list_empty(head)) {
-		spin_unlock(&sdp->sd_ail_lock);
-		return;
-	}
-	sync_gen = sdp->sd_ail_sync_gen++;
-
-	while(!done) {
-		done = 1;
-		list_for_each_entry_reverse(ai, head, ai_list) {
-			if (ai->ai_sync_gen >= sync_gen)
-				continue;
-			ai->ai_sync_gen = sync_gen;
-			gfs2_ail1_start_one(sdp, ai); /* This may drop ail lock */
-			done = 0;
-			break;
-		}
-	}
-
-	spin_unlock(&sdp->sd_ail_lock);
-}
+/**
+ * gfs2_ail1_empty - Try to empty the ail1 lists
+ * @sdp: The superblock
+ *
+ * Tries to empty the ail1 lists, starting with the oldest first
+ */
 
-static int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags)
+static int gfs2_ail1_empty(struct gfs2_sbd *sdp)
 {
 	struct gfs2_ail *ai, *s;
 	int ret;
 
 	spin_lock(&sdp->sd_ail_lock);
-
 	list_for_each_entry_safe_reverse(ai, s, &sdp->sd_ail1_list, ai_list) {
-		if (gfs2_ail1_empty_one(sdp, ai, flags))
+		gfs2_ail1_empty_one(sdp, ai);
+		if (list_empty(&ai->ai_ail1_list))
 			list_move(&ai->ai_list, &sdp->sd_ail2_list);
-		else if (!(flags & DIO_ALL))
+		else
 			break;
 	}
-
 	ret = list_empty(&sdp->sd_ail1_list);
-
 	spin_unlock(&sdp->sd_ail_lock);
 
 	return ret;
@@ -574,7 +583,7 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags, int pull)
 	set_buffer_uptodate(bh);
 	clear_buffer_dirty(bh);
 
-	gfs2_ail1_empty(sdp, 0);
+	gfs2_ail1_empty(sdp);
 	tail = current_tail(sdp);
 
 	lh = (struct gfs2_log_header *)bh->b_data;
@@ -869,7 +878,7 @@ void gfs2_meta_syncfs(struct gfs2_sbd *sdp)
 	gfs2_log_flush(sdp, NULL);
 	for (;;) {
 		gfs2_ail1_start(sdp);
-		if (gfs2_ail1_empty(sdp, DIO_ALL))
+		if (gfs2_ail1_empty(sdp))
 			break;
 		msleep(10);
 	}
@@ -905,17 +914,15 @@ int gfs2_logd(void *data)
 
 		preflush = atomic_read(&sdp->sd_log_pinned);
 		if (gfs2_jrnl_flush_reqd(sdp) || t == 0) {
-			gfs2_ail1_empty(sdp, DIO_ALL);
+			gfs2_ail1_empty(sdp);
 			gfs2_log_flush(sdp, NULL);
-			gfs2_ail1_empty(sdp, DIO_ALL);
 		}
 
 		if (gfs2_ail_flush_reqd(sdp)) {
 			gfs2_ail1_start(sdp);
 			io_schedule();
-			gfs2_ail1_empty(sdp, 0);
+			gfs2_ail1_empty(sdp);
 			gfs2_log_flush(sdp, NULL);
-			gfs2_ail1_empty(sdp, DIO_ALL);
 		}
 
 		wake_up(&sdp->sd_log_waitq);
diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h
index 0d007f92023..ab0621698b7 100644
--- a/fs/gfs2/log.h
+++ b/fs/gfs2/log.h
@@ -12,6 +12,7 @@
 
 #include <linux/list.h>
 #include <linux/spinlock.h>
+#include <linux/writeback.h>
 #include "incore.h"
 
 /**
@@ -59,6 +60,7 @@ extern struct buffer_head *gfs2_log_fake_buf(struct gfs2_sbd *sdp,
 extern void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl);
 extern void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans);
 extern void gfs2_remove_from_ail(struct gfs2_bufdata *bd);
+extern void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc);
 
 extern void gfs2_log_shutdown(struct gfs2_sbd *sdp);
 extern void gfs2_meta_syncfs(struct gfs2_sbd *sdp);
diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c
index 51d27f00ebb..05bbb124699 100644
--- a/fs/gfs2/lops.c
+++ b/fs/gfs2/lops.c
@@ -40,7 +40,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh)
 {
 	struct gfs2_bufdata *bd;
 
-	gfs2_assert_withdraw(sdp, test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags));
+	BUG_ON(!current->journal_info);
 
 	clear_buffer_dirty(bh);
 	if (test_set_buffer_pinned(bh))
@@ -65,6 +65,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh)
  * @sdp: the filesystem the buffer belongs to
  * @bh: The buffer to unpin
  * @ai:
+ * @flags: The inode dirty flags
  *
  */
 
@@ -73,10 +74,8 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh,
 {
 	struct gfs2_bufdata *bd = bh->b_private;
 
-	gfs2_assert_withdraw(sdp, buffer_uptodate(bh));
-
-	if (!buffer_pinned(bh))
-		gfs2_assert_withdraw(sdp, 0);
+	BUG_ON(!buffer_uptodate(bh));
+	BUG_ON(!buffer_pinned(bh));
 
 	lock_buffer(bh);
 	mark_buffer_dirty(bh);
@@ -95,8 +94,7 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh,
 	list_add(&bd->bd_ail_st_list, &ai->ai_ail1_list);
 	spin_unlock(&sdp->sd_ail_lock);
 
-	if (test_and_clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags))
-		gfs2_glock_schedule_for_reclaim(bd->bd_gl);
+	clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags);
 	trace_gfs2_pin(bd, 0);
 	unlock_buffer(bh);
 	atomic_dec(&sdp->sd_log_pinned);
@@ -322,12 +320,16 @@ static void buf_lo_after_scan(struct gfs2_jdesc *jd, int error, int pass)
 
 static void revoke_lo_add(struct gfs2_sbd *sdp, struct gfs2_log_element *le)
 {
+	struct gfs2_bufdata *bd = container_of(le, struct gfs2_bufdata, bd_le);
+	struct gfs2_glock *gl = bd->bd_gl;
 	struct gfs2_trans *tr;
 
 	tr = current->journal_info;
 	tr->tr_touched = 1;
 	tr->tr_num_revoke++;
 	sdp->sd_log_num_revoke++;
+	atomic_inc(&gl->gl_revokes);
+	set_bit(GLF_LFLUSH, &gl->gl_flags);
 	list_add(&le->le_list, &sdp->sd_log_le_revoke);
 }
 
@@ -350,9 +352,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp)
 	ld->ld_data1 = cpu_to_be32(sdp->sd_log_num_revoke);
 	offset = sizeof(struct gfs2_log_descriptor);
 
-	while (!list_empty(head)) {
-		bd = list_entry(head->next, struct gfs2_bufdata, bd_le.le_list);
-		list_del_init(&bd->bd_le.le_list);
+	list_for_each_entry(bd, head, bd_le.le_list) {
 		sdp->sd_log_num_revoke--;
 
 		if (offset + sizeof(u64) > sdp->sd_sb.sb_bsize) {
@@ -367,8 +367,6 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp)
 		}
 
 		*(__be64 *)(bh->b_data + offset) = cpu_to_be64(bd->bd_blkno);
-		kmem_cache_free(gfs2_bufdata_cachep, bd);
-
 		offset += sizeof(u64);
 	}
 	gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke);
@@ -376,6 +374,22 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp)
 	submit_bh(WRITE_SYNC, bh);
 }
 
+static void revoke_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
+{
+	struct list_head *head = &sdp->sd_log_le_revoke;
+	struct gfs2_bufdata *bd;
+	struct gfs2_glock *gl;
+
+	while (!list_empty(head)) {
+		bd = list_entry(head->next, struct gfs2_bufdata, bd_le.le_list);
+		list_del_init(&bd->bd_le.le_list);
+		gl = bd->bd_gl;
+		atomic_dec(&gl->gl_revokes);
+		clear_bit(GLF_LFLUSH, &gl->gl_flags);
+		kmem_cache_free(gfs2_bufdata_cachep, bd);
+	}
+}
+
 static void revoke_lo_before_scan(struct gfs2_jdesc *jd,
 				  struct gfs2_log_header_host *head, int pass)
 {
@@ -749,6 +763,7 @@ const struct gfs2_log_operations gfs2_buf_lops = {
 const struct gfs2_log_operations gfs2_revoke_lops = {
 	.lo_add = revoke_lo_add,
 	.lo_before_commit = revoke_lo_before_commit,
+	.lo_after_commit = revoke_lo_after_commit,
 	.lo_before_scan = revoke_lo_before_scan,
 	.lo_scan_elements = revoke_lo_scan_elements,
 	.lo_after_scan = revoke_lo_after_scan,
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index 888a5f5a1a5..cfa327d3319 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -53,6 +53,7 @@ static void gfs2_init_glock_once(void *foo)
 	INIT_LIST_HEAD(&gl->gl_lru);
 	INIT_LIST_HEAD(&gl->gl_ail_list);
 	atomic_set(&gl->gl_ail_count, 0);
+	atomic_set(&gl->gl_revokes, 0);
 }
 
 static void gfs2_init_gl_aspace_once(void *foo)
diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c
index 675349b5a13..747238cd9f9 100644
--- a/fs/gfs2/meta_io.c
+++ b/fs/gfs2/meta_io.c
@@ -31,6 +31,7 @@
 #include "rgrp.h"
 #include "trans.h"
 #include "util.h"
+#include "trace_gfs2.h"
 
 static int gfs2_aspace_writepage(struct page *page, struct writeback_control *wbc)
 {
@@ -310,6 +311,7 @@ void gfs2_remove_from_journal(struct buffer_head *bh, struct gfs2_trans *tr, int
 	struct gfs2_bufdata *bd = bh->b_private;
 
 	if (test_clear_buffer_pinned(bh)) {
+		trace_gfs2_pin(bd, 0);
 		atomic_dec(&sdp->sd_log_pinned);
 		list_del_init(&bd->bd_le.le_list);
 		if (meta) {
diff --git a/fs/gfs2/meta_io.h b/fs/gfs2/meta_io.h
index 6a1d9ba1641..22c52659313 100644
--- a/fs/gfs2/meta_io.h
+++ b/fs/gfs2/meta_io.h
@@ -77,8 +77,6 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen);
 
 #define buffer_busy(bh) \
 ((bh)->b_state & ((1ul << BH_Dirty) | (1ul << BH_Lock) | (1ul << BH_Pinned)))
-#define buffer_in_io(bh) \
-((bh)->b_state & ((1ul << BH_Dirty) | (1ul << BH_Lock)))
 
 #endif /* __DIO_DOT_H__ */
 
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 42ef24355af..8ac9ae189b5 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -126,8 +126,10 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb)
  * changed.
  */
 
-static int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent)
+static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent)
 {
+	struct gfs2_sb_host *sb = &sdp->sd_sb;
+
 	if (sb->sb_magic != GFS2_MAGIC ||
 	    sb->sb_type != GFS2_METATYPE_SB) {
 		if (!silent)
@@ -157,8 +159,10 @@ static void end_bio_io_page(struct bio *bio, int error)
 	unlock_page(page);
 }
 
-static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf)
+static void gfs2_sb_in(struct gfs2_sbd *sdp, const void *buf)
 {
+	struct gfs2_sb_host *sb = &sdp->sd_sb;
+	struct super_block *s = sdp->sd_vfs;
 	const struct gfs2_sb *str = buf;
 
 	sb->sb_magic = be32_to_cpu(str->sb_header.mh_magic);
@@ -175,7 +179,7 @@ static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf)
 
 	memcpy(sb->sb_lockproto, str->sb_lockproto, GFS2_LOCKNAME_LEN);
 	memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN);
-	memcpy(sb->sb_uuid, str->sb_uuid, 16);
+	memcpy(s->s_uuid, str->sb_uuid, 16);
 }
 
 /**
@@ -197,7 +201,7 @@ static void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf)
  * Returns: 0 on success or error
  */
 
-static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector)
+static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector, int silent)
 {
 	struct super_block *sb = sdp->sd_vfs;
 	struct gfs2_sb *p;
@@ -227,10 +231,10 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector)
 		return -EIO;
 	}
 	p = kmap(page);
-	gfs2_sb_in(&sdp->sd_sb, p);
+	gfs2_sb_in(sdp, p);
 	kunmap(page);
 	__free_page(page);
-	return 0;
+	return gfs2_check_sb(sdp, silent);
 }
 
 /**
@@ -247,17 +251,13 @@ static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent)
 	unsigned int x;
 	int error;
 
-	error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift);
+	error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift, silent);
 	if (error) {
 		if (!silent)
 			fs_err(sdp, "can't read superblock\n");
 		return error;
 	}
 
-	error = gfs2_check_sb(sdp, &sdp->sd_sb, silent);
-	if (error)
-		return error;
-
 	sdp->sd_fsb2bb_shift = sdp->sd_sb.sb_bsize_shift -
 			       GFS2_BASIC_BLOCK_SHIFT;
 	sdp->sd_fsb2bb = 1 << sdp->sd_fsb2bb_shift;
@@ -340,14 +340,10 @@ static int init_names(struct gfs2_sbd *sdp, int silent)
 	/*  Try to autodetect  */
 
 	if (!proto[0] || !table[0]) {
-		error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift);
+		error = gfs2_read_super(sdp, GFS2_SB_ADDR >> sdp->sd_fsb2bb_shift, silent);
 		if (error)
 			return error;
 
-		error = gfs2_check_sb(sdp, &sdp->sd_sb, silent);
-		if (error)
-			goto out;
-
 		if (!proto[0])
 			proto = sdp->sd_sb.sb_lockproto;
 		if (!table[0])
@@ -364,7 +360,6 @@ static int init_names(struct gfs2_sbd *sdp, int silent)
 	while ((table = strchr(table, '/')))
 		*table = '_';
 
-out:
 	return error;
 }
 
@@ -430,7 +425,7 @@ static int gfs2_lookup_root(struct super_block *sb, struct dentry **dptr,
 	struct dentry *dentry;
 	struct inode *inode;
 
-	inode = gfs2_inode_lookup(sb, DT_DIR, no_addr, 0);
+	inode = gfs2_inode_lookup(sb, DT_DIR, no_addr, 0, 0);
 	if (IS_ERR(inode)) {
 		fs_err(sdp, "can't read in %s inode: %ld\n", name, PTR_ERR(inode));
 		return PTR_ERR(inode);
@@ -1119,8 +1114,7 @@ static int fill_super(struct super_block *sb, struct gfs2_args *args, int silent
 	if (sdp->sd_args.ar_statfs_quantum) {
 		sdp->sd_tune.gt_statfs_slow = 0;
 		sdp->sd_tune.gt_statfs_quantum = sdp->sd_args.ar_statfs_quantum;
-	}
-	else {
+	} else {
 		sdp->sd_tune.gt_statfs_slow = 1;
 		sdp->sd_tune.gt_statfs_quantum = 30;
 	}
diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
deleted file mode 100644
index 09e436a5072..00000000000
--- a/fs/gfs2/ops_inode.c
+++ /dev/null
@@ -1,1344 +0,0 @@
-/*
- * Copyright (C) Sistina Software, Inc.  1997-2003 All rights reserved.
- * Copyright (C) 2004-2006 Red Hat, Inc.  All rights reserved.
- *
- * This copyrighted material is made available to anyone wishing to use,
- * modify, copy, or redistribute it subject to the terms and conditions
- * of the GNU General Public License version 2.
- */
-
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/completion.h>
-#include <linux/buffer_head.h>
-#include <linux/namei.h>
-#include <linux/mm.h>
-#include <linux/xattr.h>
-#include <linux/posix_acl.h>
-#include <linux/gfs2_ondisk.h>
-#include <linux/crc32.h>
-#include <linux/fiemap.h>
-#include <asm/uaccess.h>
-
-#include "gfs2.h"
-#include "incore.h"
-#include "acl.h"
-#include "bmap.h"
-#include "dir.h"
-#include "xattr.h"
-#include "glock.h"
-#include "inode.h"
-#include "meta_io.h"
-#include "quota.h"
-#include "rgrp.h"
-#include "trans.h"
-#include "util.h"
-#include "super.h"
-
-/**
- * gfs2_create - Create a file
- * @dir: The directory in which to create the file
- * @dentry: The dentry of the new file
- * @mode: The mode of the new file
- *
- * Returns: errno
- */
-
-static int gfs2_create(struct inode *dir, struct dentry *dentry,
-		       int mode, struct nameidata *nd)
-{
-	struct gfs2_inode *dip = GFS2_I(dir);
-	struct gfs2_sbd *sdp = GFS2_SB(dir);
-	struct gfs2_holder ghs[2];
-	struct inode *inode;
-
-	gfs2_holder_init(dip->i_gl, 0, 0, ghs);
-
-	for (;;) {
-		inode = gfs2_createi(ghs, &dentry->d_name, S_IFREG | mode, 0);
-		if (!IS_ERR(inode)) {
-			gfs2_trans_end(sdp);
-			if (dip->i_alloc->al_rgd)
-				gfs2_inplace_release(dip);
-			gfs2_quota_unlock(dip);
-			gfs2_alloc_put(dip);
-			gfs2_glock_dq_uninit_m(2, ghs);
-			mark_inode_dirty(inode);
-			break;
-		} else if (PTR_ERR(inode) != -EEXIST ||
-			   (nd && nd->flags & LOOKUP_EXCL)) {
-			gfs2_holder_uninit(ghs);
-			return PTR_ERR(inode);
-		}
-
-		inode = gfs2_lookupi(dir, &dentry->d_name, 0);
-		if (inode) {
-			if (!IS_ERR(inode)) {
-				gfs2_holder_uninit(ghs);
-				break;
-			} else {
-				gfs2_holder_uninit(ghs);
-				return PTR_ERR(inode);
-			}
-		}
-	}
-
-	d_instantiate(dentry, inode);
-
-	return 0;
-}
-
-/**
- * gfs2_lookup - Look up a filename in a directory and return its inode
- * @dir: The directory inode
- * @dentry: The dentry of the new inode
- * @nd: passed from Linux VFS, ignored by us
- *
- * Called by the VFS layer. Lock dir and call gfs2_lookupi()
- *
- * Returns: errno
- */
-
-static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry,
-				  struct nameidata *nd)
-{
-	struct inode *inode = NULL;
-
-	inode = gfs2_lookupi(dir, &dentry->d_name, 0);
-	if (inode && IS_ERR(inode))
-		return ERR_CAST(inode);
-
-	if (inode) {
-		struct gfs2_glock *gl = GFS2_I(inode)->i_gl;
-		struct gfs2_holder gh;
-		int error;
-		error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
-		if (error) {
-			iput(inode);
-			return ERR_PTR(error);
-		}
-		gfs2_glock_dq_uninit(&gh);
-		return d_splice_alias(inode, dentry);
-	}
-	d_add(dentry, inode);
-
-	return NULL;
-}
-
-/**
- * gfs2_link - Link to a file
- * @old_dentry: The inode to link
- * @dir: Add link to this directory
- * @dentry: The name of the link
- *
- * Link the inode in "old_dentry" into the directory "dir" with the
- * name in "dentry".
- *
- * Returns: errno
- */
-
-static int gfs2_link(struct dentry *old_dentry, struct inode *dir,
-		     struct dentry *dentry)
-{
-	struct gfs2_inode *dip = GFS2_I(dir);
-	struct gfs2_sbd *sdp = GFS2_SB(dir);
-	struct inode *inode = old_dentry->d_inode;
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_holder ghs[2];
-	int alloc_required;
-	int error;
-
-	if (S_ISDIR(inode->i_mode))
-		return -EPERM;
-
-	gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
-	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1);
-
-	error = gfs2_glock_nq(ghs); /* parent */
-	if (error)
-		goto out_parent;
-
-	error = gfs2_glock_nq(ghs + 1); /* child */
-	if (error)
-		goto out_child;
-
-	error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC, 0);
-	if (error)
-		goto out_gunlock;
-
-	error = gfs2_dir_check(dir, &dentry->d_name, NULL);
-	switch (error) {
-	case -ENOENT:
-		break;
-	case 0:
-		error = -EEXIST;
-	default:
-		goto out_gunlock;
-	}
-
-	error = -EINVAL;
-	if (!dip->i_inode.i_nlink)
-		goto out_gunlock;
-	error = -EFBIG;
-	if (dip->i_entries == (u32)-1)
-		goto out_gunlock;
-	error = -EPERM;
-	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		goto out_gunlock;
-	error = -EINVAL;
-	if (!ip->i_inode.i_nlink)
-		goto out_gunlock;
-	error = -EMLINK;
-	if (ip->i_inode.i_nlink == (u32)-1)
-		goto out_gunlock;
-
-	alloc_required = error = gfs2_diradd_alloc_required(dir, &dentry->d_name);
-	if (error < 0)
-		goto out_gunlock;
-	error = 0;
-
-	if (alloc_required) {
-		struct gfs2_alloc *al = gfs2_alloc_get(dip);
-		if (!al) {
-			error = -ENOMEM;
-			goto out_gunlock;
-		}
-
-		error = gfs2_quota_lock_check(dip);
-		if (error)
-			goto out_alloc;
-
-		al->al_requested = sdp->sd_max_dirres;
-
-		error = gfs2_inplace_reserve(dip);
-		if (error)
-			goto out_gunlock_q;
-
-		error = gfs2_trans_begin(sdp, sdp->sd_max_dirres +
-					 gfs2_rg_blocks(al) +
-					 2 * RES_DINODE + RES_STATFS +
-					 RES_QUOTA, 0);
-		if (error)
-			goto out_ipres;
-	} else {
-		error = gfs2_trans_begin(sdp, 2 * RES_DINODE + RES_LEAF, 0);
-		if (error)
-			goto out_ipres;
-	}
-
-	error = gfs2_dir_add(dir, &dentry->d_name, ip, IF2DT(inode->i_mode));
-	if (error)
-		goto out_end_trans;
-
-	error = gfs2_change_nlink(ip, +1);
-
-out_end_trans:
-	gfs2_trans_end(sdp);
-out_ipres:
-	if (alloc_required)
-		gfs2_inplace_release(dip);
-out_gunlock_q:
-	if (alloc_required)
-		gfs2_quota_unlock(dip);
-out_alloc:
-	if (alloc_required)
-		gfs2_alloc_put(dip);
-out_gunlock:
-	gfs2_glock_dq(ghs + 1);
-out_child:
-	gfs2_glock_dq(ghs);
-out_parent:
-	gfs2_holder_uninit(ghs);
-	gfs2_holder_uninit(ghs + 1);
-	if (!error) {
-		ihold(inode);
-		d_instantiate(dentry, inode);
-		mark_inode_dirty(inode);
-	}
-	return error;
-}
-
-/*
- * gfs2_unlink_ok - check to see that a inode is still in a directory
- * @dip: the directory
- * @name: the name of the file
- * @ip: the inode
- *
- * Assumes that the lock on (at least) @dip is held.
- *
- * Returns: 0 if the parent/child relationship is correct, errno if it isn't
- */
-
-static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name,
-			  const struct gfs2_inode *ip)
-{
-	int error;
-
-	if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode))
-		return -EPERM;
-
-	if ((dip->i_inode.i_mode & S_ISVTX) &&
-	    dip->i_inode.i_uid != current_fsuid() &&
-	    ip->i_inode.i_uid != current_fsuid() && !capable(CAP_FOWNER))
-		return -EPERM;
-
-	if (IS_APPEND(&dip->i_inode))
-		return -EPERM;
-
-	error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0);
-	if (error)
-		return error;
-
-	error = gfs2_dir_check(&dip->i_inode, name, ip);
-	if (error)
-		return error;
-
-	return 0;
-}
-
-/**
- * gfs2_unlink - Unlink a file
- * @dir: The inode of the directory containing the file to unlink
- * @dentry: The file itself
- *
- * Unlink a file.  Call gfs2_unlinki()
- *
- * Returns: errno
- */
-
-static int gfs2_unlink(struct inode *dir, struct dentry *dentry)
-{
-	struct gfs2_inode *dip = GFS2_I(dir);
-	struct gfs2_sbd *sdp = GFS2_SB(dir);
-	struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
-	struct gfs2_holder ghs[3];
-	struct gfs2_rgrpd *rgd;
-	struct gfs2_holder ri_gh;
-	int error;
-
-	error = gfs2_rindex_hold(sdp, &ri_gh);
-	if (error)
-		return error;
-
-	gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
-	gfs2_holder_init(ip->i_gl,  LM_ST_EXCLUSIVE, 0, ghs + 1);
-
-	rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr);
-	gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2);
-
-
-	error = gfs2_glock_nq(ghs); /* parent */
-	if (error)
-		goto out_parent;
-
-	error = gfs2_glock_nq(ghs + 1); /* child */
-	if (error)
-		goto out_child;
-
-	error = gfs2_glock_nq(ghs + 2); /* rgrp */
-	if (error)
-		goto out_rgrp;
-
-	error = gfs2_unlink_ok(dip, &dentry->d_name, ip);
-	if (error)
-		goto out_gunlock;
-
-	error = gfs2_trans_begin(sdp, 2*RES_DINODE + RES_LEAF + RES_RG_BIT, 0);
-	if (error)
-		goto out_gunlock;
-
-	error = gfs2_dir_del(dip, &dentry->d_name);
-        if (error)
-                goto out_end_trans;
-
-	error = gfs2_change_nlink(ip, -1);
-
-out_end_trans:
-	gfs2_trans_end(sdp);
-out_gunlock:
-	gfs2_glock_dq(ghs + 2);
-out_rgrp:
-	gfs2_holder_uninit(ghs + 2);
-	gfs2_glock_dq(ghs + 1);
-out_child:
-	gfs2_holder_uninit(ghs + 1);
-	gfs2_glock_dq(ghs);
-out_parent:
-	gfs2_holder_uninit(ghs);
-	gfs2_glock_dq_uninit(&ri_gh);
-	return error;
-}
-
-/**
- * gfs2_symlink - Create a symlink
- * @dir: The directory to create the symlink in
- * @dentry: The dentry to put the symlink in
- * @symname: The thing which the link points to
- *
- * Returns: errno
- */
-
-static int gfs2_symlink(struct inode *dir, struct dentry *dentry,
-			const char *symname)
-{
-	struct gfs2_inode *dip = GFS2_I(dir), *ip;
-	struct gfs2_sbd *sdp = GFS2_SB(dir);
-	struct gfs2_holder ghs[2];
-	struct inode *inode;
-	struct buffer_head *dibh;
-	int size;
-	int error;
-
-	/* Must be stuffed with a null terminator for gfs2_follow_link() */
-	size = strlen(symname);
-	if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode) - 1)
-		return -ENAMETOOLONG;
-
-	gfs2_holder_init(dip->i_gl, 0, 0, ghs);
-
-	inode = gfs2_createi(ghs, &dentry->d_name, S_IFLNK | S_IRWXUGO, 0);
-	if (IS_ERR(inode)) {
-		gfs2_holder_uninit(ghs);
-		return PTR_ERR(inode);
-	}
-
-	ip = ghs[1].gh_gl->gl_object;
-
-	i_size_write(inode, size);
-
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-
-	if (!gfs2_assert_withdraw(sdp, !error)) {
-		gfs2_dinode_out(ip, dibh->b_data);
-		memcpy(dibh->b_data + sizeof(struct gfs2_dinode), symname,
-		       size);
-		brelse(dibh);
-	}
-
-	gfs2_trans_end(sdp);
-	if (dip->i_alloc->al_rgd)
-		gfs2_inplace_release(dip);
-	gfs2_quota_unlock(dip);
-	gfs2_alloc_put(dip);
-
-	gfs2_glock_dq_uninit_m(2, ghs);
-
-	d_instantiate(dentry, inode);
-	mark_inode_dirty(inode);
-
-	return 0;
-}
-
-/**
- * gfs2_mkdir - Make a directory
- * @dir: The parent directory of the new one
- * @dentry: The dentry of the new directory
- * @mode: The mode of the new directory
- *
- * Returns: errno
- */
-
-static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode)
-{
-	struct gfs2_inode *dip = GFS2_I(dir), *ip;
-	struct gfs2_sbd *sdp = GFS2_SB(dir);
-	struct gfs2_holder ghs[2];
-	struct inode *inode;
-	struct buffer_head *dibh;
-	int error;
-
-	gfs2_holder_init(dip->i_gl, 0, 0, ghs);
-
-	inode = gfs2_createi(ghs, &dentry->d_name, S_IFDIR | mode, 0);
-	if (IS_ERR(inode)) {
-		gfs2_holder_uninit(ghs);
-		return PTR_ERR(inode);
-	}
-
-	ip = ghs[1].gh_gl->gl_object;
-
-	ip->i_inode.i_nlink = 2;
-	i_size_write(inode, sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode));
-	ip->i_diskflags |= GFS2_DIF_JDATA;
-	ip->i_entries = 2;
-
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-
-	if (!gfs2_assert_withdraw(sdp, !error)) {
-		struct gfs2_dinode *di = (struct gfs2_dinode *)dibh->b_data;
-		struct gfs2_dirent *dent = (struct gfs2_dirent *)(di+1);
-
-		gfs2_trans_add_bh(ip->i_gl, dibh, 1);
-		gfs2_qstr2dirent(&gfs2_qdot, GFS2_DIRENT_SIZE(gfs2_qdot.len), dent);
-		dent->de_inum = di->di_num; /* already GFS2 endian */
-		dent->de_type = cpu_to_be16(DT_DIR);
-		di->di_entries = cpu_to_be32(1);
-
-		dent = (struct gfs2_dirent *)((char*)dent + GFS2_DIRENT_SIZE(1));
-		gfs2_qstr2dirent(&gfs2_qdotdot, dibh->b_size - GFS2_DIRENT_SIZE(1) - sizeof(struct gfs2_dinode), dent);
-
-		gfs2_inum_out(dip, dent);
-		dent->de_type = cpu_to_be16(DT_DIR);
-
-		gfs2_dinode_out(ip, di);
-
-		brelse(dibh);
-	}
-
-	error = gfs2_change_nlink(dip, +1);
-	gfs2_assert_withdraw(sdp, !error); /* dip already pinned */
-
-	gfs2_trans_end(sdp);
-	if (dip->i_alloc->al_rgd)
-		gfs2_inplace_release(dip);
-	gfs2_quota_unlock(dip);
-	gfs2_alloc_put(dip);
-
-	gfs2_glock_dq_uninit_m(2, ghs);
-
-	d_instantiate(dentry, inode);
-	mark_inode_dirty(inode);
-
-	return 0;
-}
-
-/**
- * gfs2_rmdiri - Remove a directory
- * @dip: The parent directory of the directory to be removed
- * @name: The name of the directory to be removed
- * @ip: The GFS2 inode of the directory to be removed
- *
- * Assumes Glocks on dip and ip are held
- *
- * Returns: errno
- */
-
-static int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name,
-		       struct gfs2_inode *ip)
-{
-	int error;
-
-	if (ip->i_entries != 2) {
-		if (gfs2_consist_inode(ip))
-			gfs2_dinode_print(ip);
-		return -EIO;
-	}
-
-	error = gfs2_dir_del(dip, name);
-	if (error)
-		return error;
-
-	error = gfs2_change_nlink(dip, -1);
-	if (error)
-		return error;
-
-	error = gfs2_dir_del(ip, &gfs2_qdot);
-	if (error)
-		return error;
-
-	error = gfs2_dir_del(ip, &gfs2_qdotdot);
-	if (error)
-		return error;
-
-	/* It looks odd, but it really should be done twice */
-	error = gfs2_change_nlink(ip, -1);
-	if (error)
-		return error;
-
-	error = gfs2_change_nlink(ip, -1);
-	if (error)
-		return error;
-
-	return error;
-}
-
-/**
- * gfs2_rmdir - Remove a directory
- * @dir: The parent directory of the directory to be removed
- * @dentry: The dentry of the directory to remove
- *
- * Remove a directory. Call gfs2_rmdiri()
- *
- * Returns: errno
- */
-
-static int gfs2_rmdir(struct inode *dir, struct dentry *dentry)
-{
-	struct gfs2_inode *dip = GFS2_I(dir);
-	struct gfs2_sbd *sdp = GFS2_SB(dir);
-	struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
-	struct gfs2_holder ghs[3];
-	struct gfs2_rgrpd *rgd;
-	struct gfs2_holder ri_gh;
-	int error;
-
-	error = gfs2_rindex_hold(sdp, &ri_gh);
-	if (error)
-		return error;
-	gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
-	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + 1);
-
-	rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr);
-	gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2);
-
-	error = gfs2_glock_nq(ghs); /* parent */
-	if (error)
-		goto out_parent;
-
-	error = gfs2_glock_nq(ghs + 1); /* child */
-	if (error)
-		goto out_child;
-
-	error = gfs2_glock_nq(ghs + 2); /* rgrp */
-	if (error)
-		goto out_rgrp;
-
-	error = gfs2_unlink_ok(dip, &dentry->d_name, ip);
-	if (error)
-		goto out_gunlock;
-
-	if (ip->i_entries < 2) {
-		if (gfs2_consist_inode(ip))
-			gfs2_dinode_print(ip);
-		error = -EIO;
-		goto out_gunlock;
-	}
-	if (ip->i_entries > 2) {
-		error = -ENOTEMPTY;
-		goto out_gunlock;
-	}
-
-	error = gfs2_trans_begin(sdp, 2 * RES_DINODE + 3 * RES_LEAF + RES_RG_BIT, 0);
-	if (error)
-		goto out_gunlock;
-
-	error = gfs2_rmdiri(dip, &dentry->d_name, ip);
-
-	gfs2_trans_end(sdp);
-
-out_gunlock:
-	gfs2_glock_dq(ghs + 2);
-out_rgrp:
-	gfs2_holder_uninit(ghs + 2);
-	gfs2_glock_dq(ghs + 1);
-out_child:
-	gfs2_holder_uninit(ghs + 1);
-	gfs2_glock_dq(ghs);
-out_parent:
-	gfs2_holder_uninit(ghs);
-	gfs2_glock_dq_uninit(&ri_gh);
-	return error;
-}
-
-/**
- * gfs2_mknod - Make a special file
- * @dir: The directory in which the special file will reside
- * @dentry: The dentry of the special file
- * @mode: The mode of the special file
- * @rdev: The device specification of the special file
- *
- */
-
-static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode,
-		      dev_t dev)
-{
-	struct gfs2_inode *dip = GFS2_I(dir);
-	struct gfs2_sbd *sdp = GFS2_SB(dir);
-	struct gfs2_holder ghs[2];
-	struct inode *inode;
-
-	gfs2_holder_init(dip->i_gl, 0, 0, ghs);
-
-	inode = gfs2_createi(ghs, &dentry->d_name, mode, dev);
-	if (IS_ERR(inode)) {
-		gfs2_holder_uninit(ghs);
-		return PTR_ERR(inode);
-	}
-
-	gfs2_trans_end(sdp);
-	if (dip->i_alloc->al_rgd)
-		gfs2_inplace_release(dip);
-	gfs2_quota_unlock(dip);
-	gfs2_alloc_put(dip);
-
-	gfs2_glock_dq_uninit_m(2, ghs);
-
-	d_instantiate(dentry, inode);
-	mark_inode_dirty(inode);
-
-	return 0;
-}
-
-/*
- * gfs2_ok_to_move - check if it's ok to move a directory to another directory
- * @this: move this
- * @to: to here
- *
- * Follow @to back to the root and make sure we don't encounter @this
- * Assumes we already hold the rename lock.
- *
- * Returns: errno
- */
-
-static int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to)
-{
-	struct inode *dir = &to->i_inode;
-	struct super_block *sb = dir->i_sb;
-	struct inode *tmp;
-	int error = 0;
-
-	igrab(dir);
-
-	for (;;) {
-		if (dir == &this->i_inode) {
-			error = -EINVAL;
-			break;
-		}
-		if (dir == sb->s_root->d_inode) {
-			error = 0;
-			break;
-		}
-
-		tmp = gfs2_lookupi(dir, &gfs2_qdotdot, 1);
-		if (IS_ERR(tmp)) {
-			error = PTR_ERR(tmp);
-			break;
-		}
-
-		iput(dir);
-		dir = tmp;
-	}
-
-	iput(dir);
-
-	return error;
-}
-
-/**
- * gfs2_rename - Rename a file
- * @odir: Parent directory of old file name
- * @odentry: The old dentry of the file
- * @ndir: Parent directory of new file name
- * @ndentry: The new dentry of the file
- *
- * Returns: errno
- */
-
-static int gfs2_rename(struct inode *odir, struct dentry *odentry,
-		       struct inode *ndir, struct dentry *ndentry)
-{
-	struct gfs2_inode *odip = GFS2_I(odir);
-	struct gfs2_inode *ndip = GFS2_I(ndir);
-	struct gfs2_inode *ip = GFS2_I(odentry->d_inode);
-	struct gfs2_inode *nip = NULL;
-	struct gfs2_sbd *sdp = GFS2_SB(odir);
-	struct gfs2_holder ghs[5], r_gh = { .gh_gl = NULL, }, ri_gh;
-	struct gfs2_rgrpd *nrgd;
-	unsigned int num_gh;
-	int dir_rename = 0;
-	int alloc_required = 0;
-	unsigned int x;
-	int error;
-
-	if (ndentry->d_inode) {
-		nip = GFS2_I(ndentry->d_inode);
-		if (ip == nip)
-			return 0;
-	}
-
-	error = gfs2_rindex_hold(sdp, &ri_gh);
-	if (error)
-		return error;
-
-	if (odip != ndip) {
-		error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE,
-					   0, &r_gh);
-		if (error)
-			goto out;
-
-		if (S_ISDIR(ip->i_inode.i_mode)) {
-			dir_rename = 1;
-			/* don't move a dirctory into it's subdir */
-			error = gfs2_ok_to_move(ip, ndip);
-			if (error)
-				goto out_gunlock_r;
-		}
-	}
-
-	num_gh = 1;
-	gfs2_holder_init(odip->i_gl, LM_ST_EXCLUSIVE, 0, ghs);
-	if (odip != ndip) {
-		gfs2_holder_init(ndip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh);
-		num_gh++;
-	}
-	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh);
-	num_gh++;
-
-	if (nip) {
-		gfs2_holder_init(nip->i_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh);
-		num_gh++;
-		/* grab the resource lock for unlink flag twiddling 
-		 * this is the case of the target file already existing
-		 * so we unlink before doing the rename
-		 */
-		nrgd = gfs2_blk2rgrpd(sdp, nip->i_no_addr);
-		if (nrgd)
-			gfs2_holder_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + num_gh++);
-	}
-
-	for (x = 0; x < num_gh; x++) {
-		error = gfs2_glock_nq(ghs + x);
-		if (error)
-			goto out_gunlock;
-	}
-
-	/* Check out the old directory */
-
-	error = gfs2_unlink_ok(odip, &odentry->d_name, ip);
-	if (error)
-		goto out_gunlock;
-
-	/* Check out the new directory */
-
-	if (nip) {
-		error = gfs2_unlink_ok(ndip, &ndentry->d_name, nip);
-		if (error)
-			goto out_gunlock;
-
-		if (S_ISDIR(nip->i_inode.i_mode)) {
-			if (nip->i_entries < 2) {
-				if (gfs2_consist_inode(nip))
-					gfs2_dinode_print(nip);
-				error = -EIO;
-				goto out_gunlock;
-			}
-			if (nip->i_entries > 2) {
-				error = -ENOTEMPTY;
-				goto out_gunlock;
-			}
-		}
-	} else {
-		error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC, 0);
-		if (error)
-			goto out_gunlock;
-
-		error = gfs2_dir_check(ndir, &ndentry->d_name, NULL);
-		switch (error) {
-		case -ENOENT:
-			error = 0;
-			break;
-		case 0:
-			error = -EEXIST;
-		default:
-			goto out_gunlock;
-		};
-
-		if (odip != ndip) {
-			if (!ndip->i_inode.i_nlink) {
-				error = -EINVAL;
-				goto out_gunlock;
-			}
-			if (ndip->i_entries == (u32)-1) {
-				error = -EFBIG;
-				goto out_gunlock;
-			}
-			if (S_ISDIR(ip->i_inode.i_mode) &&
-			    ndip->i_inode.i_nlink == (u32)-1) {
-				error = -EMLINK;
-				goto out_gunlock;
-			}
-		}
-	}
-
-	/* Check out the dir to be renamed */
-
-	if (dir_rename) {
-		error = gfs2_permission(odentry->d_inode, MAY_WRITE, 0);
-		if (error)
-			goto out_gunlock;
-	}
-
-	if (nip == NULL)
-		alloc_required = gfs2_diradd_alloc_required(ndir, &ndentry->d_name);
-	error = alloc_required;
-	if (error < 0)
-		goto out_gunlock;
-	error = 0;
-
-	if (alloc_required) {
-		struct gfs2_alloc *al = gfs2_alloc_get(ndip);
-		if (!al) {
-			error = -ENOMEM;
-			goto out_gunlock;
-		}
-
-		error = gfs2_quota_lock_check(ndip);
-		if (error)
-			goto out_alloc;
-
-		al->al_requested = sdp->sd_max_dirres;
-
-		error = gfs2_inplace_reserve_ri(ndip);
-		if (error)
-			goto out_gunlock_q;
-
-		error = gfs2_trans_begin(sdp, sdp->sd_max_dirres +
-					 gfs2_rg_blocks(al) +
-					 4 * RES_DINODE + 4 * RES_LEAF +
-					 RES_STATFS + RES_QUOTA + 4, 0);
-		if (error)
-			goto out_ipreserv;
-	} else {
-		error = gfs2_trans_begin(sdp, 4 * RES_DINODE +
-					 5 * RES_LEAF + 4, 0);
-		if (error)
-			goto out_gunlock;
-	}
-
-	/* Remove the target file, if it exists */
-
-	if (nip) {
-		if (S_ISDIR(nip->i_inode.i_mode))
-			error = gfs2_rmdiri(ndip, &ndentry->d_name, nip);
-		else {
-			error = gfs2_dir_del(ndip, &ndentry->d_name);
-			if (error)
-				goto out_end_trans;
-			error = gfs2_change_nlink(nip, -1);
-		}
-		if (error)
-			goto out_end_trans;
-	}
-
-	if (dir_rename) {
-		error = gfs2_change_nlink(ndip, +1);
-		if (error)
-			goto out_end_trans;
-		error = gfs2_change_nlink(odip, -1);
-		if (error)
-			goto out_end_trans;
-
-		error = gfs2_dir_mvino(ip, &gfs2_qdotdot, ndip, DT_DIR);
-		if (error)
-			goto out_end_trans;
-	} else {
-		struct buffer_head *dibh;
-		error = gfs2_meta_inode_buffer(ip, &dibh);
-		if (error)
-			goto out_end_trans;
-		ip->i_inode.i_ctime = CURRENT_TIME;
-		gfs2_trans_add_bh(ip->i_gl, dibh, 1);
-		gfs2_dinode_out(ip, dibh->b_data);
-		brelse(dibh);
-	}
-
-	error = gfs2_dir_del(odip, &odentry->d_name);
-	if (error)
-		goto out_end_trans;
-
-	error = gfs2_dir_add(ndir, &ndentry->d_name, ip, IF2DT(ip->i_inode.i_mode));
-	if (error)
-		goto out_end_trans;
-
-out_end_trans:
-	gfs2_trans_end(sdp);
-out_ipreserv:
-	if (alloc_required)
-		gfs2_inplace_release(ndip);
-out_gunlock_q:
-	if (alloc_required)
-		gfs2_quota_unlock(ndip);
-out_alloc:
-	if (alloc_required)
-		gfs2_alloc_put(ndip);
-out_gunlock:
-	while (x--) {
-		gfs2_glock_dq(ghs + x);
-		gfs2_holder_uninit(ghs + x);
-	}
-out_gunlock_r:
-	if (r_gh.gh_gl)
-		gfs2_glock_dq_uninit(&r_gh);
-out:
-	gfs2_glock_dq_uninit(&ri_gh);
-	return error;
-}
-
-/**
- * gfs2_follow_link - Follow a symbolic link
- * @dentry: The dentry of the link
- * @nd: Data that we pass to vfs_follow_link()
- *
- * This can handle symlinks of any size.
- *
- * Returns: 0 on success or error code
- */
-
-static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd)
-{
-	struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
-	struct gfs2_holder i_gh;
-	struct buffer_head *dibh;
-	unsigned int x, size;
-	char *buf;
-	int error;
-
-	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, 0, &i_gh);
-	error = gfs2_glock_nq(&i_gh);
-	if (error) {
-		gfs2_holder_uninit(&i_gh);
-		nd_set_link(nd, ERR_PTR(error));
-		return NULL;
-	}
-
-	size = (unsigned int)i_size_read(&ip->i_inode);
-	if (size == 0) {
-		gfs2_consist_inode(ip);
-		buf = ERR_PTR(-EIO);
-		goto out;
-	}
-
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-	if (error) {
-		buf = ERR_PTR(error);
-		goto out;
-	}
-
-	x = size + 1;
-	buf = kmalloc(x, GFP_NOFS);
-	if (!buf)
-		buf = ERR_PTR(-ENOMEM);
-	else
-		memcpy(buf, dibh->b_data + sizeof(struct gfs2_dinode), x);
-	brelse(dibh);
-out:
-	gfs2_glock_dq_uninit(&i_gh);
-	nd_set_link(nd, buf);
-	return NULL;
-}
-
-static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p)
-{
-	char *s = nd_get_link(nd);
-	if (!IS_ERR(s))
-		kfree(s);
-}
-
-/**
- * gfs2_permission -
- * @inode: The inode
- * @mask: The mask to be tested
- * @flags: Indicates whether this is an RCU path walk or not
- *
- * This may be called from the VFS directly, or from within GFS2 with the
- * inode locked, so we look to see if the glock is already locked and only
- * lock the glock if its not already been done.
- *
- * Returns: errno
- */
-
-int gfs2_permission(struct inode *inode, int mask, unsigned int flags)
-{
-	struct gfs2_inode *ip;
-	struct gfs2_holder i_gh;
-	int error;
-	int unlock = 0;
-
-
-	ip = GFS2_I(inode);
-	if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) {
-		if (flags & IPERM_FLAG_RCU)
-			return -ECHILD;
-		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
-		if (error)
-			return error;
-		unlock = 1;
-	}
-
-	if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode))
-		error = -EACCES;
-	else
-		error = generic_permission(inode, mask, flags, gfs2_check_acl);
-	if (unlock)
-		gfs2_glock_dq_uninit(&i_gh);
-
-	return error;
-}
-
-static int setattr_chown(struct inode *inode, struct iattr *attr)
-{
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_sbd *sdp = GFS2_SB(inode);
-	u32 ouid, ogid, nuid, ngid;
-	int error;
-
-	ouid = inode->i_uid;
-	ogid = inode->i_gid;
-	nuid = attr->ia_uid;
-	ngid = attr->ia_gid;
-
-	if (!(attr->ia_valid & ATTR_UID) || ouid == nuid)
-		ouid = nuid = NO_QUOTA_CHANGE;
-	if (!(attr->ia_valid & ATTR_GID) || ogid == ngid)
-		ogid = ngid = NO_QUOTA_CHANGE;
-
-	if (!gfs2_alloc_get(ip))
-		return -ENOMEM;
-
-	error = gfs2_quota_lock(ip, nuid, ngid);
-	if (error)
-		goto out_alloc;
-
-	if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) {
-		error = gfs2_quota_check(ip, nuid, ngid);
-		if (error)
-			goto out_gunlock_q;
-	}
-
-	error = gfs2_trans_begin(sdp, RES_DINODE + 2 * RES_QUOTA, 0);
-	if (error)
-		goto out_gunlock_q;
-
-	error = gfs2_setattr_simple(ip, attr);
-	if (error)
-		goto out_end_trans;
-
-	if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) {
-		u64 blocks = gfs2_get_inode_blocks(&ip->i_inode);
-		gfs2_quota_change(ip, -blocks, ouid, ogid);
-		gfs2_quota_change(ip, blocks, nuid, ngid);
-	}
-
-out_end_trans:
-	gfs2_trans_end(sdp);
-out_gunlock_q:
-	gfs2_quota_unlock(ip);
-out_alloc:
-	gfs2_alloc_put(ip);
-	return error;
-}
-
-/**
- * gfs2_setattr - Change attributes on an inode
- * @dentry: The dentry which is changing
- * @attr: The structure describing the change
- *
- * The VFS layer wants to change one or more of an inodes attributes.  Write
- * that change out to disk.
- *
- * Returns: errno
- */
-
-static int gfs2_setattr(struct dentry *dentry, struct iattr *attr)
-{
-	struct inode *inode = dentry->d_inode;
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_holder i_gh;
-	int error;
-
-	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh);
-	if (error)
-		return error;
-
-	error = -EPERM;
-	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		goto out;
-
-	error = inode_change_ok(inode, attr);
-	if (error)
-		goto out;
-
-	if (attr->ia_valid & ATTR_SIZE)
-		error = gfs2_setattr_size(inode, attr->ia_size);
-	else if (attr->ia_valid & (ATTR_UID | ATTR_GID))
-		error = setattr_chown(inode, attr);
-	else if ((attr->ia_valid & ATTR_MODE) && IS_POSIXACL(inode))
-		error = gfs2_acl_chmod(ip, attr);
-	else
-		error = gfs2_setattr_simple(ip, attr);
-
-out:
-	gfs2_glock_dq_uninit(&i_gh);
-	if (!error)
-		mark_inode_dirty(inode);
-	return error;
-}
-
-/**
- * gfs2_getattr - Read out an inode's attributes
- * @mnt: The vfsmount the inode is being accessed from
- * @dentry: The dentry to stat
- * @stat: The inode's stats
- *
- * This may be called from the VFS directly, or from within GFS2 with the
- * inode locked, so we look to see if the glock is already locked and only
- * lock the glock if its not already been done. Note that its the NFS
- * readdirplus operation which causes this to be called (from filldir)
- * with the glock already held.
- *
- * Returns: errno
- */
-
-static int gfs2_getattr(struct vfsmount *mnt, struct dentry *dentry,
-			struct kstat *stat)
-{
-	struct inode *inode = dentry->d_inode;
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_holder gh;
-	int error;
-	int unlock = 0;
-
-	if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) {
-		error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
-		if (error)
-			return error;
-		unlock = 1;
-	}
-
-	generic_fillattr(inode, stat);
-	if (unlock)
-		gfs2_glock_dq_uninit(&gh);
-
-	return 0;
-}
-
-static int gfs2_setxattr(struct dentry *dentry, const char *name,
-			 const void *data, size_t size, int flags)
-{
-	struct inode *inode = dentry->d_inode;
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_holder gh;
-	int ret;
-
-	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
-	ret = gfs2_glock_nq(&gh);
-	if (ret == 0) {
-		ret = generic_setxattr(dentry, name, data, size, flags);
-		gfs2_glock_dq(&gh);
-	}
-	gfs2_holder_uninit(&gh);
-	return ret;
-}
-
-static ssize_t gfs2_getxattr(struct dentry *dentry, const char *name,
-			     void *data, size_t size)
-{
-	struct inode *inode = dentry->d_inode;
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_holder gh;
-	int ret;
-
-	gfs2_holder_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh);
-	ret = gfs2_glock_nq(&gh);
-	if (ret == 0) {
-		ret = generic_getxattr(dentry, name, data, size);
-		gfs2_glock_dq(&gh);
-	}
-	gfs2_holder_uninit(&gh);
-	return ret;
-}
-
-static int gfs2_removexattr(struct dentry *dentry, const char *name)
-{
-	struct inode *inode = dentry->d_inode;
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_holder gh;
-	int ret;
-
-	gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
-	ret = gfs2_glock_nq(&gh);
-	if (ret == 0) {
-		ret = generic_removexattr(dentry, name);
-		gfs2_glock_dq(&gh);
-	}
-	gfs2_holder_uninit(&gh);
-	return ret;
-}
-
-static int gfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
-		       u64 start, u64 len)
-{
-	struct gfs2_inode *ip = GFS2_I(inode);
-	struct gfs2_holder gh;
-	int ret;
-
-	ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC);
-	if (ret)
-		return ret;
-
-	mutex_lock(&inode->i_mutex);
-
-	ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
-	if (ret)
-		goto out;
-
-	if (gfs2_is_stuffed(ip)) {
-		u64 phys = ip->i_no_addr << inode->i_blkbits;
-		u64 size = i_size_read(inode);
-		u32 flags = FIEMAP_EXTENT_LAST|FIEMAP_EXTENT_NOT_ALIGNED|
-			    FIEMAP_EXTENT_DATA_INLINE;
-		phys += sizeof(struct gfs2_dinode);
-		phys += start;
-		if (start + len > size)
-			len = size - start;
-		if (start < size)
-			ret = fiemap_fill_next_extent(fieinfo, start, phys,
-						      len, flags);
-		if (ret == 1)
-			ret = 0;
-	} else {
-		ret = __generic_block_fiemap(inode, fieinfo, start, len,
-					     gfs2_block_map);
-	}
-
-	gfs2_glock_dq_uninit(&gh);
-out:
-	mutex_unlock(&inode->i_mutex);
-	return ret;
-}
-
-const struct inode_operations gfs2_file_iops = {
-	.permission = gfs2_permission,
-	.setattr = gfs2_setattr,
-	.getattr = gfs2_getattr,
-	.setxattr = gfs2_setxattr,
-	.getxattr = gfs2_getxattr,
-	.listxattr = gfs2_listxattr,
-	.removexattr = gfs2_removexattr,
-	.fiemap = gfs2_fiemap,
-};
-
-const struct inode_operations gfs2_dir_iops = {
-	.create = gfs2_create,
-	.lookup = gfs2_lookup,
-	.link = gfs2_link,
-	.unlink = gfs2_unlink,
-	.symlink = gfs2_symlink,
-	.mkdir = gfs2_mkdir,
-	.rmdir = gfs2_rmdir,
-	.mknod = gfs2_mknod,
-	.rename = gfs2_rename,
-	.permission = gfs2_permission,
-	.setattr = gfs2_setattr,
-	.getattr = gfs2_getattr,
-	.setxattr = gfs2_setxattr,
-	.getxattr = gfs2_getxattr,
-	.listxattr = gfs2_listxattr,
-	.removexattr = gfs2_removexattr,
-	.fiemap = gfs2_fiemap,
-};
-
-const struct inode_operations gfs2_symlink_iops = {
-	.readlink = generic_readlink,
-	.follow_link = gfs2_follow_link,
-	.put_link = gfs2_put_link,
-	.permission = gfs2_permission,
-	.setattr = gfs2_setattr,
-	.getattr = gfs2_getattr,
-	.setxattr = gfs2_setxattr,
-	.getxattr = gfs2_getxattr,
-	.listxattr = gfs2_listxattr,
-	.removexattr = gfs2_removexattr,
-	.fiemap = gfs2_fiemap,
-};
-
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index cf930cd9664..7273ad3c85b 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -78,10 +78,11 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal,
 
 static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1,
 			       unsigned char *buf2, unsigned int offset,
-			       unsigned int buflen, u32 block,
+			       struct gfs2_bitmap *bi, u32 block,
 			       unsigned char new_state)
 {
 	unsigned char *byte1, *byte2, *end, cur_state;
+	unsigned int buflen = bi->bi_len;
 	const unsigned int bit = (block % GFS2_NBBY) * GFS2_BIT_SIZE;
 
 	byte1 = buf1 + offset + (block / GFS2_NBBY);
@@ -92,6 +93,16 @@ static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1,
 	cur_state = (*byte1 >> bit) & GFS2_BIT_MASK;
 
 	if (unlikely(!valid_change[new_state * 4 + cur_state])) {
+		printk(KERN_WARNING "GFS2: buf_blk = 0x%llx old_state=%d, "
+		       "new_state=%d\n",
+		       (unsigned long long)block, cur_state, new_state);
+		printk(KERN_WARNING "GFS2: rgrp=0x%llx bi_start=0x%lx\n",
+		       (unsigned long long)rgd->rd_addr,
+		       (unsigned long)bi->bi_start);
+		printk(KERN_WARNING "GFS2: bi_offset=0x%lx bi_len=0x%lx\n",
+		       (unsigned long)bi->bi_offset,
+		       (unsigned long)bi->bi_len);
+		dump_stack();
 		gfs2_consist_rgrpd(rgd);
 		return;
 	}
@@ -381,6 +392,7 @@ static void clear_rgrpdi(struct gfs2_sbd *sdp)
 
 		if (gl) {
 			gl->gl_object = NULL;
+			gfs2_glock_add_to_lru(gl);
 			gfs2_glock_put(gl);
 		}
 
@@ -945,7 +957,7 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip
 		/* rgblk_search can return a block < goal, so we need to
 		   keep it marching forward. */
 		no_addr = block + rgd->rd_data0;
-		goal++;
+		goal = max(block + 1, goal + 1);
 		if (*last_unlinked != NO_BLOCK && no_addr <= *last_unlinked)
 			continue;
 		if (no_addr == skip)
@@ -971,7 +983,7 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip
 			found++;
 
 		/* Limit reclaim to sensible number of tasks */
-		if (found > 2*NR_CPUS)
+		if (found > NR_CPUS)
 			return;
 	}
 
@@ -1365,7 +1377,7 @@ skip:
 
 	gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1);
 	gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset,
-		    bi->bi_len, blk, new_state);
+		    bi, blk, new_state);
 	goal = blk;
 	while (*n < elen) {
 		goal++;
@@ -1375,7 +1387,7 @@ skip:
 		    GFS2_BLKST_FREE)
 			break;
 		gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset,
-			    bi->bi_len, goal, new_state);
+			    bi, goal, new_state);
 		(*n)++;
 	}
 out:
@@ -1432,7 +1444,7 @@ static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart,
 		}
 		gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1);
 		gfs2_setbit(rgd, bi->bi_bh->b_data, NULL, bi->bi_offset,
-			    bi->bi_len, buf_blk, new_state);
+			    bi, buf_blk, new_state);
 	}
 
 	return rgd;
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index a4e23d68a39..ed540e7018b 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -23,6 +23,7 @@
 #include <linux/time.h>
 #include <linux/wait.h>
 #include <linux/writeback.h>
+#include <linux/backing-dev.h>
 
 #include "gfs2.h"
 #include "incore.h"
@@ -700,11 +701,47 @@ void gfs2_unfreeze_fs(struct gfs2_sbd *sdp)
 	mutex_unlock(&sdp->sd_freeze_lock);
 }
 
+void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf)
+{
+	struct gfs2_dinode *str = buf;
+
+	str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC);
+	str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI);
+	str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI);
+	str->di_num.no_addr = cpu_to_be64(ip->i_no_addr);
+	str->di_num.no_formal_ino = cpu_to_be64(ip->i_no_formal_ino);
+	str->di_mode = cpu_to_be32(ip->i_inode.i_mode);
+	str->di_uid = cpu_to_be32(ip->i_inode.i_uid);
+	str->di_gid = cpu_to_be32(ip->i_inode.i_gid);
+	str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink);
+	str->di_size = cpu_to_be64(i_size_read(&ip->i_inode));
+	str->di_blocks = cpu_to_be64(gfs2_get_inode_blocks(&ip->i_inode));
+	str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec);
+	str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec);
+	str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec);
+
+	str->di_goal_meta = cpu_to_be64(ip->i_goal);
+	str->di_goal_data = cpu_to_be64(ip->i_goal);
+	str->di_generation = cpu_to_be64(ip->i_generation);
+
+	str->di_flags = cpu_to_be32(ip->i_diskflags);
+	str->di_height = cpu_to_be16(ip->i_height);
+	str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) &&
+					     !(ip->i_diskflags & GFS2_DIF_EXHASH) ?
+					     GFS2_FORMAT_DE : 0);
+	str->di_depth = cpu_to_be16(ip->i_depth);
+	str->di_entries = cpu_to_be32(ip->i_entries);
+
+	str->di_eattr = cpu_to_be64(ip->i_eattr);
+	str->di_atime_nsec = cpu_to_be32(ip->i_inode.i_atime.tv_nsec);
+	str->di_mtime_nsec = cpu_to_be32(ip->i_inode.i_mtime.tv_nsec);
+	str->di_ctime_nsec = cpu_to_be32(ip->i_inode.i_ctime.tv_nsec);
+}
 
 /**
  * gfs2_write_inode - Make sure the inode is stable on the disk
  * @inode: The inode
- * @sync: synchronous write flag
+ * @wbc: The writeback control structure
  *
  * Returns: errno
  */
@@ -713,15 +750,17 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
+	struct address_space *metamapping = gfs2_glock2aspace(ip->i_gl);
+	struct backing_dev_info *bdi = metamapping->backing_dev_info;
 	struct gfs2_holder gh;
 	struct buffer_head *bh;
 	struct timespec atime;
 	struct gfs2_dinode *di;
-	int ret = 0;
+	int ret = -EAGAIN;
 
-	/* Check this is a "normal" inode, etc */
+	/* Skip timestamp update, if this is from a memalloc */
 	if (current->flags & PF_MEMALLOC)
-		return 0;
+		goto do_flush;
 	ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
 	if (ret)
 		goto do_flush;
@@ -745,6 +784,13 @@ do_unlock:
 do_flush:
 	if (wbc->sync_mode == WB_SYNC_ALL)
 		gfs2_log_flush(GFS2_SB(inode), ip->i_gl);
+	filemap_fdatawrite(metamapping);
+	if (bdi->dirty_exceeded)
+		gfs2_ail1_flush(sdp, wbc);
+	if (!ret && (wbc->sync_mode == WB_SYNC_ALL))
+		ret = filemap_fdatawait(metamapping);
+	if (ret)
+		mark_inode_dirty_sync(inode);
 	return ret;
 }
 
@@ -874,8 +920,9 @@ restart:
 
 static int gfs2_sync_fs(struct super_block *sb, int wait)
 {
-	if (wait && sb->s_fs_info)
-		gfs2_log_flush(sb->s_fs_info, NULL);
+	struct gfs2_sbd *sdp = sb->s_fs_info;
+	if (wait && sdp)
+		gfs2_log_flush(sdp, NULL);
 	return 0;
 }
 
@@ -1308,6 +1355,78 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt)
 	return 0;
 }
 
+static void gfs2_final_release_pages(struct gfs2_inode *ip)
+{
+	struct inode *inode = &ip->i_inode;
+	struct gfs2_glock *gl = ip->i_gl;
+
+	truncate_inode_pages(gfs2_glock2aspace(ip->i_gl), 0);
+	truncate_inode_pages(&inode->i_data, 0);
+
+	if (atomic_read(&gl->gl_revokes) == 0) {
+		clear_bit(GLF_LFLUSH, &gl->gl_flags);
+		clear_bit(GLF_DIRTY, &gl->gl_flags);
+	}
+}
+
+static int gfs2_dinode_dealloc(struct gfs2_inode *ip)
+{
+	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
+	struct gfs2_alloc *al;
+	struct gfs2_rgrpd *rgd;
+	int error;
+
+	if (gfs2_get_inode_blocks(&ip->i_inode) != 1) {
+		gfs2_consist_inode(ip);
+		return -EIO;
+	}
+
+	al = gfs2_alloc_get(ip);
+	if (!al)
+		return -ENOMEM;
+
+	error = gfs2_quota_hold(ip, NO_QUOTA_CHANGE, NO_QUOTA_CHANGE);
+	if (error)
+		goto out;
+
+	error = gfs2_rindex_hold(sdp, &al->al_ri_gh);
+	if (error)
+		goto out_qs;
+
+	rgd = gfs2_blk2rgrpd(sdp, ip->i_no_addr);
+	if (!rgd) {
+		gfs2_consist_inode(ip);
+		error = -EIO;
+		goto out_rindex_relse;
+	}
+
+	error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0,
+				   &al->al_rgd_gh);
+	if (error)
+		goto out_rindex_relse;
+
+	error = gfs2_trans_begin(sdp, RES_RG_BIT + RES_STATFS + RES_QUOTA,
+				 sdp->sd_jdesc->jd_blocks);
+	if (error)
+		goto out_rg_gunlock;
+
+	gfs2_free_di(rgd, ip);
+
+	gfs2_final_release_pages(ip);
+
+	gfs2_trans_end(sdp);
+
+out_rg_gunlock:
+	gfs2_glock_dq_uninit(&al->al_rgd_gh);
+out_rindex_relse:
+	gfs2_glock_dq_uninit(&al->al_ri_gh);
+out_qs:
+	gfs2_quota_unhold(ip);
+out:
+	gfs2_alloc_put(ip);
+	return error;
+}
+
 /*
  * We have to (at the moment) hold the inodes main lock to cover
  * the gap between unlocking the shared lock on the iopen lock and
@@ -1318,15 +1437,17 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt)
 
 static void gfs2_evict_inode(struct inode *inode)
 {
-	struct gfs2_sbd *sdp = inode->i_sb->s_fs_info;
+	struct super_block *sb = inode->i_sb;
+	struct gfs2_sbd *sdp = sb->s_fs_info;
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_holder gh;
 	int error;
 
-	if (inode->i_nlink)
+	if (inode->i_nlink || (sb->s_flags & MS_RDONLY))
 		goto out;
 
-	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+	/* Must not read inode block until block type has been verified */
+	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_SKIP, &gh);
 	if (unlikely(error)) {
 		gfs2_glock_dq_uninit(&ip->i_iopen_gh);
 		goto out;
@@ -1336,6 +1457,12 @@ static void gfs2_evict_inode(struct inode *inode)
 	if (error)
 		goto out_truncate;
 
+	if (test_bit(GIF_INVALID, &ip->i_flags)) {
+		error = gfs2_inode_refresh(ip);
+		if (error)
+			goto out_truncate;
+	}
+
 	ip->i_iopen_gh.gh_flags |= GL_NOCACHE;
 	gfs2_glock_dq_wait(&ip->i_iopen_gh);
 	gfs2_holder_reinit(LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB | GL_NOCACHE, &ip->i_iopen_gh);
@@ -1363,15 +1490,13 @@ static void gfs2_evict_inode(struct inode *inode)
 	}
 
 	error = gfs2_dinode_dealloc(ip);
-	if (error)
-		goto out_unlock;
+	goto out_unlock;
 
 out_truncate:
 	error = gfs2_trans_begin(sdp, 0, sdp->sd_jdesc->jd_blocks);
 	if (error)
 		goto out_unlock;
-	/* Needs to be done before glock release & also in a transaction */
-	truncate_inode_pages(&inode->i_data, 0);
+	gfs2_final_release_pages(ip);
 	gfs2_trans_end(sdp);
 
 out_unlock:
@@ -1386,6 +1511,7 @@ out:
 	end_writeback(inode);
 
 	ip->i_gl->gl_object = NULL;
+	gfs2_glock_add_to_lru(ip->i_gl);
 	gfs2_glock_put(ip->i_gl);
 	ip->i_gl = NULL;
 	if (ip->i_iopen_gh.gh_gl) {
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index 748ccb557c1..e20eab37bc8 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -81,7 +81,8 @@ static int gfs2_uuid_valid(const u8 *uuid)
 
 static ssize_t uuid_show(struct gfs2_sbd *sdp, char *buf)
 {
-	const u8 *uuid = sdp->sd_sb.sb_uuid;
+	struct super_block *s = sdp->sd_vfs;
+	const u8 *uuid = s->s_uuid;
 	buf[0] = '\0';
 	if (!gfs2_uuid_valid(uuid))
 		return 0;
@@ -616,7 +617,8 @@ static int gfs2_uevent(struct kset *kset, struct kobject *kobj,
 		       struct kobj_uevent_env *env)
 {
 	struct gfs2_sbd *sdp = container_of(kobj, struct gfs2_sbd, sd_kobj);
-	const u8 *uuid = sdp->sd_sb.sb_uuid;
+	struct super_block *s = sdp->sd_vfs;
+	const u8 *uuid = s->s_uuid;
 
 	add_uevent_var(env, "LOCKTABLE=%s", sdp->sd_table_name);
 	add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name);
diff --git a/fs/gfs2/trace_gfs2.h b/fs/gfs2/trace_gfs2.h
index cedb0bb96d9..5d07609ec57 100644
--- a/fs/gfs2/trace_gfs2.h
+++ b/fs/gfs2/trace_gfs2.h
@@ -10,6 +10,7 @@
 #include <linux/buffer_head.h>
 #include <linux/dlmconstants.h>
 #include <linux/gfs2_ondisk.h>
+#include <linux/writeback.h>
 #include "incore.h"
 #include "glock.h"
 
@@ -40,7 +41,9 @@
 	{(1UL << GLF_REPLY_PENDING),		"r" },		\
 	{(1UL << GLF_INITIAL),			"I" },		\
 	{(1UL << GLF_FROZEN),			"F" },		\
-	{(1UL << GLF_QUEUED),			"q" })
+	{(1UL << GLF_QUEUED),			"q" },		\
+	{(1UL << GLF_LRU),			"L" },		\
+	{(1UL << GLF_OBJECT),			"o" })
 
 #ifndef NUMPTY
 #define NUMPTY
@@ -94,7 +97,7 @@ TRACE_EVENT(gfs2_glock_state_change,
 		__entry->new_state	= glock_trace_state(new_state);
 		__entry->tgt_state	= glock_trace_state(gl->gl_target);
 		__entry->dmt_state	= glock_trace_state(gl->gl_demote_state);
-		__entry->flags		= gl->gl_flags;
+		__entry->flags		= gl->gl_flags | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0);
 	),
 
 	TP_printk("%u,%u glock %d:%lld state %s to %s tgt:%s dmt:%s flags:%s",
@@ -127,7 +130,7 @@ TRACE_EVENT(gfs2_glock_put,
 		__entry->gltype		= gl->gl_name.ln_type;
 		__entry->glnum		= gl->gl_name.ln_number;
 		__entry->cur_state	= glock_trace_state(gl->gl_state);
-		__entry->flags		= gl->gl_flags;
+		__entry->flags		= gl->gl_flags  | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0);
 	),
 
 	TP_printk("%u,%u glock %d:%lld state %s => %s flags:%s",
@@ -161,7 +164,7 @@ TRACE_EVENT(gfs2_demote_rq,
 		__entry->glnum		= gl->gl_name.ln_number;
 		__entry->cur_state	= glock_trace_state(gl->gl_state);
 		__entry->dmt_state	= glock_trace_state(gl->gl_demote_state);
-		__entry->flags		= gl->gl_flags;
+		__entry->flags		= gl->gl_flags  | (gl->gl_object ? (1UL<<GLF_OBJECT) : 0);
 	),
 
 	TP_printk("%u,%u glock %d:%lld demote %s to %s flags:%s",
@@ -318,6 +321,33 @@ TRACE_EVENT(gfs2_log_blocks,
 		  MINOR(__entry->dev), __entry->blocks)
 );
 
+/* Writing back the AIL */
+TRACE_EVENT(gfs2_ail_flush,
+
+	TP_PROTO(const struct gfs2_sbd *sdp, const struct writeback_control *wbc, int start),
+
+	TP_ARGS(sdp, wbc, start),
+
+	TP_STRUCT__entry(
+		__field(	dev_t,	dev			)
+		__field(	int, start			)
+		__field(	int, sync_mode			)
+		__field(	long, nr_to_write		)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= sdp->sd_vfs->s_dev;
+		__entry->start		= start;
+		__entry->sync_mode	= wbc->sync_mode;
+		__entry->nr_to_write	= wbc->nr_to_write;
+	),
+
+	TP_printk("%u,%u ail flush %s %s %ld", MAJOR(__entry->dev),
+		  MINOR(__entry->dev), __entry->start ? "start" : "end",
+		  __entry->sync_mode == WB_SYNC_ALL ? "all" : "none",
+		  __entry->nr_to_write)
+);
+
 /* Section 3 - bmap
  *
  * Objectives:
diff --git a/fs/hpfs/Kconfig b/fs/hpfs/Kconfig
index 0c39dc3ef7d..56bd15c5bf6 100644
--- a/fs/hpfs/Kconfig
+++ b/fs/hpfs/Kconfig
@@ -1,7 +1,6 @@
 config HPFS_FS
 	tristate "OS/2 HPFS file system support"
 	depends on BLOCK
-	depends on BROKEN || !PREEMPT
 	help
 	  OS/2 is IBM's operating system for PC's, the same as Warp, and HPFS
 	  is the file system used for organizing files on OS/2 hard disk
diff --git a/fs/hpfs/alloc.c b/fs/hpfs/alloc.c
index 5503e2c2891..7a5eb2c718c 100644
--- a/fs/hpfs/alloc.c
+++ b/fs/hpfs/alloc.c
@@ -8,8 +8,6 @@
 
 #include "hpfs_fn.h"
 
-static int hpfs_alloc_if_possible_nolock(struct super_block *s, secno sec);
-
 /*
  * Check if a sector is allocated in bitmap
  * This is really slow. Turned on only if chk==2
@@ -18,9 +16,9 @@ static int hpfs_alloc_if_possible_nolock(struct super_block *s, secno sec);
 static int chk_if_allocated(struct super_block *s, secno sec, char *msg)
 {
 	struct quad_buffer_head qbh;
-	unsigned *bmp;
+	u32 *bmp;
 	if (!(bmp = hpfs_map_bitmap(s, sec >> 14, &qbh, "chk"))) goto fail;
-	if ((bmp[(sec & 0x3fff) >> 5] >> (sec & 0x1f)) & 1) {
+	if ((cpu_to_le32(bmp[(sec & 0x3fff) >> 5]) >> (sec & 0x1f)) & 1) {
 		hpfs_error(s, "sector '%s' - %08x not allocated in bitmap", msg, sec);
 		goto fail1;
 	}
@@ -28,7 +26,7 @@ static int chk_if_allocated(struct super_block *s, secno sec, char *msg)
 	if (sec >= hpfs_sb(s)->sb_dirband_start && sec < hpfs_sb(s)->sb_dirband_start + hpfs_sb(s)->sb_dirband_size) {
 		unsigned ssec = (sec - hpfs_sb(s)->sb_dirband_start) / 4;
 		if (!(bmp = hpfs_map_dnode_bitmap(s, &qbh))) goto fail;
-		if ((bmp[ssec >> 5] >> (ssec & 0x1f)) & 1) {
+		if ((le32_to_cpu(bmp[ssec >> 5]) >> (ssec & 0x1f)) & 1) {
 			hpfs_error(s, "sector '%s' - %08x not allocated in directory bitmap", msg, sec);
 			goto fail1;
 		}
@@ -75,7 +73,6 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne
 		hpfs_error(s, "Bad allocation size: %d", n);
 		return 0;
 	}
-	lock_super(s);
 	if (bs != ~0x3fff) {
 		if (!(bmp = hpfs_map_bitmap(s, near >> 14, &qbh, "aib"))) goto uls;
 	} else {
@@ -85,10 +82,6 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne
 		ret = bs + nr;
 		goto rt;
 	}
-	/*if (!tstbits(bmp, nr + n, n + forward)) {
-		ret = bs + nr + n;
-		goto rt;
-	}*/
 	q = nr + n; b = 0;
 	while ((a = tstbits(bmp, q, n + forward)) != 0) {
 		q += a;
@@ -105,14 +98,14 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne
 		goto rt;
 	}
 	nr >>= 5;
-	/*for (i = nr + 1; i != nr; i++, i &= 0x1ff) {*/
+	/*for (i = nr + 1; i != nr; i++, i &= 0x1ff) */
 	i = nr;
 	do {
-		if (!bmp[i]) goto cont;
-		if (n + forward >= 0x3f && bmp[i] != -1) goto cont;
+		if (!le32_to_cpu(bmp[i])) goto cont;
+		if (n + forward >= 0x3f && le32_to_cpu(bmp[i]) != 0xffffffff) goto cont;
 		q = i<<5;
 		if (i > 0) {
-			unsigned k = bmp[i-1];
+			unsigned k = le32_to_cpu(bmp[i-1]);
 			while (k & 0x80000000) {
 				q--; k <<= 1;
 			}
@@ -132,18 +125,17 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne
 	} while (i != nr);
 	rt:
 	if (ret) {
-		if (hpfs_sb(s)->sb_chk && ((ret >> 14) != (bs >> 14) || (bmp[(ret & 0x3fff) >> 5] | ~(((1 << n) - 1) << (ret & 0x1f))) != 0xffffffff)) {
+		if (hpfs_sb(s)->sb_chk && ((ret >> 14) != (bs >> 14) || (le32_to_cpu(bmp[(ret & 0x3fff) >> 5]) | ~(((1 << n) - 1) << (ret & 0x1f))) != 0xffffffff)) {
 			hpfs_error(s, "Allocation doesn't work! Wanted %d, allocated at %08x", n, ret);
 			ret = 0;
 			goto b;
 		}
-		bmp[(ret & 0x3fff) >> 5] &= ~(((1 << n) - 1) << (ret & 0x1f));
+		bmp[(ret & 0x3fff) >> 5] &= cpu_to_le32(~(((1 << n) - 1) << (ret & 0x1f)));
 		hpfs_mark_4buffers_dirty(&qbh);
 	}
 	b:
 	hpfs_brelse4(&qbh);
 	uls:
-	unlock_super(s);
 	return ret;
 }
 
@@ -155,7 +147,7 @@ static secno alloc_in_bmp(struct super_block *s, secno near, unsigned n, unsigne
  *				sectors
  */
 
-secno hpfs_alloc_sector(struct super_block *s, secno near, unsigned n, int forward, int lock)
+secno hpfs_alloc_sector(struct super_block *s, secno near, unsigned n, int forward)
 {
 	secno sec;
 	int i;
@@ -167,7 +159,6 @@ secno hpfs_alloc_sector(struct super_block *s, secno near, unsigned n, int forwa
 		forward = -forward;
 		f_p = 1;
 	}
-	if (lock) hpfs_lock_creation(s);
 	n_bmps = (sbi->sb_fs_size + 0x4000 - 1) >> 14;
 	if (near && near < sbi->sb_fs_size) {
 		if ((sec = alloc_in_bmp(s, near, n, f_p ? forward : forward/4))) goto ret;
@@ -214,18 +205,17 @@ secno hpfs_alloc_sector(struct super_block *s, secno near, unsigned n, int forwa
 	ret:
 	if (sec && f_p) {
 		for (i = 0; i < forward; i++) {
-			if (!hpfs_alloc_if_possible_nolock(s, sec + i + 1)) {
+			if (!hpfs_alloc_if_possible(s, sec + i + 1)) {
 				hpfs_error(s, "Prealloc doesn't work! Wanted %d, allocated at %08x, can't allocate %d", forward, sec, i);
 				sec = 0;
 				break;
 			}
 		}
 	}
-	if (lock) hpfs_unlock_creation(s);
 	return sec;
 }
 
-static secno alloc_in_dirband(struct super_block *s, secno near, int lock)
+static secno alloc_in_dirband(struct super_block *s, secno near)
 {
 	unsigned nr = near;
 	secno sec;
@@ -236,49 +226,35 @@ static secno alloc_in_dirband(struct super_block *s, secno near, int lock)
 		nr = sbi->sb_dirband_start + sbi->sb_dirband_size - 4;
 	nr -= sbi->sb_dirband_start;
 	nr >>= 2;
-	if (lock) hpfs_lock_creation(s);
 	sec = alloc_in_bmp(s, (~0x3fff) | nr, 1, 0);
-	if (lock) hpfs_unlock_creation(s);
 	if (!sec) return 0;
 	return ((sec & 0x3fff) << 2) + sbi->sb_dirband_start;
 }
 
 /* Alloc sector if it's free */
 
-static int hpfs_alloc_if_possible_nolock(struct super_block *s, secno sec)
+int hpfs_alloc_if_possible(struct super_block *s, secno sec)
 {
 	struct quad_buffer_head qbh;
-	unsigned *bmp;
-	lock_super(s);
+	u32 *bmp;
 	if (!(bmp = hpfs_map_bitmap(s, sec >> 14, &qbh, "aip"))) goto end;
-	if (bmp[(sec & 0x3fff) >> 5] & (1 << (sec & 0x1f))) {
-		bmp[(sec & 0x3fff) >> 5] &= ~(1 << (sec & 0x1f));
+	if (le32_to_cpu(bmp[(sec & 0x3fff) >> 5]) & (1 << (sec & 0x1f))) {
+		bmp[(sec & 0x3fff) >> 5] &= cpu_to_le32(~(1 << (sec & 0x1f)));
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
-		unlock_super(s);
 		return 1;
 	}
 	hpfs_brelse4(&qbh);
 	end:
-	unlock_super(s);
 	return 0;
 }
 
-int hpfs_alloc_if_possible(struct super_block *s, secno sec)
-{
-	int r;
-	hpfs_lock_creation(s);
-	r = hpfs_alloc_if_possible_nolock(s, sec);
-	hpfs_unlock_creation(s);
-	return r;
-}
-
 /* Free sectors in bitmaps */
 
 void hpfs_free_sectors(struct super_block *s, secno sec, unsigned n)
 {
 	struct quad_buffer_head qbh;
-	unsigned *bmp;
+	u32 *bmp;
 	struct hpfs_sb_info *sbi = hpfs_sb(s);
 	/*printk("2 - ");*/
 	if (!n) return;
@@ -286,26 +262,22 @@ void hpfs_free_sectors(struct super_block *s, secno sec, unsigned n)
 		hpfs_error(s, "Trying to free reserved sector %08x", sec);
 		return;
 	}
-	lock_super(s);
 	sbi->sb_max_fwd_alloc += n > 0xffff ? 0xffff : n;
 	if (sbi->sb_max_fwd_alloc > 0xffffff) sbi->sb_max_fwd_alloc = 0xffffff;
 	new_map:
 	if (!(bmp = hpfs_map_bitmap(s, sec >> 14, &qbh, "free"))) {
-		unlock_super(s);
 		return;
 	}	
 	new_tst:
-	if ((bmp[(sec & 0x3fff) >> 5] >> (sec & 0x1f) & 1)) {
+	if ((le32_to_cpu(bmp[(sec & 0x3fff) >> 5]) >> (sec & 0x1f) & 1)) {
 		hpfs_error(s, "sector %08x not allocated", sec);
 		hpfs_brelse4(&qbh);
-		unlock_super(s);
 		return;
 	}
-	bmp[(sec & 0x3fff) >> 5] |= 1 << (sec & 0x1f);
+	bmp[(sec & 0x3fff) >> 5] |= cpu_to_le32(1 << (sec & 0x1f));
 	if (!--n) {
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
-		unlock_super(s);
 		return;
 	}	
 	if (!(++sec & 0x3fff)) {
@@ -327,13 +299,13 @@ int hpfs_check_free_dnodes(struct super_block *s, int n)
 	int n_bmps = (hpfs_sb(s)->sb_fs_size + 0x4000 - 1) >> 14;
 	int b = hpfs_sb(s)->sb_c_bitmap & 0x0fffffff;
 	int i, j;
-	unsigned *bmp;
+	u32 *bmp;
 	struct quad_buffer_head qbh;
 	if ((bmp = hpfs_map_dnode_bitmap(s, &qbh))) {
 		for (j = 0; j < 512; j++) {
 			unsigned k;
-			if (!bmp[j]) continue;
-			for (k = bmp[j]; k; k >>= 1) if (k & 1) if (!--n) {
+			if (!le32_to_cpu(bmp[j])) continue;
+			for (k = le32_to_cpu(bmp[j]); k; k >>= 1) if (k & 1) if (!--n) {
 				hpfs_brelse4(&qbh);
 				return 0;
 			}
@@ -352,10 +324,10 @@ int hpfs_check_free_dnodes(struct super_block *s, int n)
 	chk_bmp:
 	if (bmp) {
 		for (j = 0; j < 512; j++) {
-			unsigned k;
-			if (!bmp[j]) continue;
+			u32 k;
+			if (!le32_to_cpu(bmp[j])) continue;
 			for (k = 0xf; k; k <<= 4)
-				if ((bmp[j] & k) == k) {
+				if ((le32_to_cpu(bmp[j]) & k) == k) {
 					if (!--n) {
 						hpfs_brelse4(&qbh);
 						return 0;
@@ -379,44 +351,40 @@ void hpfs_free_dnode(struct super_block *s, dnode_secno dno)
 		hpfs_free_sectors(s, dno, 4);
 	} else {
 		struct quad_buffer_head qbh;
-		unsigned *bmp;
+		u32 *bmp;
 		unsigned ssec = (dno - hpfs_sb(s)->sb_dirband_start) / 4;
-		lock_super(s);
 		if (!(bmp = hpfs_map_dnode_bitmap(s, &qbh))) {
-			unlock_super(s);
 			return;
 		}
-		bmp[ssec >> 5] |= 1 << (ssec & 0x1f);
+		bmp[ssec >> 5] |= cpu_to_le32(1 << (ssec & 0x1f));
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
-		unlock_super(s);
 	}
 }
 
 struct dnode *hpfs_alloc_dnode(struct super_block *s, secno near,
-			 dnode_secno *dno, struct quad_buffer_head *qbh,
-			 int lock)
+			 dnode_secno *dno, struct quad_buffer_head *qbh)
 {
 	struct dnode *d;
 	if (hpfs_count_one_bitmap(s, hpfs_sb(s)->sb_dmap) > FREE_DNODES_ADD) {
-		if (!(*dno = alloc_in_dirband(s, near, lock)))
-			if (!(*dno = hpfs_alloc_sector(s, near, 4, 0, lock))) return NULL;
+		if (!(*dno = alloc_in_dirband(s, near)))
+			if (!(*dno = hpfs_alloc_sector(s, near, 4, 0))) return NULL;
 	} else {
-		if (!(*dno = hpfs_alloc_sector(s, near, 4, 0, lock)))
-			if (!(*dno = alloc_in_dirband(s, near, lock))) return NULL;
+		if (!(*dno = hpfs_alloc_sector(s, near, 4, 0)))
+			if (!(*dno = alloc_in_dirband(s, near))) return NULL;
 	}
 	if (!(d = hpfs_get_4sectors(s, *dno, qbh))) {
 		hpfs_free_dnode(s, *dno);
 		return NULL;
 	}
 	memset(d, 0, 2048);
-	d->magic = DNODE_MAGIC;
-	d->first_free = 52;
+	d->magic = cpu_to_le32(DNODE_MAGIC);
+	d->first_free = cpu_to_le32(52);
 	d->dirent[0] = 32;
 	d->dirent[2] = 8;
 	d->dirent[30] = 1;
 	d->dirent[31] = 255;
-	d->self = *dno;
+	d->self = cpu_to_le32(*dno);
 	return d;
 }
 
@@ -424,16 +392,16 @@ struct fnode *hpfs_alloc_fnode(struct super_block *s, secno near, fnode_secno *f
 			  struct buffer_head **bh)
 {
 	struct fnode *f;
-	if (!(*fno = hpfs_alloc_sector(s, near, 1, FNODE_ALLOC_FWD, 1))) return NULL;
+	if (!(*fno = hpfs_alloc_sector(s, near, 1, FNODE_ALLOC_FWD))) return NULL;
 	if (!(f = hpfs_get_sector(s, *fno, bh))) {
 		hpfs_free_sectors(s, *fno, 1);
 		return NULL;
 	}	
 	memset(f, 0, 512);
-	f->magic = FNODE_MAGIC;
-	f->ea_offs = 0xc4;
+	f->magic = cpu_to_le32(FNODE_MAGIC);
+	f->ea_offs = cpu_to_le16(0xc4);
 	f->btree.n_free_nodes = 8;
-	f->btree.first_free = 8;
+	f->btree.first_free = cpu_to_le16(8);
 	return f;
 }
 
@@ -441,16 +409,16 @@ struct anode *hpfs_alloc_anode(struct super_block *s, secno near, anode_secno *a
 			  struct buffer_head **bh)
 {
 	struct anode *a;
-	if (!(*ano = hpfs_alloc_sector(s, near, 1, ANODE_ALLOC_FWD, 1))) return NULL;
+	if (!(*ano = hpfs_alloc_sector(s, near, 1, ANODE_ALLOC_FWD))) return NULL;
 	if (!(a = hpfs_get_sector(s, *ano, bh))) {
 		hpfs_free_sectors(s, *ano, 1);
 		return NULL;
 	}
 	memset(a, 0, 512);
-	a->magic = ANODE_MAGIC;
-	a->self = *ano;
+	a->magic = cpu_to_le32(ANODE_MAGIC);
+	a->self = cpu_to_le32(*ano);
 	a->btree.n_free_nodes = 40;
 	a->btree.n_used_nodes = 0;
-	a->btree.first_free = 8;
+	a->btree.first_free = cpu_to_le16(8);
 	return a;
 }
diff --git a/fs/hpfs/anode.c b/fs/hpfs/anode.c
index 6a2f04bf3df..08b503e8ed2 100644
--- a/fs/hpfs/anode.c
+++ b/fs/hpfs/anode.c
@@ -22,8 +22,8 @@ secno hpfs_bplus_lookup(struct super_block *s, struct inode *inode,
 	if (hpfs_sb(s)->sb_chk) if (hpfs_stop_cycles(s, a, &c1, &c2, "hpfs_bplus_lookup")) return -1;
 	if (btree->internal) {
 		for (i = 0; i < btree->n_used_nodes; i++)
-			if (btree->u.internal[i].file_secno > sec) {
-				a = btree->u.internal[i].down;
+			if (le32_to_cpu(btree->u.internal[i].file_secno) > sec) {
+				a = le32_to_cpu(btree->u.internal[i].down);
 				brelse(bh);
 				if (!(anode = hpfs_map_anode(s, a, &bh))) return -1;
 				btree = &anode->btree;
@@ -34,18 +34,18 @@ secno hpfs_bplus_lookup(struct super_block *s, struct inode *inode,
 		return -1;
 	}
 	for (i = 0; i < btree->n_used_nodes; i++)
-		if (btree->u.external[i].file_secno <= sec &&
-		    btree->u.external[i].file_secno + btree->u.external[i].length > sec) {
-			a = btree->u.external[i].disk_secno + sec - btree->u.external[i].file_secno;
+		if (le32_to_cpu(btree->u.external[i].file_secno) <= sec &&
+		    le32_to_cpu(btree->u.external[i].file_secno) + le32_to_cpu(btree->u.external[i].length) > sec) {
+			a = le32_to_cpu(btree->u.external[i].disk_secno) + sec - le32_to_cpu(btree->u.external[i].file_secno);
 			if (hpfs_sb(s)->sb_chk) if (hpfs_chk_sectors(s, a, 1, "data")) {
 				brelse(bh);
 				return -1;
 			}
 			if (inode) {
 				struct hpfs_inode_info *hpfs_inode = hpfs_i(inode);
-				hpfs_inode->i_file_sec = btree->u.external[i].file_secno;
-				hpfs_inode->i_disk_sec = btree->u.external[i].disk_secno;
-				hpfs_inode->i_n_secs = btree->u.external[i].length;
+				hpfs_inode->i_file_sec = le32_to_cpu(btree->u.external[i].file_secno);
+				hpfs_inode->i_disk_sec = le32_to_cpu(btree->u.external[i].disk_secno);
+				hpfs_inode->i_n_secs = le32_to_cpu(btree->u.external[i].length);
 			}
 			brelse(bh);
 			return a;
@@ -83,8 +83,8 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi
 		return -1;
 	}
 	if (btree->internal) {
-		a = btree->u.internal[n].down;
-		btree->u.internal[n].file_secno = -1;
+		a = le32_to_cpu(btree->u.internal[n].down);
+		btree->u.internal[n].file_secno = cpu_to_le32(-1);
 		mark_buffer_dirty(bh);
 		brelse(bh);
 		if (hpfs_sb(s)->sb_chk)
@@ -94,15 +94,15 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi
 		goto go_down;
 	}
 	if (n >= 0) {
-		if (btree->u.external[n].file_secno + btree->u.external[n].length != fsecno) {
+		if (le32_to_cpu(btree->u.external[n].file_secno) + le32_to_cpu(btree->u.external[n].length) != fsecno) {
 			hpfs_error(s, "allocated size %08x, trying to add sector %08x, %cnode %08x",
-				btree->u.external[n].file_secno + btree->u.external[n].length, fsecno,
+				le32_to_cpu(btree->u.external[n].file_secno) + le32_to_cpu(btree->u.external[n].length), fsecno,
 				fnod?'f':'a', node);
 			brelse(bh);
 			return -1;
 		}
-		if (hpfs_alloc_if_possible(s, se = btree->u.external[n].disk_secno + btree->u.external[n].length)) {
-			btree->u.external[n].length++;
+		if (hpfs_alloc_if_possible(s, se = le32_to_cpu(btree->u.external[n].disk_secno) + le32_to_cpu(btree->u.external[n].length))) {
+			btree->u.external[n].length = cpu_to_le32(le32_to_cpu(btree->u.external[n].length) + 1);
 			mark_buffer_dirty(bh);
 			brelse(bh);
 			return se;
@@ -115,20 +115,20 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi
 		}
 		se = !fnod ? node : (node + 16384) & ~16383;
 	}	
-	if (!(se = hpfs_alloc_sector(s, se, 1, fsecno*ALLOC_M>ALLOC_FWD_MAX ? ALLOC_FWD_MAX : fsecno*ALLOC_M<ALLOC_FWD_MIN ? ALLOC_FWD_MIN : fsecno*ALLOC_M, 1))) {
+	if (!(se = hpfs_alloc_sector(s, se, 1, fsecno*ALLOC_M>ALLOC_FWD_MAX ? ALLOC_FWD_MAX : fsecno*ALLOC_M<ALLOC_FWD_MIN ? ALLOC_FWD_MIN : fsecno*ALLOC_M))) {
 		brelse(bh);
 		return -1;
 	}
-	fs = n < 0 ? 0 : btree->u.external[n].file_secno + btree->u.external[n].length;
+	fs = n < 0 ? 0 : le32_to_cpu(btree->u.external[n].file_secno) + le32_to_cpu(btree->u.external[n].length);
 	if (!btree->n_free_nodes) {
-		up = a != node ? anode->up : -1;
+		up = a != node ? le32_to_cpu(anode->up) : -1;
 		if (!(anode = hpfs_alloc_anode(s, a, &na, &bh1))) {
 			brelse(bh);
 			hpfs_free_sectors(s, se, 1);
 			return -1;
 		}
 		if (a == node && fnod) {
-			anode->up = node;
+			anode->up = cpu_to_le32(node);
 			anode->btree.fnode_parent = 1;
 			anode->btree.n_used_nodes = btree->n_used_nodes;
 			anode->btree.first_free = btree->first_free;
@@ -137,9 +137,9 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi
 			btree->internal = 1;
 			btree->n_free_nodes = 11;
 			btree->n_used_nodes = 1;
-			btree->first_free = (char *)&(btree->u.internal[1]) - (char *)btree;
-			btree->u.internal[0].file_secno = -1;
-			btree->u.internal[0].down = na;
+			btree->first_free = cpu_to_le16((char *)&(btree->u.internal[1]) - (char *)btree);
+			btree->u.internal[0].file_secno = cpu_to_le32(-1);
+			btree->u.internal[0].down = cpu_to_le32(na);
 			mark_buffer_dirty(bh);
 		} else if (!(ranode = hpfs_alloc_anode(s, /*a*/0, &ra, &bh2))) {
 			brelse(bh);
@@ -153,15 +153,15 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi
 		btree = &anode->btree;
 	}
 	btree->n_free_nodes--; n = btree->n_used_nodes++;
-	btree->first_free += 12;
-	btree->u.external[n].disk_secno = se;
-	btree->u.external[n].file_secno = fs;
-	btree->u.external[n].length = 1;
+	btree->first_free = cpu_to_le16(le16_to_cpu(btree->first_free) + 12);
+	btree->u.external[n].disk_secno = cpu_to_le32(se);
+	btree->u.external[n].file_secno = cpu_to_le32(fs);
+	btree->u.external[n].length = cpu_to_le32(1);
 	mark_buffer_dirty(bh);
 	brelse(bh);
 	if ((a == node && fnod) || na == -1) return se;
 	c2 = 0;
-	while (up != -1) {
+	while (up != (anode_secno)-1) {
 		struct anode *new_anode;
 		if (hpfs_sb(s)->sb_chk)
 			if (hpfs_stop_cycles(s, up, &c1, &c2, "hpfs_add_sector_to_btree #2")) return -1;
@@ -174,47 +174,47 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi
 		}
 		if (btree->n_free_nodes) {
 			btree->n_free_nodes--; n = btree->n_used_nodes++;
-			btree->first_free += 8;
-			btree->u.internal[n].file_secno = -1;
-			btree->u.internal[n].down = na;
-			btree->u.internal[n-1].file_secno = fs;
+			btree->first_free = cpu_to_le16(le16_to_cpu(btree->first_free) + 8);
+			btree->u.internal[n].file_secno = cpu_to_le32(-1);
+			btree->u.internal[n].down = cpu_to_le32(na);
+			btree->u.internal[n-1].file_secno = cpu_to_le32(fs);
 			mark_buffer_dirty(bh);
 			brelse(bh);
 			brelse(bh2);
 			hpfs_free_sectors(s, ra, 1);
 			if ((anode = hpfs_map_anode(s, na, &bh))) {
-				anode->up = up;
+				anode->up = cpu_to_le32(up);
 				anode->btree.fnode_parent = up == node && fnod;
 				mark_buffer_dirty(bh);
 				brelse(bh);
 			}
 			return se;
 		}
-		up = up != node ? anode->up : -1;
-		btree->u.internal[btree->n_used_nodes - 1].file_secno = /*fs*/-1;
+		up = up != node ? le32_to_cpu(anode->up) : -1;
+		btree->u.internal[btree->n_used_nodes - 1].file_secno = cpu_to_le32(/*fs*/-1);
 		mark_buffer_dirty(bh);
 		brelse(bh);
 		a = na;
 		if ((new_anode = hpfs_alloc_anode(s, a, &na, &bh))) {
 			anode = new_anode;
-			/*anode->up = up != -1 ? up : ra;*/
+			/*anode->up = cpu_to_le32(up != -1 ? up : ra);*/
 			anode->btree.internal = 1;
 			anode->btree.n_used_nodes = 1;
 			anode->btree.n_free_nodes = 59;
-			anode->btree.first_free = 16;
-			anode->btree.u.internal[0].down = a;
-			anode->btree.u.internal[0].file_secno = -1;
+			anode->btree.first_free = cpu_to_le16(16);
+			anode->btree.u.internal[0].down = cpu_to_le32(a);
+			anode->btree.u.internal[0].file_secno = cpu_to_le32(-1);
 			mark_buffer_dirty(bh);
 			brelse(bh);
 			if ((anode = hpfs_map_anode(s, a, &bh))) {
-				anode->up = na;
+				anode->up = cpu_to_le32(na);
 				mark_buffer_dirty(bh);
 				brelse(bh);
 			}
 		} else na = a;
 	}
 	if ((anode = hpfs_map_anode(s, na, &bh))) {
-		anode->up = node;
+		anode->up = cpu_to_le32(node);
 		if (fnod) anode->btree.fnode_parent = 1;
 		mark_buffer_dirty(bh);
 		brelse(bh);
@@ -232,14 +232,14 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi
 		}
 		btree = &fnode->btree;
 	}
-	ranode->up = node;
-	memcpy(&ranode->btree, btree, btree->first_free);
+	ranode->up = cpu_to_le32(node);
+	memcpy(&ranode->btree, btree, le16_to_cpu(btree->first_free));
 	if (fnod) ranode->btree.fnode_parent = 1;
 	ranode->btree.n_free_nodes = (ranode->btree.internal ? 60 : 40) - ranode->btree.n_used_nodes;
 	if (ranode->btree.internal) for (n = 0; n < ranode->btree.n_used_nodes; n++) {
 		struct anode *unode;
-		if ((unode = hpfs_map_anode(s, ranode->u.internal[n].down, &bh1))) {
-			unode->up = ra;
+		if ((unode = hpfs_map_anode(s, le32_to_cpu(ranode->u.internal[n].down), &bh1))) {
+			unode->up = cpu_to_le32(ra);
 			unode->btree.fnode_parent = 0;
 			mark_buffer_dirty(bh1);
 			brelse(bh1);
@@ -248,11 +248,11 @@ secno hpfs_add_sector_to_btree(struct super_block *s, secno node, int fnod, unsi
 	btree->internal = 1;
 	btree->n_free_nodes = fnod ? 10 : 58;
 	btree->n_used_nodes = 2;
-	btree->first_free = (char *)&btree->u.internal[2] - (char *)btree;
-	btree->u.internal[0].file_secno = fs;
-	btree->u.internal[0].down = ra;
-	btree->u.internal[1].file_secno = -1;
-	btree->u.internal[1].down = na;
+	btree->first_free = cpu_to_le16((char *)&btree->u.internal[2] - (char *)btree);
+	btree->u.internal[0].file_secno = cpu_to_le32(fs);
+	btree->u.internal[0].down = cpu_to_le32(ra);
+	btree->u.internal[1].file_secno = cpu_to_le32(-1);
+	btree->u.internal[1].down = cpu_to_le32(na);
 	mark_buffer_dirty(bh);
 	brelse(bh);
 	mark_buffer_dirty(bh2);
@@ -279,7 +279,7 @@ void hpfs_remove_btree(struct super_block *s, struct bplus_header *btree)
 	go_down:
 	d2 = 0;
 	while (btree1->internal) {
-		ano = btree1->u.internal[pos].down;
+		ano = le32_to_cpu(btree1->u.internal[pos].down);
 		if (level) brelse(bh);
 		if (hpfs_sb(s)->sb_chk)
 			if (hpfs_stop_cycles(s, ano, &d1, &d2, "hpfs_remove_btree #1"))
@@ -290,7 +290,7 @@ void hpfs_remove_btree(struct super_block *s, struct bplus_header *btree)
 		pos = 0;
 	}
 	for (i = 0; i < btree1->n_used_nodes; i++)
-		hpfs_free_sectors(s, btree1->u.external[i].disk_secno, btree1->u.external[i].length);
+		hpfs_free_sectors(s, le32_to_cpu(btree1->u.external[i].disk_secno), le32_to_cpu(btree1->u.external[i].length));
 	go_up:
 	if (!level) return;
 	brelse(bh);
@@ -298,13 +298,13 @@ void hpfs_remove_btree(struct super_block *s, struct bplus_header *btree)
 		if (hpfs_stop_cycles(s, ano, &c1, &c2, "hpfs_remove_btree #2")) return;
 	hpfs_free_sectors(s, ano, 1);
 	oano = ano;
-	ano = anode->up;
+	ano = le32_to_cpu(anode->up);
 	if (--level) {
 		if (!(anode = hpfs_map_anode(s, ano, &bh))) return;
 		btree1 = &anode->btree;
 	} else btree1 = btree;
 	for (i = 0; i < btree1->n_used_nodes; i++) {
-		if (btree1->u.internal[i].down == oano) {
+		if (le32_to_cpu(btree1->u.internal[i].down) == oano) {
 			if ((pos = i + 1) < btree1->n_used_nodes)
 				goto go_down;
 			else
@@ -411,7 +411,7 @@ void hpfs_truncate_btree(struct super_block *s, secno f, int fno, unsigned secs)
 		if (fno) {
 			btree->n_free_nodes = 8;
 			btree->n_used_nodes = 0;
-			btree->first_free = 8;
+			btree->first_free = cpu_to_le16(8);
 			btree->internal = 0;
 			mark_buffer_dirty(bh);
 		} else hpfs_free_sectors(s, f, 1);
@@ -421,22 +421,22 @@ void hpfs_truncate_btree(struct super_block *s, secno f, int fno, unsigned secs)
 	while (btree->internal) {
 		nodes = btree->n_used_nodes + btree->n_free_nodes;
 		for (i = 0; i < btree->n_used_nodes; i++)
-			if (btree->u.internal[i].file_secno >= secs) goto f;
+			if (le32_to_cpu(btree->u.internal[i].file_secno) >= secs) goto f;
 		brelse(bh);
 		hpfs_error(s, "internal btree %08x doesn't end with -1", node);
 		return;
 		f:
 		for (j = i + 1; j < btree->n_used_nodes; j++)
-			hpfs_ea_remove(s, btree->u.internal[j].down, 1, 0);
+			hpfs_ea_remove(s, le32_to_cpu(btree->u.internal[j].down), 1, 0);
 		btree->n_used_nodes = i + 1;
 		btree->n_free_nodes = nodes - btree->n_used_nodes;
-		btree->first_free = 8 + 8 * btree->n_used_nodes;
+		btree->first_free = cpu_to_le16(8 + 8 * btree->n_used_nodes);
 		mark_buffer_dirty(bh);
-		if (btree->u.internal[i].file_secno == secs) {
+		if (btree->u.internal[i].file_secno == cpu_to_le32(secs)) {
 			brelse(bh);
 			return;
 		}
-		node = btree->u.internal[i].down;
+		node = le32_to_cpu(btree->u.internal[i].down);
 		brelse(bh);
 		if (hpfs_sb(s)->sb_chk)
 			if (hpfs_stop_cycles(s, node, &c1, &c2, "hpfs_truncate_btree"))
@@ -446,25 +446,25 @@ void hpfs_truncate_btree(struct super_block *s, secno f, int fno, unsigned secs)
 	}	
 	nodes = btree->n_used_nodes + btree->n_free_nodes;
 	for (i = 0; i < btree->n_used_nodes; i++)
-		if (btree->u.external[i].file_secno + btree->u.external[i].length >= secs) goto ff;
+		if (le32_to_cpu(btree->u.external[i].file_secno) + le32_to_cpu(btree->u.external[i].length) >= secs) goto ff;
 	brelse(bh);
 	return;
 	ff:
-	if (secs <= btree->u.external[i].file_secno) {
+	if (secs <= le32_to_cpu(btree->u.external[i].file_secno)) {
 		hpfs_error(s, "there is an allocation error in file %08x, sector %08x", f, secs);
 		if (i) i--;
 	}
-	else if (btree->u.external[i].file_secno + btree->u.external[i].length > secs) {
-		hpfs_free_sectors(s, btree->u.external[i].disk_secno + secs -
-			btree->u.external[i].file_secno, btree->u.external[i].length
-			- secs + btree->u.external[i].file_secno); /* I hope gcc optimizes this :-) */
-		btree->u.external[i].length = secs - btree->u.external[i].file_secno;
+	else if (le32_to_cpu(btree->u.external[i].file_secno) + le32_to_cpu(btree->u.external[i].length) > secs) {
+		hpfs_free_sectors(s, le32_to_cpu(btree->u.external[i].disk_secno) + secs -
+			le32_to_cpu(btree->u.external[i].file_secno), le32_to_cpu(btree->u.external[i].length)
+			- secs + le32_to_cpu(btree->u.external[i].file_secno)); /* I hope gcc optimizes this :-) */
+		btree->u.external[i].length = cpu_to_le32(secs - le32_to_cpu(btree->u.external[i].file_secno));
 	}
 	for (j = i + 1; j < btree->n_used_nodes; j++)
-		hpfs_free_sectors(s, btree->u.external[j].disk_secno, btree->u.external[j].length);
+		hpfs_free_sectors(s, le32_to_cpu(btree->u.external[j].disk_secno), le32_to_cpu(btree->u.external[j].length));
 	btree->n_used_nodes = i + 1;
 	btree->n_free_nodes = nodes - btree->n_used_nodes;
-	btree->first_free = 8 + 12 * btree->n_used_nodes;
+	btree->first_free = cpu_to_le16(8 + 12 * btree->n_used_nodes);
 	mark_buffer_dirty(bh);
 	brelse(bh);
 }
@@ -480,12 +480,12 @@ void hpfs_remove_fnode(struct super_block *s, fnode_secno fno)
 	struct extended_attribute *ea_end;
 	if (!(fnode = hpfs_map_fnode(s, fno, &bh))) return;
 	if (!fnode->dirflag) hpfs_remove_btree(s, &fnode->btree);
-	else hpfs_remove_dtree(s, fnode->u.external[0].disk_secno);
+	else hpfs_remove_dtree(s, le32_to_cpu(fnode->u.external[0].disk_secno));
 	ea_end = fnode_end_ea(fnode);
 	for (ea = fnode_ea(fnode); ea < ea_end; ea = next_ea(ea))
 		if (ea->indirect)
 			hpfs_ea_remove(s, ea_sec(ea), ea->anode, ea_len(ea));
-	hpfs_ea_ext_remove(s, fnode->ea_secno, fnode->ea_anode, fnode->ea_size_l);
+	hpfs_ea_ext_remove(s, le32_to_cpu(fnode->ea_secno), fnode->ea_anode, le32_to_cpu(fnode->ea_size_l));
 	brelse(bh);
 	hpfs_free_sectors(s, fno, 1);
 }
diff --git a/fs/hpfs/buffer.c b/fs/hpfs/buffer.c
index 793cb9d943d..9ecde27d1e2 100644
--- a/fs/hpfs/buffer.c
+++ b/fs/hpfs/buffer.c
@@ -9,22 +9,6 @@
 #include <linux/slab.h>
 #include "hpfs_fn.h"
 
-void hpfs_lock_creation(struct super_block *s)
-{
-#ifdef DEBUG_LOCKS
-	printk("lock creation\n");
-#endif
-	mutex_lock(&hpfs_sb(s)->hpfs_creation_de);
-}
-
-void hpfs_unlock_creation(struct super_block *s)
-{
-#ifdef DEBUG_LOCKS
-	printk("unlock creation\n");
-#endif
-	mutex_unlock(&hpfs_sb(s)->hpfs_creation_de);
-}
-
 /* Map a sector into a buffer and return pointers to it and to the buffer. */
 
 void *hpfs_map_sector(struct super_block *s, unsigned secno, struct buffer_head **bhp,
@@ -32,6 +16,8 @@ void *hpfs_map_sector(struct super_block *s, unsigned secno, struct buffer_head
 {
 	struct buffer_head *bh;
 
+	hpfs_lock_assert(s);
+
 	cond_resched();
 
 	*bhp = bh = sb_bread(s, secno);
@@ -50,6 +36,8 @@ void *hpfs_get_sector(struct super_block *s, unsigned secno, struct buffer_head
 	struct buffer_head *bh;
 	/*return hpfs_map_sector(s, secno, bhp, 0);*/
 
+	hpfs_lock_assert(s);
+
 	cond_resched();
 
 	if ((*bhp = bh = sb_getblk(s, secno)) != NULL) {
@@ -70,6 +58,8 @@ void *hpfs_map_4sectors(struct super_block *s, unsigned secno, struct quad_buffe
 	struct buffer_head *bh;
 	char *data;
 
+	hpfs_lock_assert(s);
+
 	cond_resched();
 
 	if (secno & 3) {
@@ -125,6 +115,8 @@ void *hpfs_get_4sectors(struct super_block *s, unsigned secno,
 {
 	cond_resched();
 
+	hpfs_lock_assert(s);
+
 	if (secno & 3) {
 		printk("HPFS: hpfs_get_4sectors: unaligned read\n");
 		return NULL;
diff --git a/fs/hpfs/dir.c b/fs/hpfs/dir.c
index b3d7c0ddb60..f46ae025bfb 100644
--- a/fs/hpfs/dir.c
+++ b/fs/hpfs/dir.c
@@ -88,9 +88,9 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 			hpfs_error(inode->i_sb, "not a directory, fnode %08lx",
 					(unsigned long)inode->i_ino);
 		}
-		if (hpfs_inode->i_dno != fno->u.external[0].disk_secno) {
+		if (hpfs_inode->i_dno != le32_to_cpu(fno->u.external[0].disk_secno)) {
 			e = 1;
-			hpfs_error(inode->i_sb, "corrupted inode: i_dno == %08x, fnode -> dnode == %08x", hpfs_inode->i_dno, fno->u.external[0].disk_secno);
+			hpfs_error(inode->i_sb, "corrupted inode: i_dno == %08x, fnode -> dnode == %08x", hpfs_inode->i_dno, le32_to_cpu(fno->u.external[0].disk_secno));
 		}
 		brelse(bh);
 		if (e) {
@@ -156,7 +156,7 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 			goto again;
 		}
 		tempname = hpfs_translate_name(inode->i_sb, de->name, de->namelen, lc, de->not_8x3);
-		if (filldir(dirent, tempname, de->namelen, old_pos, de->fnode, DT_UNKNOWN) < 0) {
+		if (filldir(dirent, tempname, de->namelen, old_pos, le32_to_cpu(de->fnode), DT_UNKNOWN) < 0) {
 			filp->f_pos = old_pos;
 			if (tempname != de->name) kfree(tempname);
 			hpfs_brelse4(&qbh);
@@ -221,7 +221,7 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name
 	 * Get inode number, what we're after.
 	 */
 
-	ino = de->fnode;
+	ino = le32_to_cpu(de->fnode);
 
 	/*
 	 * Go find or make an inode.
@@ -236,7 +236,7 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name
 		hpfs_init_inode(result);
 		if (de->directory)
 			hpfs_read_inode(result);
-		else if (de->ea_size && hpfs_sb(dir->i_sb)->sb_eas)
+		else if (le32_to_cpu(de->ea_size) && hpfs_sb(dir->i_sb)->sb_eas)
 			hpfs_read_inode(result);
 		else {
 			result->i_mode |= S_IFREG;
@@ -250,8 +250,6 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name
 	hpfs_result = hpfs_i(result);
 	if (!de->directory) hpfs_result->i_parent_dir = dir->i_ino;
 
-	hpfs_decide_conv(result, name, len);
-
 	if (de->has_acl || de->has_xtd_perm) if (!(dir->i_sb->s_flags & MS_RDONLY)) {
 		hpfs_error(result->i_sb, "ACLs or XPERM found. This is probably HPFS386. This driver doesn't support it now. Send me some info on these structures");
 		goto bail1;
@@ -263,19 +261,19 @@ struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct name
 	 */
 
 	if (!result->i_ctime.tv_sec) {
-		if (!(result->i_ctime.tv_sec = local_to_gmt(dir->i_sb, de->creation_date)))
+		if (!(result->i_ctime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->creation_date))))
 			result->i_ctime.tv_sec = 1;
 		result->i_ctime.tv_nsec = 0;
-		result->i_mtime.tv_sec = local_to_gmt(dir->i_sb, de->write_date);
+		result->i_mtime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->write_date));
 		result->i_mtime.tv_nsec = 0;
-		result->i_atime.tv_sec = local_to_gmt(dir->i_sb, de->read_date);
+		result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(de->read_date));
 		result->i_atime.tv_nsec = 0;
-		hpfs_result->i_ea_size = de->ea_size;
+		hpfs_result->i_ea_size = le32_to_cpu(de->ea_size);
 		if (!hpfs_result->i_ea_mode && de->read_only)
 			result->i_mode &= ~0222;
 		if (!de->directory) {
 			if (result->i_size == -1) {
-				result->i_size = de->file_size;
+				result->i_size = le32_to_cpu(de->file_size);
 				result->i_data.a_ops = &hpfs_aops;
 				hpfs_i(result)->mmu_private = result->i_size;
 			/*
diff --git a/fs/hpfs/dnode.c b/fs/hpfs/dnode.c
index 9b2ffadfc8c..1e0e2ac30fd 100644
--- a/fs/hpfs/dnode.c
+++ b/fs/hpfs/dnode.c
@@ -14,11 +14,11 @@ static loff_t get_pos(struct dnode *d, struct hpfs_dirent *fde)
 	struct hpfs_dirent *de_end = dnode_end_de(d);
 	int i = 1;
 	for (de = dnode_first_de(d); de < de_end; de = de_next_de(de)) {
-		if (de == fde) return ((loff_t) d->self << 4) | (loff_t)i;
+		if (de == fde) return ((loff_t) le32_to_cpu(d->self) << 4) | (loff_t)i;
 		i++;
 	}
 	printk("HPFS: get_pos: not_found\n");
-	return ((loff_t)d->self << 4) | (loff_t)1;
+	return ((loff_t)le32_to_cpu(d->self) << 4) | (loff_t)1;
 }
 
 void hpfs_add_pos(struct inode *inode, loff_t *pos)
@@ -130,29 +130,30 @@ static void set_last_pointer(struct super_block *s, struct dnode *d, dnode_secno
 {
 	struct hpfs_dirent *de;
 	if (!(de = dnode_last_de(d))) {
-		hpfs_error(s, "set_last_pointer: empty dnode %08x", d->self);
+		hpfs_error(s, "set_last_pointer: empty dnode %08x", le32_to_cpu(d->self));
 		return;
 	}
 	if (hpfs_sb(s)->sb_chk) {
 		if (de->down) {
 			hpfs_error(s, "set_last_pointer: dnode %08x has already last pointer %08x",
-				d->self, de_down_pointer(de));
+				le32_to_cpu(d->self), de_down_pointer(de));
 			return;
 		}
-		if (de->length != 32) {
-			hpfs_error(s, "set_last_pointer: bad last dirent in dnode %08x", d->self);
+		if (le16_to_cpu(de->length) != 32) {
+			hpfs_error(s, "set_last_pointer: bad last dirent in dnode %08x", le32_to_cpu(d->self));
 			return;
 		}
 	}
 	if (ptr) {
-		if ((d->first_free += 4) > 2048) {
-			hpfs_error(s,"set_last_pointer: too long dnode %08x", d->self);
-			d->first_free -= 4;
+		d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) + 4);
+		if (le32_to_cpu(d->first_free) > 2048) {
+			hpfs_error(s, "set_last_pointer: too long dnode %08x", le32_to_cpu(d->self));
+			d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) - 4);
 			return;
 		}
-		de->length = 36;
+		de->length = cpu_to_le16(36);
 		de->down = 1;
-		*(dnode_secno *)((char *)de + 32) = ptr;
+		*(dnode_secno *)((char *)de + 32) = cpu_to_le32(ptr);
 	}
 }
 
@@ -168,7 +169,7 @@ struct hpfs_dirent *hpfs_add_de(struct super_block *s, struct dnode *d,
 	for (de = dnode_first_de(d); de < de_end; de = de_next_de(de)) {
 		int c = hpfs_compare_names(s, name, namelen, de->name, de->namelen, de->last);
 		if (!c) {
-			hpfs_error(s, "name (%c,%d) already exists in dnode %08x", *name, namelen, d->self);
+			hpfs_error(s, "name (%c,%d) already exists in dnode %08x", *name, namelen, le32_to_cpu(d->self));
 			return NULL;
 		}
 		if (c < 0) break;
@@ -176,15 +177,14 @@ struct hpfs_dirent *hpfs_add_de(struct super_block *s, struct dnode *d,
 	memmove((char *)de + d_size, de, (char *)de_end - (char *)de);
 	memset(de, 0, d_size);
 	if (down_ptr) {
-		*(int *)((char *)de + d_size - 4) = down_ptr;
+		*(dnode_secno *)((char *)de + d_size - 4) = cpu_to_le32(down_ptr);
 		de->down = 1;
 	}
-	de->length = d_size;
-	if (down_ptr) de->down = 1;
+	de->length = cpu_to_le16(d_size);
 	de->not_8x3 = hpfs_is_name_long(name, namelen);
 	de->namelen = namelen;
 	memcpy(de->name, name, namelen);
-	d->first_free += d_size;
+	d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) + d_size);
 	return de;
 }
 
@@ -194,25 +194,25 @@ static void hpfs_delete_de(struct super_block *s, struct dnode *d,
 			   struct hpfs_dirent *de)
 {
 	if (de->last) {
-		hpfs_error(s, "attempt to delete last dirent in dnode %08x", d->self);
+		hpfs_error(s, "attempt to delete last dirent in dnode %08x", le32_to_cpu(d->self));
 		return;
 	}
-	d->first_free -= de->length;
-	memmove(de, de_next_de(de), d->first_free + (char *)d - (char *)de);
+	d->first_free = cpu_to_le32(le32_to_cpu(d->first_free) - le16_to_cpu(de->length));
+	memmove(de, de_next_de(de), le32_to_cpu(d->first_free) + (char *)d - (char *)de);
 }
 
 static void fix_up_ptrs(struct super_block *s, struct dnode *d)
 {
 	struct hpfs_dirent *de;
 	struct hpfs_dirent *de_end = dnode_end_de(d);
-	dnode_secno dno = d->self;
+	dnode_secno dno = le32_to_cpu(d->self);
 	for (de = dnode_first_de(d); de < de_end; de = de_next_de(de))
 		if (de->down) {
 			struct quad_buffer_head qbh;
 			struct dnode *dd;
 			if ((dd = hpfs_map_dnode(s, de_down_pointer(de), &qbh))) {
-				if (dd->up != dno || dd->root_dnode) {
-					dd->up = dno;
+				if (le32_to_cpu(dd->up) != dno || dd->root_dnode) {
+					dd->up = cpu_to_le32(dno);
 					dd->root_dnode = 0;
 					hpfs_mark_4buffers_dirty(&qbh);
 				}
@@ -262,7 +262,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
 			kfree(nname);
 			return 1;
 		}
-	if (d->first_free + de_size(namelen, down_ptr) <= 2048) {
+	if (le32_to_cpu(d->first_free) + de_size(namelen, down_ptr) <= 2048) {
 		loff_t t;
 		copy_de(de=hpfs_add_de(i->i_sb, d, name, namelen, down_ptr), new_de);
 		t = get_pos(d, de);
@@ -286,11 +286,11 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
 		kfree(nname);
 		return 1;
 	}	
-	memcpy(nd, d, d->first_free);
+	memcpy(nd, d, le32_to_cpu(d->first_free));
 	copy_de(de = hpfs_add_de(i->i_sb, nd, name, namelen, down_ptr), new_de);
 	for_all_poss(i, hpfs_pos_ins, get_pos(nd, de), 1);
 	h = ((char *)dnode_last_de(nd) - (char *)nd) / 2 + 10;
-	if (!(ad = hpfs_alloc_dnode(i->i_sb, d->up, &adno, &qbh1, 0))) {
+	if (!(ad = hpfs_alloc_dnode(i->i_sb, le32_to_cpu(d->up), &adno, &qbh1))) {
 		hpfs_error(i->i_sb, "unable to alloc dnode - dnode tree will be corrupted");
 		hpfs_brelse4(&qbh);
 		kfree(nd);
@@ -313,20 +313,21 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
 	down_ptr = adno;
 	set_last_pointer(i->i_sb, ad, de->down ? de_down_pointer(de) : 0);
 	de = de_next_de(de);
-	memmove((char *)nd + 20, de, nd->first_free + (char *)nd - (char *)de);
-	nd->first_free -= (char *)de - (char *)nd - 20;
-	memcpy(d, nd, nd->first_free);
+	memmove((char *)nd + 20, de, le32_to_cpu(nd->first_free) + (char *)nd - (char *)de);
+	nd->first_free = cpu_to_le32(le32_to_cpu(nd->first_free) - ((char *)de - (char *)nd - 20));
+	memcpy(d, nd, le32_to_cpu(nd->first_free));
 	for_all_poss(i, hpfs_pos_del, (loff_t)dno << 4, pos);
 	fix_up_ptrs(i->i_sb, ad);
 	if (!d->root_dnode) {
-		dno = ad->up = d->up;
+		ad->up = d->up;
+		dno = le32_to_cpu(ad->up);
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
 		hpfs_mark_4buffers_dirty(&qbh1);
 		hpfs_brelse4(&qbh1);
 		goto go_up;
 	}
-	if (!(rd = hpfs_alloc_dnode(i->i_sb, d->up, &rdno, &qbh2, 0))) {
+	if (!(rd = hpfs_alloc_dnode(i->i_sb, le32_to_cpu(d->up), &rdno, &qbh2))) {
 		hpfs_error(i->i_sb, "unable to alloc dnode - dnode tree will be corrupted");
 		hpfs_brelse4(&qbh);
 		hpfs_brelse4(&qbh1);
@@ -338,7 +339,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
 	i->i_blocks += 4;
 	rd->root_dnode = 1;
 	rd->up = d->up;
-	if (!(fnode = hpfs_map_fnode(i->i_sb, d->up, &bh))) {
+	if (!(fnode = hpfs_map_fnode(i->i_sb, le32_to_cpu(d->up), &bh))) {
 		hpfs_free_dnode(i->i_sb, rdno);
 		hpfs_brelse4(&qbh);
 		hpfs_brelse4(&qbh1);
@@ -347,10 +348,11 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
 		kfree(nname);
 		return 1;
 	}
-	fnode->u.external[0].disk_secno = rdno;
+	fnode->u.external[0].disk_secno = cpu_to_le32(rdno);
 	mark_buffer_dirty(bh);
 	brelse(bh);
-	d->up = ad->up = hpfs_i(i)->i_dno = rdno;
+	hpfs_i(i)->i_dno = rdno;
+	d->up = ad->up = cpu_to_le32(rdno);
 	d->root_dnode = ad->root_dnode = 0;
 	hpfs_mark_4buffers_dirty(&qbh);
 	hpfs_brelse4(&qbh);
@@ -373,7 +375,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno,
 
 int hpfs_add_dirent(struct inode *i,
 		    const unsigned char *name, unsigned namelen,
-		    struct hpfs_dirent *new_de, int cdepth)
+		    struct hpfs_dirent *new_de)
 {
 	struct hpfs_inode_info *hpfs_inode = hpfs_i(i);
 	struct dnode *d;
@@ -403,7 +405,6 @@ int hpfs_add_dirent(struct inode *i,
 		}
 	}
 	hpfs_brelse4(&qbh);
-	if (!cdepth) hpfs_lock_creation(i->i_sb);
 	if (hpfs_check_free_dnodes(i->i_sb, FREE_DNODES_ADD)) {
 		c = 1;
 		goto ret;
@@ -411,7 +412,6 @@ int hpfs_add_dirent(struct inode *i,
 	i->i_version++;
 	c = hpfs_add_to_dnode(i, dno, name, namelen, new_de, 0);
 	ret:
-	if (!cdepth) hpfs_unlock_creation(i->i_sb);
 	return c;
 }
 
@@ -437,9 +437,9 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to)
 				return 0;
 		if (!(dnode = hpfs_map_dnode(i->i_sb, dno, &qbh))) return 0;
 		if (hpfs_sb(i->i_sb)->sb_chk) {
-			if (dnode->up != chk_up) {
+			if (le32_to_cpu(dnode->up) != chk_up) {
 				hpfs_error(i->i_sb, "move_to_top: up pointer from %08x should be %08x, is %08x",
-					dno, chk_up, dnode->up);
+					dno, chk_up, le32_to_cpu(dnode->up));
 				hpfs_brelse4(&qbh);
 				return 0;
 			}
@@ -455,7 +455,7 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to)
 		hpfs_brelse4(&qbh);
 	}
 	while (!(de = dnode_pre_last_de(dnode))) {
-		dnode_secno up = dnode->up;
+		dnode_secno up = le32_to_cpu(dnode->up);
 		hpfs_brelse4(&qbh);
 		hpfs_free_dnode(i->i_sb, dno);
 		i->i_size -= 2048;
@@ -474,8 +474,8 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to)
 			hpfs_brelse4(&qbh);
 			return 0;
 		}
-		dnode->first_free -= 4;
-		de->length -= 4;
+		dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) - 4);
+		de->length = cpu_to_le16(le16_to_cpu(de->length) - 4);
 		de->down = 0;
 		hpfs_mark_4buffers_dirty(&qbh);
 		dno = up;
@@ -483,12 +483,12 @@ static secno move_to_top(struct inode *i, dnode_secno from, dnode_secno to)
 	t = get_pos(dnode, de);
 	for_all_poss(i, hpfs_pos_subst, t, 4);
 	for_all_poss(i, hpfs_pos_subst, t + 1, 5);
-	if (!(nde = kmalloc(de->length, GFP_NOFS))) {
+	if (!(nde = kmalloc(le16_to_cpu(de->length), GFP_NOFS))) {
 		hpfs_error(i->i_sb, "out of memory for dirent - directory will be corrupted");
 		hpfs_brelse4(&qbh);
 		return 0;
 	}
-	memcpy(nde, de, de->length);
+	memcpy(nde, de, le16_to_cpu(de->length));
 	ddno = de->down ? de_down_pointer(de) : 0;
 	hpfs_delete_de(i->i_sb, dnode, de);
 	set_last_pointer(i->i_sb, dnode, ddno);
@@ -517,11 +517,11 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno)
 	try_it_again:
 	if (hpfs_stop_cycles(i->i_sb, dno, &c1, &c2, "delete_empty_dnode")) return;
 	if (!(dnode = hpfs_map_dnode(i->i_sb, dno, &qbh))) return;
-	if (dnode->first_free > 56) goto end;
-	if (dnode->first_free == 52 || dnode->first_free == 56) {
+	if (le32_to_cpu(dnode->first_free) > 56) goto end;
+	if (le32_to_cpu(dnode->first_free) == 52 || le32_to_cpu(dnode->first_free) == 56) {
 		struct hpfs_dirent *de_end;
 		int root = dnode->root_dnode;
-		up = dnode->up;
+		up = le32_to_cpu(dnode->up);
 		de = dnode_first_de(dnode);
 		down = de->down ? de_down_pointer(de) : 0;
 		if (hpfs_sb(i->i_sb)->sb_chk) if (root && !down) {
@@ -545,13 +545,13 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno)
 				return;
 			    }
 			if ((d1 = hpfs_map_dnode(i->i_sb, down, &qbh1))) {
-				d1->up = up;
+				d1->up = cpu_to_le32(up);
 				d1->root_dnode = 1;
 				hpfs_mark_4buffers_dirty(&qbh1);
 				hpfs_brelse4(&qbh1);
 			}
 			if ((fnode = hpfs_map_fnode(i->i_sb, up, &bh))) {
-				fnode->u.external[0].disk_secno = down;
+				fnode->u.external[0].disk_secno = cpu_to_le32(down);
 				mark_buffer_dirty(bh);
 				brelse(bh);
 			}
@@ -570,22 +570,22 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno)
 		for_all_poss(i, hpfs_pos_subst, ((loff_t)dno << 4) | 1, ((loff_t)up << 4) | p);
 		if (!down) {
 			de->down = 0;
-			de->length -= 4;
-			dnode->first_free -= 4;
+			de->length = cpu_to_le16(le16_to_cpu(de->length) - 4);
+			dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) - 4);
 			memmove(de_next_de(de), (char *)de_next_de(de) + 4,
-				(char *)dnode + dnode->first_free - (char *)de_next_de(de));
+				(char *)dnode + le32_to_cpu(dnode->first_free) - (char *)de_next_de(de));
 		} else {
 			struct dnode *d1;
 			struct quad_buffer_head qbh1;
-			*(dnode_secno *) ((void *) de + de->length - 4) = down;
+			*(dnode_secno *) ((void *) de + le16_to_cpu(de->length) - 4) = down;
 			if ((d1 = hpfs_map_dnode(i->i_sb, down, &qbh1))) {
-				d1->up = up;
+				d1->up = cpu_to_le32(up);
 				hpfs_mark_4buffers_dirty(&qbh1);
 				hpfs_brelse4(&qbh1);
 			}
 		}
 	} else {
-		hpfs_error(i->i_sb, "delete_empty_dnode: dnode %08x, first_free == %03x", dno, dnode->first_free);
+		hpfs_error(i->i_sb, "delete_empty_dnode: dnode %08x, first_free == %03x", dno, le32_to_cpu(dnode->first_free));
 		goto end;
 	}
 
@@ -596,18 +596,18 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno)
 		struct quad_buffer_head qbh1;
 		if (!de_next->down) goto endm;
 		ndown = de_down_pointer(de_next);
-		if (!(de_cp = kmalloc(de->length, GFP_NOFS))) {
+		if (!(de_cp = kmalloc(le16_to_cpu(de->length), GFP_NOFS))) {
 			printk("HPFS: out of memory for dtree balancing\n");
 			goto endm;
 		}
-		memcpy(de_cp, de, de->length);
+		memcpy(de_cp, de, le16_to_cpu(de->length));
 		hpfs_delete_de(i->i_sb, dnode, de);
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
 		for_all_poss(i, hpfs_pos_subst, ((loff_t)up << 4) | p, 4);
 		for_all_poss(i, hpfs_pos_del, ((loff_t)up << 4) | p, 1);
 		if (de_cp->down) if ((d1 = hpfs_map_dnode(i->i_sb, de_down_pointer(de_cp), &qbh1))) {
-			d1->up = ndown;
+			d1->up = cpu_to_le32(ndown);
 			hpfs_mark_4buffers_dirty(&qbh1);
 			hpfs_brelse4(&qbh1);
 		}
@@ -635,7 +635,7 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno)
 			struct hpfs_dirent *del = dnode_last_de(d1);
 			dlp = del->down ? de_down_pointer(del) : 0;
 			if (!dlp && down) {
-				if (d1->first_free > 2044) {
+				if (le32_to_cpu(d1->first_free) > 2044) {
 					if (hpfs_sb(i->i_sb)->sb_chk >= 2) {
 						printk("HPFS: warning: unbalanced dnode tree, see hpfs.txt 4 more info\n");
 						printk("HPFS: warning: terminating balancing operation\n");
@@ -647,38 +647,38 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno)
 					printk("HPFS: warning: unbalanced dnode tree, see hpfs.txt 4 more info\n");
 					printk("HPFS: warning: goin'on\n");
 				}
-				del->length += 4;
+				del->length = cpu_to_le16(le16_to_cpu(del->length) + 4);
 				del->down = 1;
-				d1->first_free += 4;
+				d1->first_free = cpu_to_le32(le32_to_cpu(d1->first_free) + 4);
 			}
 			if (dlp && !down) {
-				del->length -= 4;
+				del->length = cpu_to_le16(le16_to_cpu(del->length) - 4);
 				del->down = 0;
-				d1->first_free -= 4;
+				d1->first_free = cpu_to_le32(le32_to_cpu(d1->first_free) - 4);
 			} else if (down)
-				*(dnode_secno *) ((void *) del + del->length - 4) = down;
+				*(dnode_secno *) ((void *) del + le16_to_cpu(del->length) - 4) = cpu_to_le32(down);
 		} else goto endm;
-		if (!(de_cp = kmalloc(de_prev->length, GFP_NOFS))) {
+		if (!(de_cp = kmalloc(le16_to_cpu(de_prev->length), GFP_NOFS))) {
 			printk("HPFS: out of memory for dtree balancing\n");
 			hpfs_brelse4(&qbh1);
 			goto endm;
 		}
 		hpfs_mark_4buffers_dirty(&qbh1);
 		hpfs_brelse4(&qbh1);
-		memcpy(de_cp, de_prev, de_prev->length);
+		memcpy(de_cp, de_prev, le16_to_cpu(de_prev->length));
 		hpfs_delete_de(i->i_sb, dnode, de_prev);
 		if (!de_prev->down) {
-			de_prev->length += 4;
+			de_prev->length = cpu_to_le16(le16_to_cpu(de_prev->length) + 4);
 			de_prev->down = 1;
-			dnode->first_free += 4;
+			dnode->first_free = cpu_to_le32(le32_to_cpu(dnode->first_free) + 4);
 		}
-		*(dnode_secno *) ((void *) de_prev + de_prev->length - 4) = ndown;
+		*(dnode_secno *) ((void *) de_prev + le16_to_cpu(de_prev->length) - 4) = cpu_to_le32(ndown);
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
 		for_all_poss(i, hpfs_pos_subst, ((loff_t)up << 4) | (p - 1), 4);
 		for_all_poss(i, hpfs_pos_subst, ((loff_t)up << 4) | p, ((loff_t)up << 4) | (p - 1));
 		if (down) if ((d1 = hpfs_map_dnode(i->i_sb, de_down_pointer(de), &qbh1))) {
-			d1->up = ndown;
+			d1->up = cpu_to_le32(ndown);
 			hpfs_mark_4buffers_dirty(&qbh1);
 			hpfs_brelse4(&qbh1);
 		}
@@ -701,7 +701,6 @@ int hpfs_remove_dirent(struct inode *i, dnode_secno dno, struct hpfs_dirent *de,
 {
 	struct dnode *dnode = qbh->data;
 	dnode_secno down = 0;
-	int lock = 0;
 	loff_t t;
 	if (de->first || de->last) {
 		hpfs_error(i->i_sb, "hpfs_remove_dirent: attempt to delete first or last dirent in dnode %08x", dno);
@@ -710,11 +709,8 @@ int hpfs_remove_dirent(struct inode *i, dnode_secno dno, struct hpfs_dirent *de,
 	}
 	if (de->down) down = de_down_pointer(de);
 	if (depth && (de->down || (de == dnode_first_de(dnode) && de_next_de(de)->last))) {
-		lock = 1;
-		hpfs_lock_creation(i->i_sb);
 		if (hpfs_check_free_dnodes(i->i_sb, FREE_DNODES_DEL)) {
 			hpfs_brelse4(qbh);
-			hpfs_unlock_creation(i->i_sb);
 			return 2;
 		}
 	}
@@ -727,11 +723,9 @@ int hpfs_remove_dirent(struct inode *i, dnode_secno dno, struct hpfs_dirent *de,
 		dnode_secno a = move_to_top(i, down, dno);
 		for_all_poss(i, hpfs_pos_subst, 5, t);
 		if (a) delete_empty_dnode(i, a);
-		if (lock) hpfs_unlock_creation(i->i_sb);
 		return !a;
 	}
 	delete_empty_dnode(i, dno);
-	if (lock) hpfs_unlock_creation(i->i_sb);
 	return 0;
 }
 
@@ -751,8 +745,8 @@ void hpfs_count_dnodes(struct super_block *s, dnode_secno dno, int *n_dnodes,
 	ptr = 0;
 	go_up:
 	if (!(dnode = hpfs_map_dnode(s, dno, &qbh))) return;
-	if (hpfs_sb(s)->sb_chk) if (odno && odno != -1 && dnode->up != odno)
-		hpfs_error(s, "hpfs_count_dnodes: bad up pointer; dnode %08x, down %08x points to %08x", odno, dno, dnode->up);
+	if (hpfs_sb(s)->sb_chk) if (odno && odno != -1 && le32_to_cpu(dnode->up) != odno)
+		hpfs_error(s, "hpfs_count_dnodes: bad up pointer; dnode %08x, down %08x points to %08x", odno, dno, le32_to_cpu(dnode->up));
 	de = dnode_first_de(dnode);
 	if (ptr) while(1) {
 		if (de->down) if (de_down_pointer(de) == ptr) goto process_de;
@@ -776,7 +770,7 @@ void hpfs_count_dnodes(struct super_block *s, dnode_secno dno, int *n_dnodes,
 	if (!de->first && !de->last && n_items) (*n_items)++;
 	if ((de = de_next_de(de)) < dnode_end_de(dnode)) goto next_de;
 	ptr = dno;
-	dno = dnode->up;
+	dno = le32_to_cpu(dnode->up);
 	if (dnode->root_dnode) {
 		hpfs_brelse4(&qbh);
 		return;
@@ -824,8 +818,8 @@ dnode_secno hpfs_de_as_down_as_possible(struct super_block *s, dnode_secno dno)
 			return d;
 	if (!(de = map_nth_dirent(s, d, 1, &qbh, NULL))) return dno;
 	if (hpfs_sb(s)->sb_chk)
-		if (up && ((struct dnode *)qbh.data)->up != up)
-			hpfs_error(s, "hpfs_de_as_down_as_possible: bad up pointer; dnode %08x, down %08x points to %08x", up, d, ((struct dnode *)qbh.data)->up);
+		if (up && le32_to_cpu(((struct dnode *)qbh.data)->up) != up)
+			hpfs_error(s, "hpfs_de_as_down_as_possible: bad up pointer; dnode %08x, down %08x points to %08x", up, d, le32_to_cpu(((struct dnode *)qbh.data)->up));
 	if (!de->down) {
 		hpfs_brelse4(&qbh);
 		return d;
@@ -874,7 +868,7 @@ struct hpfs_dirent *map_pos_dirent(struct inode *inode, loff_t *posp,
 	/* Going up */
 	if (dnode->root_dnode) goto bail;
 
-	if (!(up_dnode = hpfs_map_dnode(inode->i_sb, dnode->up, &qbh0)))
+	if (!(up_dnode = hpfs_map_dnode(inode->i_sb, le32_to_cpu(dnode->up), &qbh0)))
 		goto bail;
 
 	end_up_de = dnode_end_de(up_dnode);
@@ -882,16 +876,16 @@ struct hpfs_dirent *map_pos_dirent(struct inode *inode, loff_t *posp,
 	for (up_de = dnode_first_de(up_dnode); up_de < end_up_de;
 	     up_de = de_next_de(up_de)) {
 		if (!(++c & 077)) hpfs_error(inode->i_sb,
-			"map_pos_dirent: pos crossed dnode boundary; dnode = %08x", dnode->up);
+			"map_pos_dirent: pos crossed dnode boundary; dnode = %08x", le32_to_cpu(dnode->up));
 		if (up_de->down && de_down_pointer(up_de) == dno) {
-			*posp = ((loff_t) dnode->up << 4) + c;
+			*posp = ((loff_t) le32_to_cpu(dnode->up) << 4) + c;
 			hpfs_brelse4(&qbh0);
 			return de;
 		}
 	}
 	
 	hpfs_error(inode->i_sb, "map_pos_dirent: pointer to dnode %08x not found in parent dnode %08x",
-		dno, dnode->up);
+		dno, le32_to_cpu(dnode->up));
 	hpfs_brelse4(&qbh0);
 	
 	bail:
@@ -1017,17 +1011,17 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno,
 		/*name2[15] = 0xff;*/
 		name1len = 15; name2len = 256;
 	}
-	if (!(upf = hpfs_map_fnode(s, f->up, &bh))) {
+	if (!(upf = hpfs_map_fnode(s, le32_to_cpu(f->up), &bh))) {
 		kfree(name2);
 		return NULL;
 	}	
 	if (!upf->dirflag) {
 		brelse(bh);
-		hpfs_error(s, "fnode %08x has non-directory parent %08x", fno, f->up);
+		hpfs_error(s, "fnode %08x has non-directory parent %08x", fno, le32_to_cpu(f->up));
 		kfree(name2);
 		return NULL;
 	}
-	dno = upf->u.external[0].disk_secno;
+	dno = le32_to_cpu(upf->u.external[0].disk_secno);
 	brelse(bh);
 	go_down:
 	downd = 0;
@@ -1049,7 +1043,7 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno,
 		return NULL;
 	}
 	next_de:
-	if (de->fnode == fno) {
+	if (le32_to_cpu(de->fnode) == fno) {
 		kfree(name2);
 		return de;
 	}
@@ -1065,7 +1059,7 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno,
 		goto go_down;
 	}
 	f:
-	if (de->fnode == fno) {
+	if (le32_to_cpu(de->fnode) == fno) {
 		kfree(name2);
 		return de;
 	}
@@ -1074,7 +1068,7 @@ struct hpfs_dirent *map_fnode_dirent(struct super_block *s, fnode_secno fno,
 	if ((de = de_next_de(de)) < de_end) goto next_de;
 	if (d->root_dnode) goto not_found;
 	downd = dno;
-	dno = d->up;
+	dno = le32_to_cpu(d->up);
 	hpfs_brelse4(qbh);
 	if (hpfs_sb(s)->sb_chk)
 		if (hpfs_stop_cycles(s, downd, &d1, &d2, "map_fnode_dirent #2")) {
diff --git a/fs/hpfs/ea.c b/fs/hpfs/ea.c
index 45e53d972b4..d8b84d113c8 100644
--- a/fs/hpfs/ea.c
+++ b/fs/hpfs/ea.c
@@ -24,7 +24,7 @@ void hpfs_ea_ext_remove(struct super_block *s, secno a, int ano, unsigned len)
 		}
 		if (hpfs_ea_read(s, a, ano, pos, 4, ex)) return;
 		if (ea->indirect) {
-			if (ea->valuelen != 8) {
+			if (ea_valuelen(ea) != 8) {
 				hpfs_error(s, "ea->indirect set while ea->valuelen!=8, %s %08x, pos %08x",
 					ano ? "anode" : "sectors", a, pos);
 				return;
@@ -33,7 +33,7 @@ void hpfs_ea_ext_remove(struct super_block *s, secno a, int ano, unsigned len)
 				return;
 			hpfs_ea_remove(s, ea_sec(ea), ea->anode, ea_len(ea));
 		}
-		pos += ea->namelen + ea->valuelen + 5;
+		pos += ea->namelen + ea_valuelen(ea) + 5;
 	}
 	if (!ano) hpfs_free_sectors(s, a, (len+511) >> 9);
 	else {
@@ -76,24 +76,24 @@ int hpfs_read_ea(struct super_block *s, struct fnode *fnode, char *key,
 	unsigned pos;
 	int ano, len;
 	secno a;
+	char ex[4 + 255 + 1 + 8];
 	struct extended_attribute *ea;
 	struct extended_attribute *ea_end = fnode_end_ea(fnode);
 	for (ea = fnode_ea(fnode); ea < ea_end; ea = next_ea(ea))
 		if (!strcmp(ea->name, key)) {
 			if (ea->indirect)
 				goto indirect;
-			if (ea->valuelen >= size)
+			if (ea_valuelen(ea) >= size)
 				return -EINVAL;
-			memcpy(buf, ea_data(ea), ea->valuelen);
-			buf[ea->valuelen] = 0;
+			memcpy(buf, ea_data(ea), ea_valuelen(ea));
+			buf[ea_valuelen(ea)] = 0;
 			return 0;
 		}
-	a = fnode->ea_secno;
-	len = fnode->ea_size_l;
+	a = le32_to_cpu(fnode->ea_secno);
+	len = le32_to_cpu(fnode->ea_size_l);
 	ano = fnode->ea_anode;
 	pos = 0;
 	while (pos < len) {
-		char ex[4 + 255 + 1 + 8];
 		ea = (struct extended_attribute *)ex;
 		if (pos + 4 > len) {
 			hpfs_error(s, "EAs don't end correctly, %s %08x, len %08x",
@@ -106,14 +106,14 @@ int hpfs_read_ea(struct super_block *s, struct fnode *fnode, char *key,
 		if (!strcmp(ea->name, key)) {
 			if (ea->indirect)
 				goto indirect;
-			if (ea->valuelen >= size)
+			if (ea_valuelen(ea) >= size)
 				return -EINVAL;
-			if (hpfs_ea_read(s, a, ano, pos + 4 + ea->namelen + 1, ea->valuelen, buf))
+			if (hpfs_ea_read(s, a, ano, pos + 4 + ea->namelen + 1, ea_valuelen(ea), buf))
 				return -EIO;
-			buf[ea->valuelen] = 0;
+			buf[ea_valuelen(ea)] = 0;
 			return 0;
 		}
-		pos += ea->namelen + ea->valuelen + 5;
+		pos += ea->namelen + ea_valuelen(ea) + 5;
 	}
 	return -ENOENT;
 indirect:
@@ -138,16 +138,16 @@ char *hpfs_get_ea(struct super_block *s, struct fnode *fnode, char *key, int *si
 		if (!strcmp(ea->name, key)) {
 			if (ea->indirect)
 				return get_indirect_ea(s, ea->anode, ea_sec(ea), *size = ea_len(ea));
-			if (!(ret = kmalloc((*size = ea->valuelen) + 1, GFP_NOFS))) {
+			if (!(ret = kmalloc((*size = ea_valuelen(ea)) + 1, GFP_NOFS))) {
 				printk("HPFS: out of memory for EA\n");
 				return NULL;
 			}
-			memcpy(ret, ea_data(ea), ea->valuelen);
-			ret[ea->valuelen] = 0;
+			memcpy(ret, ea_data(ea), ea_valuelen(ea));
+			ret[ea_valuelen(ea)] = 0;
 			return ret;
 		}
-	a = fnode->ea_secno;
-	len = fnode->ea_size_l;
+	a = le32_to_cpu(fnode->ea_secno);
+	len = le32_to_cpu(fnode->ea_size_l);
 	ano = fnode->ea_anode;
 	pos = 0;
 	while (pos < len) {
@@ -164,18 +164,18 @@ char *hpfs_get_ea(struct super_block *s, struct fnode *fnode, char *key, int *si
 		if (!strcmp(ea->name, key)) {
 			if (ea->indirect)
 				return get_indirect_ea(s, ea->anode, ea_sec(ea), *size = ea_len(ea));
-			if (!(ret = kmalloc((*size = ea->valuelen) + 1, GFP_NOFS))) {
+			if (!(ret = kmalloc((*size = ea_valuelen(ea)) + 1, GFP_NOFS))) {
 				printk("HPFS: out of memory for EA\n");
 				return NULL;
 			}
-			if (hpfs_ea_read(s, a, ano, pos + 4 + ea->namelen + 1, ea->valuelen, ret)) {
+			if (hpfs_ea_read(s, a, ano, pos + 4 + ea->namelen + 1, ea_valuelen(ea), ret)) {
 				kfree(ret);
 				return NULL;
 			}
-			ret[ea->valuelen] = 0;
+			ret[ea_valuelen(ea)] = 0;
 			return ret;
 		}
-		pos += ea->namelen + ea->valuelen + 5;
+		pos += ea->namelen + ea_valuelen(ea) + 5;
 	}
 	return NULL;
 }
@@ -202,13 +202,13 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key,
 			if (ea->indirect) {
 				if (ea_len(ea) == size)
 					set_indirect_ea(s, ea->anode, ea_sec(ea), data, size);
-			} else if (ea->valuelen == size) {
+			} else if (ea_valuelen(ea) == size) {
 				memcpy(ea_data(ea), data, size);
 			}
 			return;
 		}
-	a = fnode->ea_secno;
-	len = fnode->ea_size_l;
+	a = le32_to_cpu(fnode->ea_secno);
+	len = le32_to_cpu(fnode->ea_size_l);
 	ano = fnode->ea_anode;
 	pos = 0;
 	while (pos < len) {
@@ -228,68 +228,70 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key,
 					set_indirect_ea(s, ea->anode, ea_sec(ea), data, size);
 			}
 			else {
-				if (ea->valuelen == size)
+				if (ea_valuelen(ea) == size)
 					hpfs_ea_write(s, a, ano, pos + 4 + ea->namelen + 1, size, data);
 			}
 			return;
 		}
-		pos += ea->namelen + ea->valuelen + 5;
+		pos += ea->namelen + ea_valuelen(ea) + 5;
 	}
-	if (!fnode->ea_offs) {
-		/*if (fnode->ea_size_s) {
+	if (!le16_to_cpu(fnode->ea_offs)) {
+		/*if (le16_to_cpu(fnode->ea_size_s)) {
 			hpfs_error(s, "fnode %08x: ea_size_s == %03x, ea_offs == 0",
-				inode->i_ino, fnode->ea_size_s);
+				inode->i_ino, le16_to_cpu(fnode->ea_size_s));
 			return;
 		}*/
-		fnode->ea_offs = 0xc4;
+		fnode->ea_offs = cpu_to_le16(0xc4);
 	}
-	if (fnode->ea_offs < 0xc4 || fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s > 0x200) {
+	if (le16_to_cpu(fnode->ea_offs) < 0xc4 || le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s) + le16_to_cpu(fnode->ea_size_s) > 0x200) {
 		hpfs_error(s, "fnode %08lx: ea_offs == %03x, ea_size_s == %03x",
 			(unsigned long)inode->i_ino,
-			fnode->ea_offs, fnode->ea_size_s);
+			le32_to_cpu(fnode->ea_offs), le16_to_cpu(fnode->ea_size_s));
 		return;
 	}
-	if ((fnode->ea_size_s || !fnode->ea_size_l) &&
-	     fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s + strlen(key) + size + 5 <= 0x200) {
+	if ((le16_to_cpu(fnode->ea_size_s) || !le32_to_cpu(fnode->ea_size_l)) &&
+	     le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s) + le16_to_cpu(fnode->ea_size_s) + strlen(key) + size + 5 <= 0x200) {
 		ea = fnode_end_ea(fnode);
 		*(char *)ea = 0;
 		ea->namelen = strlen(key);
-		ea->valuelen = size;
+		ea->valuelen_lo = size;
+		ea->valuelen_hi = size >> 8;
 		strcpy(ea->name, key);
 		memcpy(ea_data(ea), data, size);
-		fnode->ea_size_s += strlen(key) + size + 5;
+		fnode->ea_size_s = cpu_to_le16(le16_to_cpu(fnode->ea_size_s) + strlen(key) + size + 5);
 		goto ret;
 	}
 	/* Most the code here is 99.9993422% unused. I hope there are no bugs.
 	   But what .. HPFS.IFS has also bugs in ea management. */
-	if (fnode->ea_size_s && !fnode->ea_size_l) {
+	if (le16_to_cpu(fnode->ea_size_s) && !le32_to_cpu(fnode->ea_size_l)) {
 		secno n;
 		struct buffer_head *bh;
 		char *data;
-		if (!(n = hpfs_alloc_sector(s, fno, 1, 0, 1))) return;
+		if (!(n = hpfs_alloc_sector(s, fno, 1, 0))) return;
 		if (!(data = hpfs_get_sector(s, n, &bh))) {
 			hpfs_free_sectors(s, n, 1);
 			return;
 		}
-		memcpy(data, fnode_ea(fnode), fnode->ea_size_s);
-		fnode->ea_size_l = fnode->ea_size_s;
-		fnode->ea_size_s = 0;
-		fnode->ea_secno = n;
-		fnode->ea_anode = 0;
+		memcpy(data, fnode_ea(fnode), le16_to_cpu(fnode->ea_size_s));
+		fnode->ea_size_l = cpu_to_le32(le16_to_cpu(fnode->ea_size_s));
+		fnode->ea_size_s = cpu_to_le16(0);
+		fnode->ea_secno = cpu_to_le32(n);
+		fnode->ea_anode = cpu_to_le32(0);
 		mark_buffer_dirty(bh);
 		brelse(bh);
 	}
-	pos = fnode->ea_size_l + 5 + strlen(key) + size;
-	len = (fnode->ea_size_l + 511) >> 9;
+	pos = le32_to_cpu(fnode->ea_size_l) + 5 + strlen(key) + size;
+	len = (le32_to_cpu(fnode->ea_size_l) + 511) >> 9;
 	if (pos >= 30000) goto bail;
 	while (((pos + 511) >> 9) > len) {
 		if (!len) {
-			if (!(fnode->ea_secno = hpfs_alloc_sector(s, fno, 1, 0, 1)))
-				goto bail;
+			secno q = hpfs_alloc_sector(s, fno, 1, 0);
+			if (!q) goto bail;
+			fnode->ea_secno = cpu_to_le32(q);
 			fnode->ea_anode = 0;
 			len++;
 		} else if (!fnode->ea_anode) {
-			if (hpfs_alloc_if_possible(s, fnode->ea_secno + len)) {
+			if (hpfs_alloc_if_possible(s, le32_to_cpu(fnode->ea_secno) + len)) {
 				len++;
 			} else {
 				/* Aargh... don't know how to create ea anodes :-( */
@@ -298,26 +300,26 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key,
 				anode_secno a_s;
 				if (!(anode = hpfs_alloc_anode(s, fno, &a_s, &bh)))
 					goto bail;
-				anode->up = fno;
+				anode->up = cpu_to_le32(fno);
 				anode->btree.fnode_parent = 1;
 				anode->btree.n_free_nodes--;
 				anode->btree.n_used_nodes++;
-				anode->btree.first_free += 12;
-				anode->u.external[0].disk_secno = fnode->ea_secno;
-				anode->u.external[0].file_secno = 0;
-				anode->u.external[0].length = len;
+				anode->btree.first_free = cpu_to_le16(le16_to_cpu(anode->btree.first_free) + 12);
+				anode->u.external[0].disk_secno = cpu_to_le32(le32_to_cpu(fnode->ea_secno));
+				anode->u.external[0].file_secno = cpu_to_le32(0);
+				anode->u.external[0].length = cpu_to_le32(len);
 				mark_buffer_dirty(bh);
 				brelse(bh);
 				fnode->ea_anode = 1;
-				fnode->ea_secno = a_s;*/
+				fnode->ea_secno = cpu_to_le32(a_s);*/
 				secno new_sec;
 				int i;
-				if (!(new_sec = hpfs_alloc_sector(s, fno, 1, 1 - ((pos + 511) >> 9), 1)))
+				if (!(new_sec = hpfs_alloc_sector(s, fno, 1, 1 - ((pos + 511) >> 9))))
 					goto bail;
 				for (i = 0; i < len; i++) {
 					struct buffer_head *bh1, *bh2;
 					void *b1, *b2;
-					if (!(b1 = hpfs_map_sector(s, fnode->ea_secno + i, &bh1, len - i - 1))) {
+					if (!(b1 = hpfs_map_sector(s, le32_to_cpu(fnode->ea_secno) + i, &bh1, len - i - 1))) {
 						hpfs_free_sectors(s, new_sec, (pos + 511) >> 9);
 						goto bail;
 					}
@@ -331,13 +333,13 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key,
 					mark_buffer_dirty(bh2);
 					brelse(bh2);
 				}
-				hpfs_free_sectors(s, fnode->ea_secno, len);
-				fnode->ea_secno = new_sec;
+				hpfs_free_sectors(s, le32_to_cpu(fnode->ea_secno), len);
+				fnode->ea_secno = cpu_to_le32(new_sec);
 				len = (pos + 511) >> 9;
 			}
 		}
 		if (fnode->ea_anode) {
-			if (hpfs_add_sector_to_btree(s, fnode->ea_secno,
+			if (hpfs_add_sector_to_btree(s, le32_to_cpu(fnode->ea_secno),
 						     0, len) != -1) {
 				len++;
 			} else {
@@ -349,17 +351,17 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, const char *key,
 	h[1] = strlen(key);
 	h[2] = size & 0xff;
 	h[3] = size >> 8;
-	if (hpfs_ea_write(s, fnode->ea_secno, fnode->ea_anode, fnode->ea_size_l, 4, h)) goto bail;
-	if (hpfs_ea_write(s, fnode->ea_secno, fnode->ea_anode, fnode->ea_size_l + 4, h[1] + 1, key)) goto bail;
-	if (hpfs_ea_write(s, fnode->ea_secno, fnode->ea_anode, fnode->ea_size_l + 5 + h[1], size, data)) goto bail;
-	fnode->ea_size_l = pos;
+	if (hpfs_ea_write(s, le32_to_cpu(fnode->ea_secno), fnode->ea_anode, le32_to_cpu(fnode->ea_size_l), 4, h)) goto bail;
+	if (hpfs_ea_write(s, le32_to_cpu(fnode->ea_secno), fnode->ea_anode, le32_to_cpu(fnode->ea_size_l) + 4, h[1] + 1, key)) goto bail;
+	if (hpfs_ea_write(s, le32_to_cpu(fnode->ea_secno), fnode->ea_anode, le32_to_cpu(fnode->ea_size_l) + 5 + h[1], size, data)) goto bail;
+	fnode->ea_size_l = cpu_to_le32(pos);
 	ret:
 	hpfs_i(inode)->i_ea_size += 5 + strlen(key) + size;
 	return;
 	bail:
-	if (fnode->ea_secno)
-		if (fnode->ea_anode) hpfs_truncate_btree(s, fnode->ea_secno, 1, (fnode->ea_size_l + 511) >> 9);
-		else hpfs_free_sectors(s, fnode->ea_secno + ((fnode->ea_size_l + 511) >> 9), len - ((fnode->ea_size_l + 511) >> 9));
-	else fnode->ea_secno = fnode->ea_size_l = 0;
+	if (le32_to_cpu(fnode->ea_secno))
+		if (fnode->ea_anode) hpfs_truncate_btree(s, le32_to_cpu(fnode->ea_secno), 1, (le32_to_cpu(fnode->ea_size_l) + 511) >> 9);
+		else hpfs_free_sectors(s, le32_to_cpu(fnode->ea_secno) + ((le32_to_cpu(fnode->ea_size_l) + 511) >> 9), len - ((le32_to_cpu(fnode->ea_size_l) + 511) >> 9));
+	else fnode->ea_secno = fnode->ea_size_l = cpu_to_le32(0);
 }
 	
diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c
index 9b9eb6933e4..89c500ee521 100644
--- a/fs/hpfs/file.c
+++ b/fs/hpfs/file.c
@@ -20,8 +20,8 @@ static int hpfs_file_release(struct inode *inode, struct file *file)
 
 int hpfs_file_fsync(struct file *file, int datasync)
 {
-	/*return file_fsync(file, datasync);*/
-	return 0; /* Don't fsync :-) */
+	struct inode *inode = file->f_mapping->host;
+	return sync_blockdev(inode->i_sb->s_bdev);
 }
 
 /*
@@ -48,38 +48,46 @@ static secno hpfs_bmap(struct inode *inode, unsigned file_secno)
 static void hpfs_truncate(struct inode *i)
 {
 	if (IS_IMMUTABLE(i)) return /*-EPERM*/;
-	hpfs_lock(i->i_sb);
+	hpfs_lock_assert(i->i_sb);
+
 	hpfs_i(i)->i_n_secs = 0;
 	i->i_blocks = 1 + ((i->i_size + 511) >> 9);
 	hpfs_i(i)->mmu_private = i->i_size;
 	hpfs_truncate_btree(i->i_sb, i->i_ino, 1, ((i->i_size + 511) >> 9));
 	hpfs_write_inode(i);
 	hpfs_i(i)->i_n_secs = 0;
-	hpfs_unlock(i->i_sb);
 }
 
 static int hpfs_get_block(struct inode *inode, sector_t iblock, struct buffer_head *bh_result, int create)
 {
+	int r;
 	secno s;
+	hpfs_lock(inode->i_sb);
 	s = hpfs_bmap(inode, iblock);
 	if (s) {
 		map_bh(bh_result, inode->i_sb, s);
-		return 0;
+		goto ret_0;
 	}
-	if (!create) return 0;
+	if (!create) goto ret_0;
 	if (iblock<<9 != hpfs_i(inode)->mmu_private) {
 		BUG();
-		return -EIO;
+		r = -EIO;
+		goto ret_r;
 	}
 	if ((s = hpfs_add_sector_to_btree(inode->i_sb, inode->i_ino, 1, inode->i_blocks - 1)) == -1) {
 		hpfs_truncate_btree(inode->i_sb, inode->i_ino, 1, inode->i_blocks - 1);
-		return -ENOSPC;
+		r = -ENOSPC;
+		goto ret_r;
 	}
 	inode->i_blocks++;
 	hpfs_i(inode)->mmu_private += 512;
 	set_buffer_new(bh_result);
 	map_bh(bh_result, inode->i_sb, s);
-	return 0;
+	ret_0:
+	r = 0;
+	ret_r:
+	hpfs_unlock(inode->i_sb);
+	return r;
 }
 
 static int hpfs_writepage(struct page *page, struct writeback_control *wbc)
@@ -130,8 +138,11 @@ static ssize_t hpfs_file_write(struct file *file, const char __user *buf,
 	ssize_t retval;
 
 	retval = do_sync_write(file, buf, count, ppos);
-	if (retval > 0)
+	if (retval > 0) {
+		hpfs_lock(file->f_path.dentry->d_sb);
 		hpfs_i(file->f_path.dentry->d_inode)->i_dirty = 1;
+		hpfs_unlock(file->f_path.dentry->d_sb);
+	}
 	return retval;
 }
 
diff --git a/fs/hpfs/hpfs.h b/fs/hpfs/hpfs.h
index 0e84c73cd9c..8b0650aae32 100644
--- a/fs/hpfs/hpfs.h
+++ b/fs/hpfs/hpfs.h
@@ -19,9 +19,13 @@
    For definitive information on HPFS, ask somebody else -- this is guesswork.
    There are certain to be many mistakes. */
 
+#if !defined(__LITTLE_ENDIAN) && !defined(__BIG_ENDIAN)
+#error unknown endian
+#endif
+
 /* Notation */
 
-typedef unsigned secno;			/* sector number, partition relative */
+typedef u32 secno;			/* sector number, partition relative */
 
 typedef secno dnode_secno;		/* sector number of a dnode */
 typedef secno fnode_secno;		/* sector number of an fnode */
@@ -38,28 +42,28 @@ typedef u32 time32_t;		/* 32-bit time_t type */
 
 struct hpfs_boot_block
 {
-  unsigned char jmp[3];
-  unsigned char oem_id[8];
-  unsigned char bytes_per_sector[2];	/* 512 */
-  unsigned char sectors_per_cluster;
-  unsigned char n_reserved_sectors[2];
-  unsigned char n_fats;
-  unsigned char n_rootdir_entries[2];
-  unsigned char n_sectors_s[2];
-  unsigned char media_byte;
-  unsigned short sectors_per_fat;
-  unsigned short sectors_per_track;
-  unsigned short heads_per_cyl;
-  unsigned int n_hidden_sectors;
-  unsigned int n_sectors_l;		/* size of partition */
-  unsigned char drive_number;
-  unsigned char mbz;
-  unsigned char sig_28h;		/* 28h */
-  unsigned char vol_serno[4];
-  unsigned char vol_label[11];
-  unsigned char sig_hpfs[8];		/* "HPFS    " */
-  unsigned char pad[448];
-  unsigned short magic;			/* aa55 */
+  u8 jmp[3];
+  u8 oem_id[8];
+  u8 bytes_per_sector[2];	/* 512 */
+  u8 sectors_per_cluster;
+  u8 n_reserved_sectors[2];
+  u8 n_fats;
+  u8 n_rootdir_entries[2];
+  u8 n_sectors_s[2];
+  u8 media_byte;
+  u16 sectors_per_fat;
+  u16 sectors_per_track;
+  u16 heads_per_cyl;
+  u32 n_hidden_sectors;
+  u32 n_sectors_l;		/* size of partition */
+  u8 drive_number;
+  u8 mbz;
+  u8 sig_28h;			/* 28h */
+  u8 vol_serno[4];
+  u8 vol_label[11];
+  u8 sig_hpfs[8];		/* "HPFS    " */
+  u8 pad[448];
+  u16 magic;			/* aa55 */
 };
 
 
@@ -71,31 +75,29 @@ struct hpfs_boot_block
 
 struct hpfs_super_block
 {
-  unsigned magic;			/* f995 e849 */
-  unsigned magic1;			/* fa53 e9c5, more magic? */
-  /*unsigned huh202;*/			/* ?? 202 = N. of B. in 1.00390625 S.*/
-  char version;				/* version of a filesystem  usually 2 */
-  char funcversion;			/* functional version - oldest version
+  u32 magic;				/* f995 e849 */
+  u32 magic1;				/* fa53 e9c5, more magic? */
+  u8 version;				/* version of a filesystem  usually 2 */
+  u8 funcversion;			/* functional version - oldest version
   					   of filesystem that can understand
 					   this disk */
-  unsigned short int zero;		/* 0 */
+  u16 zero;				/* 0 */
   fnode_secno root;			/* fnode of root directory */
   secno n_sectors;			/* size of filesystem */
-  unsigned n_badblocks;			/* number of bad blocks */
+  u32 n_badblocks;			/* number of bad blocks */
   secno bitmaps;			/* pointers to free space bit maps */
-  unsigned zero1;			/* 0 */
+  u32 zero1;				/* 0 */
   secno badblocks;			/* bad block list */
-  unsigned zero3;			/* 0 */
+  u32 zero3;				/* 0 */
   time32_t last_chkdsk;			/* date last checked, 0 if never */
-  /*unsigned zero4;*/			/* 0 */
-  time32_t last_optimize;			/* date last optimized, 0 if never */
+  time32_t last_optimize;		/* date last optimized, 0 if never */
   secno n_dir_band;			/* number of sectors in dir band */
   secno dir_band_start;			/* first sector in dir band */
   secno dir_band_end;			/* last sector in dir band */
   secno dir_band_bitmap;		/* free space map, 1 dnode per bit */
-  char volume_name[32];			/* not used */
+  u8 volume_name[32];			/* not used */
   secno user_id_table;			/* 8 preallocated sectors - user id */
-  unsigned zero6[103];			/* 0 */
+  u32 zero6[103];			/* 0 */
 };
 
 
@@ -107,44 +109,65 @@ struct hpfs_super_block
 
 struct hpfs_spare_block
 {
-  unsigned magic;			/* f991 1849 */
-  unsigned magic1;			/* fa52 29c5, more magic? */
-
-  unsigned dirty: 1;			/* 0 clean, 1 "improperly stopped" */
-  /*unsigned flag1234: 4;*/		/* unknown flags */
-  unsigned sparedir_used: 1;		/* spare dirblks used */
-  unsigned hotfixes_used: 1;		/* hotfixes used */
-  unsigned bad_sector: 1;		/* bad sector, corrupted disk (???) */
-  unsigned bad_bitmap: 1;		/* bad bitmap */
-  unsigned fast: 1;			/* partition was fast formatted */
-  unsigned old_wrote: 1;		/* old version wrote to partion */
-  unsigned old_wrote_1: 1;		/* old version wrote to partion (?) */
-  unsigned install_dasd_limits: 1;	/* HPFS386 flags */
-  unsigned resynch_dasd_limits: 1;
-  unsigned dasd_limits_operational: 1;
-  unsigned multimedia_active: 1;
-  unsigned dce_acls_active: 1;
-  unsigned dasd_limits_dirty: 1;
-  unsigned flag67: 2;
-  unsigned char mm_contlgulty;
-  unsigned char unused;
+  u32 magic;				/* f991 1849 */
+  u32 magic1;				/* fa52 29c5, more magic? */
+
+#ifdef __LITTLE_ENDIAN
+  u8 dirty: 1;				/* 0 clean, 1 "improperly stopped" */
+  u8 sparedir_used: 1;			/* spare dirblks used */
+  u8 hotfixes_used: 1;			/* hotfixes used */
+  u8 bad_sector: 1;			/* bad sector, corrupted disk (???) */
+  u8 bad_bitmap: 1;			/* bad bitmap */
+  u8 fast: 1;				/* partition was fast formatted */
+  u8 old_wrote: 1;			/* old version wrote to partion */
+  u8 old_wrote_1: 1;			/* old version wrote to partion (?) */
+#else
+  u8 old_wrote_1: 1;			/* old version wrote to partion (?) */
+  u8 old_wrote: 1;			/* old version wrote to partion */
+  u8 fast: 1;				/* partition was fast formatted */
+  u8 bad_bitmap: 1;			/* bad bitmap */
+  u8 bad_sector: 1;			/* bad sector, corrupted disk (???) */
+  u8 hotfixes_used: 1;			/* hotfixes used */
+  u8 sparedir_used: 1;			/* spare dirblks used */
+  u8 dirty: 1;				/* 0 clean, 1 "improperly stopped" */
+#endif
+
+#ifdef __LITTLE_ENDIAN
+  u8 install_dasd_limits: 1;		/* HPFS386 flags */
+  u8 resynch_dasd_limits: 1;
+  u8 dasd_limits_operational: 1;
+  u8 multimedia_active: 1;
+  u8 dce_acls_active: 1;
+  u8 dasd_limits_dirty: 1;
+  u8 flag67: 2;
+#else
+  u8 flag67: 2;
+  u8 dasd_limits_dirty: 1;
+  u8 dce_acls_active: 1;
+  u8 multimedia_active: 1;
+  u8 dasd_limits_operational: 1;
+  u8 resynch_dasd_limits: 1;
+  u8 install_dasd_limits: 1;		/* HPFS386 flags */
+#endif
+
+  u8 mm_contlgulty;
+  u8 unused;
 
   secno hotfix_map;			/* info about remapped bad sectors */
-  unsigned n_spares_used;		/* number of hotfixes */
-  unsigned n_spares;			/* number of spares in hotfix map */
-  unsigned n_dnode_spares_free;		/* spare dnodes unused */
-  unsigned n_dnode_spares;		/* length of spare_dnodes[] list,
+  u32 n_spares_used;			/* number of hotfixes */
+  u32 n_spares;				/* number of spares in hotfix map */
+  u32 n_dnode_spares_free;		/* spare dnodes unused */
+  u32 n_dnode_spares;			/* length of spare_dnodes[] list,
 					   follows in this block*/
   secno code_page_dir;			/* code page directory block */
-  unsigned n_code_pages;		/* number of code pages */
-  /*unsigned large_numbers[2];*/	/* ?? */
-  unsigned super_crc;			/* on HPFS386 and LAN Server this is
+  u32 n_code_pages;			/* number of code pages */
+  u32 super_crc;			/* on HPFS386 and LAN Server this is
   					   checksum of superblock, on normal
 					   OS/2 unused */
-  unsigned spare_crc;			/* on HPFS386 checksum of spareblock */
-  unsigned zero1[15];			/* unused */
+  u32 spare_crc;			/* on HPFS386 checksum of spareblock */
+  u32 zero1[15];			/* unused */
   dnode_secno spare_dnodes[100];	/* emergency free dnode list */
-  unsigned zero2[1];			/* room for more? */
+  u32 zero2[1];				/* room for more? */
 };
 
 /* The bad block list is 4 sectors long.  The first word must be zero,
@@ -179,18 +202,18 @@ struct hpfs_spare_block
 
 struct code_page_directory
 {
-  unsigned magic;			/* 4945 21f7 */
-  unsigned n_code_pages;		/* number of pointers following */
-  unsigned zero1[2];
+  u32 magic;				/* 4945 21f7 */
+  u32 n_code_pages;			/* number of pointers following */
+  u32 zero1[2];
   struct {
-    unsigned short ix;			/* index */
-    unsigned short code_page_number;	/* code page number */
-    unsigned bounds;			/* matches corresponding word
+    u16 ix;				/* index */
+    u16 code_page_number;		/* code page number */
+    u32 bounds;				/* matches corresponding word
 					   in data block */
     secno code_page_data;		/* sector number of a code_page_data
 					   containing c.p. array */
-    unsigned short index;		/* index in c.p. array in that sector*/
-    unsigned short unknown;		/* some unknown value; usually 0;
+    u16 index;				/* index in c.p. array in that sector*/
+    u16 unknown;			/* some unknown value; usually 0;
     					   2 in Japanese version */
   } array[31];				/* unknown length */
 };
@@ -201,21 +224,21 @@ struct code_page_directory
 
 struct code_page_data
 {
-  unsigned magic;			/* 8945 21f7 */
-  unsigned n_used;			/* # elements used in c_p_data[] */
-  unsigned bounds[3];			/* looks a bit like
+  u32 magic;				/* 8945 21f7 */
+  u32 n_used;				/* # elements used in c_p_data[] */
+  u32 bounds[3];			/* looks a bit like
 					     (beg1,end1), (beg2,end2)
 					   one byte each */
-  unsigned short offs[3];		/* offsets from start of sector
+  u16 offs[3];				/* offsets from start of sector
 					   to start of c_p_data[ix] */
   struct {
-    unsigned short ix;			/* index */
-    unsigned short code_page_number;	/* code page number */
-    unsigned short unknown;		/* the same as in cp directory */
-    unsigned char map[128];		/* upcase table for chars 80..ff */
-    unsigned short zero2;
+    u16 ix;				/* index */
+    u16 code_page_number;		/* code page number */
+    u16 unknown;			/* the same as in cp directory */
+    u8 map[128];			/* upcase table for chars 80..ff */
+    u16 zero2;
   } code_page[3];
-  unsigned char incognita[78];
+  u8 incognita[78];
 };
 
 
@@ -255,50 +278,84 @@ struct code_page_data
 #define DNODE_MAGIC   0x77e40aae
 
 struct dnode {
-  unsigned magic;			/* 77e4 0aae */
-  unsigned first_free;			/* offset from start of dnode to
+  u32 magic;				/* 77e4 0aae */
+  u32 first_free;			/* offset from start of dnode to
 					   first free dir entry */
-  unsigned root_dnode:1;		/* Is it root dnode? */
-  unsigned increment_me:31;		/* some kind of activity counter?
-					   Neither HPFS.IFS nor CHKDSK cares
+#ifdef __LITTLE_ENDIAN
+  u8 root_dnode: 1;			/* Is it root dnode? */
+  u8 increment_me: 7;			/* some kind of activity counter? */
+					/* Neither HPFS.IFS nor CHKDSK cares
+					   if you change this word */
+#else
+  u8 increment_me: 7;			/* some kind of activity counter? */
+					/* Neither HPFS.IFS nor CHKDSK cares
 					   if you change this word */
+  u8 root_dnode: 1;			/* Is it root dnode? */
+#endif
+  u8 increment_me2[3];
   secno up;				/* (root dnode) directory's fnode
 					   (nonroot) parent dnode */
   dnode_secno self;			/* pointer to this dnode */
-  unsigned char dirent[2028];		/* one or more dirents */
+  u8 dirent[2028];			/* one or more dirents */
 };
 
 struct hpfs_dirent {
-  unsigned short length;		/* offset to next dirent */
-  unsigned first: 1;			/* set on phony ^A^A (".") entry */
-  unsigned has_acl: 1;
-  unsigned down: 1;			/* down pointer present (after name) */
-  unsigned last: 1;			/* set on phony \377 entry */
-  unsigned has_ea: 1;			/* entry has EA */
-  unsigned has_xtd_perm: 1;		/* has extended perm list (???) */
-  unsigned has_explicit_acl: 1;
-  unsigned has_needea: 1;		/* ?? some EA has NEEDEA set
+  u16 length;				/* offset to next dirent */
+
+#ifdef __LITTLE_ENDIAN
+  u8 first: 1;				/* set on phony ^A^A (".") entry */
+  u8 has_acl: 1;
+  u8 down: 1;				/* down pointer present (after name) */
+  u8 last: 1;				/* set on phony \377 entry */
+  u8 has_ea: 1;				/* entry has EA */
+  u8 has_xtd_perm: 1;			/* has extended perm list (???) */
+  u8 has_explicit_acl: 1;
+  u8 has_needea: 1;			/* ?? some EA has NEEDEA set
+					   I have no idea why this is
+					   interesting in a dir entry */
+#else
+  u8 has_needea: 1;			/* ?? some EA has NEEDEA set
 					   I have no idea why this is
 					   interesting in a dir entry */
-  unsigned read_only: 1;		/* dos attrib */
-  unsigned hidden: 1;			/* dos attrib */
-  unsigned system: 1;			/* dos attrib */
-  unsigned flag11: 1;			/* would be volume label dos attrib */
-  unsigned directory: 1;		/* dos attrib */
-  unsigned archive: 1;			/* dos attrib */
-  unsigned not_8x3: 1;			/* name is not 8.3 */
-  unsigned flag15: 1;
+  u8 has_explicit_acl: 1;
+  u8 has_xtd_perm: 1;			/* has extended perm list (???) */
+  u8 has_ea: 1;				/* entry has EA */
+  u8 last: 1;				/* set on phony \377 entry */
+  u8 down: 1;				/* down pointer present (after name) */
+  u8 has_acl: 1;
+  u8 first: 1;				/* set on phony ^A^A (".") entry */
+#endif
+
+#ifdef __LITTLE_ENDIAN
+  u8 read_only: 1;			/* dos attrib */
+  u8 hidden: 1;				/* dos attrib */
+  u8 system: 1;				/* dos attrib */
+  u8 flag11: 1;				/* would be volume label dos attrib */
+  u8 directory: 1;			/* dos attrib */
+  u8 archive: 1;			/* dos attrib */
+  u8 not_8x3: 1;			/* name is not 8.3 */
+  u8 flag15: 1;
+#else
+  u8 flag15: 1;
+  u8 not_8x3: 1;			/* name is not 8.3 */
+  u8 archive: 1;			/* dos attrib */
+  u8 directory: 1;			/* dos attrib */
+  u8 flag11: 1;				/* would be volume label dos attrib */
+  u8 system: 1;				/* dos attrib */
+  u8 hidden: 1;				/* dos attrib */
+  u8 read_only: 1;			/* dos attrib */
+#endif
+
   fnode_secno fnode;			/* fnode giving allocation info */
   time32_t write_date;			/* mtime */
-  unsigned file_size;			/* file length, bytes */
+  u32 file_size;			/* file length, bytes */
   time32_t read_date;			/* atime */
   time32_t creation_date;			/* ctime */
-  unsigned ea_size;			/* total EA length, bytes */
-  unsigned char no_of_acls : 3;		/* number of ACL's */
-  unsigned char reserver : 5;
-  unsigned char ix;			/* code page index (of filename), see
+  u32 ea_size;				/* total EA length, bytes */
+  u8 no_of_acls;			/* number of ACL's (low 3 bits) */
+  u8 ix;				/* code page index (of filename), see
 					   struct code_page_data */
-  unsigned char namelen, name[1];	/* file name */
+  u8 namelen, name[1];			/* file name */
   /* dnode_secno down;	  btree down pointer, if present,
      			  follows name on next word boundary, or maybe it
 			  precedes next dirent, which is on a word boundary. */
@@ -318,38 +375,50 @@ struct hpfs_dirent {
 
 struct bplus_leaf_node
 {
-  unsigned file_secno;			/* first file sector in extent */
-  unsigned length;			/* length, sectors */
+  u32 file_secno;			/* first file sector in extent */
+  u32 length;				/* length, sectors */
   secno disk_secno;			/* first corresponding disk sector */
 };
 
 struct bplus_internal_node
 {
-  unsigned file_secno;			/* subtree maps sectors < this  */
+  u32 file_secno;			/* subtree maps sectors < this  */
   anode_secno down;			/* pointer to subtree */
 };
 
 struct bplus_header
 {
-  unsigned hbff: 1;	/* high bit of first free entry offset */
-  unsigned flag1: 1;
-  unsigned flag2: 1;
-  unsigned flag3: 1;
-  unsigned flag4: 1;
-  unsigned fnode_parent: 1;		/* ? we're pointed to by an fnode,
+#ifdef __LITTLE_ENDIAN
+  u8 hbff: 1;			/* high bit of first free entry offset */
+  u8 flag1234: 4;
+  u8 fnode_parent: 1;			/* ? we're pointed to by an fnode,
 					   the data btree or some ea or the
 					   main ea bootage pointer ea_secno */
 					/* also can get set in fnodes, which
 					   may be a chkdsk glitch or may mean
 					   this bit is irrelevant in fnodes,
 					   or this interpretation is all wet */
-  unsigned binary_search: 1;		/* suggest binary search (unused) */
-  unsigned internal: 1;			/* 1 -> (internal) tree of anodes
+  u8 binary_search: 1;			/* suggest binary search (unused) */
+  u8 internal: 1;			/* 1 -> (internal) tree of anodes
+					   0 -> (leaf) list of extents */
+#else
+  u8 internal: 1;			/* 1 -> (internal) tree of anodes
 					   0 -> (leaf) list of extents */
-  unsigned char fill[3];
-  unsigned char n_free_nodes;		/* free nodes in following array */
-  unsigned char n_used_nodes;		/* used nodes in following array */
-  unsigned short first_free;		/* offset from start of header to
+  u8 binary_search: 1;			/* suggest binary search (unused) */
+  u8 fnode_parent: 1;			/* ? we're pointed to by an fnode,
+					   the data btree or some ea or the
+					   main ea bootage pointer ea_secno */
+					/* also can get set in fnodes, which
+					   may be a chkdsk glitch or may mean
+					   this bit is irrelevant in fnodes,
+					   or this interpretation is all wet */
+  u8 flag1234: 4;
+  u8 hbff: 1;			/* high bit of first free entry offset */
+#endif
+  u8 fill[3];
+  u8 n_free_nodes;			/* free nodes in following array */
+  u8 n_used_nodes;			/* used nodes in following array */
+  u16 first_free;			/* offset from start of header to
 					   first free node in array */
   union {
     struct bplus_internal_node internal[0]; /* (internal) 2-word entries giving
@@ -369,37 +438,38 @@ struct bplus_header
 
 struct fnode
 {
-  unsigned magic;			/* f7e4 0aae */
-  unsigned zero1[2];			/* read history */
-  unsigned char len, name[15];		/* true length, truncated name */
+  u32 magic;				/* f7e4 0aae */
+  u32 zero1[2];				/* read history */
+  u8 len, name[15];			/* true length, truncated name */
   fnode_secno up;			/* pointer to file's directory fnode */
-  /*unsigned zero2[3];*/
   secno acl_size_l;
   secno acl_secno;
-  unsigned short acl_size_s;
-  char acl_anode;
-  char zero2;				/* history bit count */
-  unsigned ea_size_l;			/* length of disk-resident ea's */
+  u16 acl_size_s;
+  u8 acl_anode;
+  u8 zero2;				/* history bit count */
+  u32 ea_size_l;			/* length of disk-resident ea's */
   secno ea_secno;			/* first sector of disk-resident ea's*/
-  unsigned short ea_size_s;		/* length of fnode-resident ea's */
-
-  unsigned flag0: 1;
-  unsigned ea_anode: 1;			/* 1 -> ea_secno is an anode */
-  unsigned flag2: 1;
-  unsigned flag3: 1;
-  unsigned flag4: 1;
-  unsigned flag5: 1;
-  unsigned flag6: 1;
-  unsigned flag7: 1;
-  unsigned dirflag: 1;			/* 1 -> directory.  first & only extent
+  u16 ea_size_s;			/* length of fnode-resident ea's */
+
+#ifdef __LITTLE_ENDIAN
+  u8 flag0: 1;
+  u8 ea_anode: 1;			/* 1 -> ea_secno is an anode */
+  u8 flag234567: 6;
+#else
+  u8 flag234567: 6;
+  u8 ea_anode: 1;			/* 1 -> ea_secno is an anode */
+  u8 flag0: 1;
+#endif
+
+#ifdef __LITTLE_ENDIAN
+  u8 dirflag: 1;			/* 1 -> directory.  first & only extent
 					   points to dnode. */
-  unsigned flag9: 1;
-  unsigned flag10: 1;
-  unsigned flag11: 1;
-  unsigned flag12: 1;
-  unsigned flag13: 1;
-  unsigned flag14: 1;
-  unsigned flag15: 1;
+  u8 flag9012345: 7;
+#else
+  u8 flag9012345: 7;
+  u8 dirflag: 1;			/* 1 -> directory.  first & only extent
+					   points to dnode. */
+#endif
 
   struct bplus_header btree;		/* b+ tree, 8 extents or 12 subtrees */
   union {
@@ -407,17 +477,16 @@ struct fnode
     struct bplus_internal_node internal[12];
   } u;
 
-  unsigned file_size;			/* file length, bytes */
-  unsigned n_needea;			/* number of EA's with NEEDEA set */
-  char user_id[16];			/* unused */
-  unsigned short ea_offs;		/* offset from start of fnode
+  u32 file_size;			/* file length, bytes */
+  u32 n_needea;				/* number of EA's with NEEDEA set */
+  u8 user_id[16];			/* unused */
+  u16 ea_offs;				/* offset from start of fnode
 					   to first fnode-resident ea */
-  char dasd_limit_treshhold;
-  char dasd_limit_delta;
-  unsigned dasd_limit;
-  unsigned dasd_usage;
-  /*unsigned zero5[2];*/
-  unsigned char ea[316];		/* zero or more EA's, packed together
+  u8 dasd_limit_treshhold;
+  u8 dasd_limit_delta;
+  u32 dasd_limit;
+  u32 dasd_usage;
+  u8 ea[316];				/* zero or more EA's, packed together
 					   with no alignment padding.
 					   (Do not use this name, get here
 					   via fnode + ea_offs. I think.) */
@@ -430,7 +499,7 @@ struct fnode
 
 struct anode
 {
-  unsigned magic;			/* 37e4 0aae */
+  u32 magic;				/* 37e4 0aae */
   anode_secno self;			/* pointer to this anode */
   secno up;				/* parent anode or fnode */
 
@@ -440,7 +509,7 @@ struct anode
     struct bplus_internal_node internal[60];
   } u;
 
-  unsigned fill[3];			/* unused */
+  u32 fill[3];				/* unused */
 };
 
 
@@ -461,25 +530,31 @@ struct anode
 
 struct extended_attribute
 {
-  unsigned indirect: 1;			/* 1 -> value gives sector number
+#ifdef __LITTLE_ENDIAN
+  u8 indirect: 1;			/* 1 -> value gives sector number
 					   where real value starts */
-  unsigned anode: 1;			/* 1 -> sector is an anode
+  u8 anode: 1;				/* 1 -> sector is an anode
+					   that points to fragmented value */
+  u8 flag23456: 5;
+  u8 needea: 1;				/* required ea */
+#else
+  u8 needea: 1;				/* required ea */
+  u8 flag23456: 5;
+  u8 anode: 1;				/* 1 -> sector is an anode
 					   that points to fragmented value */
-  unsigned flag2: 1;
-  unsigned flag3: 1;
-  unsigned flag4: 1;
-  unsigned flag5: 1;
-  unsigned flag6: 1;
-  unsigned needea: 1;			/* required ea */
-  unsigned char namelen;		/* length of name, bytes */
-  unsigned short valuelen;		/* length of value, bytes */
-  unsigned char name[0];
+  u8 indirect: 1;			/* 1 -> value gives sector number
+					   where real value starts */
+#endif
+  u8 namelen;				/* length of name, bytes */
+  u8 valuelen_lo;			/* length of value, bytes */
+  u8 valuelen_hi;			/* length of value, bytes */
+  u8 name[0];
   /*
-    unsigned char name[namelen];	ascii attrib name
-    unsigned char nul;			terminating '\0', not counted
-    unsigned char value[valuelen];	value, arbitrary
+    u8 name[namelen];			ascii attrib name
+    u8 nul;				terminating '\0', not counted
+    u8 value[valuelen];			value, arbitrary
       if this.indirect, valuelen is 8 and the value is
-        unsigned length;		real length of value, bytes
+        u32 length;			real length of value, bytes
         secno secno;			sector address where it starts
       if this.anode, the above sector number is the root of an anode tree
         which points to the value.
diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h
index c15adbca07f..dd552f862c8 100644
--- a/fs/hpfs/hpfs_fn.h
+++ b/fs/hpfs/hpfs_fn.h
@@ -13,6 +13,7 @@
 #include <linux/pagemap.h>
 #include <linux/buffer_head.h>
 #include <linux/slab.h>
+#include <asm/unaligned.h>
 
 #include "hpfs.h"
 
@@ -51,18 +52,16 @@ struct hpfs_inode_info {
 	unsigned i_disk_sec;	/* (files) minimalist cache of alloc info */
 	unsigned i_n_secs;	/* (files) minimalist cache of alloc info */
 	unsigned i_ea_size;	/* size of extended attributes */
-	unsigned i_conv : 2;	/* (files) crlf->newline hackery */
 	unsigned i_ea_mode : 1;	/* file's permission is stored in ea */
 	unsigned i_ea_uid : 1;	/* file's uid is stored in ea */
 	unsigned i_ea_gid : 1;	/* file's gid is stored in ea */
 	unsigned i_dirty : 1;
-	struct mutex i_mutex;
-	struct mutex i_parent_mutex;
 	loff_t **i_rddir_off;
 	struct inode vfs_inode;
 };
 
 struct hpfs_sb_info {
+	struct mutex hpfs_mutex;	/* global hpfs lock */
 	ino_t sb_root;			/* inode number of root dir */
 	unsigned sb_fs_size;		/* file system size, sectors */
 	unsigned sb_bitmaps;		/* sector number of bitmap list */
@@ -74,7 +73,6 @@ struct hpfs_sb_info {
 	uid_t sb_uid;			/* uid from mount options */
 	gid_t sb_gid;			/* gid from mount options */
 	umode_t sb_mode;		/* mode from mount options */
-	unsigned sb_conv : 2;		/* crlf->newline hackery */
 	unsigned sb_eas : 2;		/* eas: 0-ignore, 1-ro, 2-rw */
 	unsigned sb_err : 2;		/* on errs: 0-cont, 1-ro, 2-panic */
 	unsigned sb_chk : 2;		/* checks: 0-no, 1-normal, 2-strict */
@@ -87,20 +85,9 @@ struct hpfs_sb_info {
 	unsigned *sb_bmp_dir;		/* main bitmap directory */
 	unsigned sb_c_bitmap;		/* current bitmap */
 	unsigned sb_max_fwd_alloc;	/* max forwad allocation */
-	struct mutex hpfs_creation_de;	/* when creating dirents, nobody else
-					   can alloc blocks */
-	/*unsigned sb_mounting : 1;*/
 	int sb_timeshift;
 };
 
-/*
- * conv= options
- */
-
-#define CONV_BINARY 0			/* no conversion */
-#define CONV_TEXT 1			/* crlf->newline */
-#define CONV_AUTO 2			/* decide based on file contents */
-
 /* Four 512-byte buffers and the 2k block obtained by concatenating them */
 
 struct quad_buffer_head {
@@ -113,7 +100,7 @@ struct quad_buffer_head {
 static inline dnode_secno de_down_pointer (struct hpfs_dirent *de)
 {
   CHKCOND(de->down,("HPFS: de_down_pointer: !de->down\n"));
-  return *(dnode_secno *) ((void *) de + de->length - 4);
+  return le32_to_cpu(*(dnode_secno *) ((void *) de + le16_to_cpu(de->length) - 4));
 }
 
 /* The first dir entry in a dnode */
@@ -127,41 +114,46 @@ static inline struct hpfs_dirent *dnode_first_de (struct dnode *dnode)
 
 static inline struct hpfs_dirent *dnode_end_de (struct dnode *dnode)
 {
-  CHKCOND(dnode->first_free>=0x14 && dnode->first_free<=0xa00,("HPFS: dnode_end_de: dnode->first_free = %d\n",(int)dnode->first_free));
-  return (void *) dnode + dnode->first_free;
+  CHKCOND(le32_to_cpu(dnode->first_free)>=0x14 && le32_to_cpu(dnode->first_free)<=0xa00,("HPFS: dnode_end_de: dnode->first_free = %x\n",(unsigned)le32_to_cpu(dnode->first_free)));
+  return (void *) dnode + le32_to_cpu(dnode->first_free);
 }
 
 /* The dir entry after dir entry de */
 
 static inline struct hpfs_dirent *de_next_de (struct hpfs_dirent *de)
 {
-  CHKCOND(de->length>=0x20 && de->length<0x800,("HPFS: de_next_de: de->length = %d\n",(int)de->length));
-  return (void *) de + de->length;
+  CHKCOND(le16_to_cpu(de->length)>=0x20 && le16_to_cpu(de->length)<0x800,("HPFS: de_next_de: de->length = %x\n",(unsigned)le16_to_cpu(de->length)));
+  return (void *) de + le16_to_cpu(de->length);
 }
 
 static inline struct extended_attribute *fnode_ea(struct fnode *fnode)
 {
-	return (struct extended_attribute *)((char *)fnode + fnode->ea_offs + fnode->acl_size_s);
+	return (struct extended_attribute *)((char *)fnode + le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s));
 }
 
 static inline struct extended_attribute *fnode_end_ea(struct fnode *fnode)
 {
-	return (struct extended_attribute *)((char *)fnode + fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s);
+	return (struct extended_attribute *)((char *)fnode + le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s) + le16_to_cpu(fnode->ea_size_s));
+}
+
+static unsigned ea_valuelen(struct extended_attribute *ea)
+{
+	return ea->valuelen_lo + 256 * ea->valuelen_hi;
 }
 
 static inline struct extended_attribute *next_ea(struct extended_attribute *ea)
 {
-	return (struct extended_attribute *)((char *)ea + 5 + ea->namelen + ea->valuelen);
+	return (struct extended_attribute *)((char *)ea + 5 + ea->namelen + ea_valuelen(ea));
 }
 
 static inline secno ea_sec(struct extended_attribute *ea)
 {
-	return *(secno *)((char *)ea + 9 + ea->namelen);
+	return le32_to_cpu(get_unaligned((secno *)((char *)ea + 9 + ea->namelen)));
 }
 
 static inline secno ea_len(struct extended_attribute *ea)
 {
-	return *(secno *)((char *)ea + 5 + ea->namelen);
+	return le32_to_cpu(get_unaligned((secno *)((char *)ea + 5 + ea->namelen)));
 }
 
 static inline char *ea_data(struct extended_attribute *ea)
@@ -186,13 +178,13 @@ static inline void copy_de(struct hpfs_dirent *dst, struct hpfs_dirent *src)
 	dst->not_8x3 = n;
 }
 
-static inline unsigned tstbits(unsigned *bmp, unsigned b, unsigned n)
+static inline unsigned tstbits(u32 *bmp, unsigned b, unsigned n)
 {
 	int i;
 	if ((b >= 0x4000) || (b + n - 1 >= 0x4000)) return n;
-	if (!((bmp[(b & 0x3fff) >> 5] >> (b & 0x1f)) & 1)) return 1;
+	if (!((le32_to_cpu(bmp[(b & 0x3fff) >> 5]) >> (b & 0x1f)) & 1)) return 1;
 	for (i = 1; i < n; i++)
-		if (/*b+i < 0x4000 &&*/ !((bmp[((b+i) & 0x3fff) >> 5] >> ((b+i) & 0x1f)) & 1))
+		if (!((le32_to_cpu(bmp[((b+i) & 0x3fff) >> 5]) >> ((b+i) & 0x1f)) & 1))
 			return i + 1;
 	return 0;
 }
@@ -200,12 +192,12 @@ static inline unsigned tstbits(unsigned *bmp, unsigned b, unsigned n)
 /* alloc.c */
 
 int hpfs_chk_sectors(struct super_block *, secno, int, char *);
-secno hpfs_alloc_sector(struct super_block *, secno, unsigned, int, int);
+secno hpfs_alloc_sector(struct super_block *, secno, unsigned, int);
 int hpfs_alloc_if_possible(struct super_block *, secno);
 void hpfs_free_sectors(struct super_block *, secno, unsigned);
 int hpfs_check_free_dnodes(struct super_block *, int);
 void hpfs_free_dnode(struct super_block *, secno);
-struct dnode *hpfs_alloc_dnode(struct super_block *, secno, dnode_secno *, struct quad_buffer_head *, int);
+struct dnode *hpfs_alloc_dnode(struct super_block *, secno, dnode_secno *, struct quad_buffer_head *);
 struct fnode *hpfs_alloc_fnode(struct super_block *, secno, fnode_secno *, struct buffer_head **);
 struct anode *hpfs_alloc_anode(struct super_block *, secno, anode_secno *, struct buffer_head **);
 
@@ -222,8 +214,6 @@ void hpfs_remove_fnode(struct super_block *, fnode_secno fno);
 
 /* buffer.c */
 
-void hpfs_lock_creation(struct super_block *);
-void hpfs_unlock_creation(struct super_block *);
 void *hpfs_map_sector(struct super_block *, unsigned, struct buffer_head **, int);
 void *hpfs_get_sector(struct super_block *, unsigned, struct buffer_head **);
 void *hpfs_map_4sectors(struct super_block *, unsigned, struct quad_buffer_head *, int);
@@ -247,7 +237,7 @@ void hpfs_del_pos(struct inode *, loff_t *);
 struct hpfs_dirent *hpfs_add_de(struct super_block *, struct dnode *,
 				const unsigned char *, unsigned, secno);
 int hpfs_add_dirent(struct inode *, const unsigned char *, unsigned,
-		    struct hpfs_dirent *, int);
+		    struct hpfs_dirent *);
 int hpfs_remove_dirent(struct inode *, dnode_secno, struct hpfs_dirent *, struct quad_buffer_head *, int);
 void hpfs_count_dnodes(struct super_block *, dnode_secno, int *, int *, int *);
 dnode_secno hpfs_de_as_down_as_possible(struct super_block *, dnode_secno dno);
@@ -303,7 +293,6 @@ int hpfs_compare_names(struct super_block *, const unsigned char *, unsigned,
 		       const unsigned char *, unsigned, int);
 int hpfs_is_name_long(const unsigned char *, unsigned);
 void hpfs_adjust_length(const unsigned char *, unsigned *);
-void hpfs_decide_conv(struct inode *, const unsigned char *, unsigned);
 
 /* namei.c */
 
@@ -346,21 +335,26 @@ static inline time32_t gmt_to_local(struct super_block *s, time_t t)
 /*
  * Locking:
  *
- * hpfs_lock() is a leftover from the big kernel lock.
- * Right now, these functions are empty and only left
- * for documentation purposes. The file system no longer
- * works on SMP systems, so the lock is not needed
- * any more.
+ * hpfs_lock() locks the whole filesystem. It must be taken
+ * on any method called by the VFS.
  *
- * If someone is interested in making it work again, this
- * would be the place to start by adding a per-superblock
- * mutex and fixing all the bugs and performance issues
- * caused by that.
+ * We don't do any per-file locking anymore, it is hard to
+ * review and HPFS is not performance-sensitive anyway.
  */
 static inline void hpfs_lock(struct super_block *s)
 {
+	struct hpfs_sb_info *sbi = hpfs_sb(s);
+	mutex_lock(&sbi->hpfs_mutex);
 }
 
 static inline void hpfs_unlock(struct super_block *s)
 {
+	struct hpfs_sb_info *sbi = hpfs_sb(s);
+	mutex_unlock(&sbi->hpfs_mutex);
+}
+
+static inline void hpfs_lock_assert(struct super_block *s)
+{
+	struct hpfs_sb_info *sbi = hpfs_sb(s);
+	WARN_ON(!mutex_is_locked(&sbi->hpfs_mutex));
 }
diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c
index 87f1f787e76..338cd836845 100644
--- a/fs/hpfs/inode.c
+++ b/fs/hpfs/inode.c
@@ -17,7 +17,6 @@ void hpfs_init_inode(struct inode *i)
 	i->i_uid = hpfs_sb(sb)->sb_uid;
 	i->i_gid = hpfs_sb(sb)->sb_gid;
 	i->i_mode = hpfs_sb(sb)->sb_mode;
-	hpfs_inode->i_conv = hpfs_sb(sb)->sb_conv;
 	i->i_size = -1;
 	i->i_blocks = -1;
 	
@@ -116,8 +115,8 @@ void hpfs_read_inode(struct inode *i)
 		i->i_mode |= S_IFDIR;
 		i->i_op = &hpfs_dir_iops;
 		i->i_fop = &hpfs_dir_ops;
-		hpfs_inode->i_parent_dir = fnode->up;
-		hpfs_inode->i_dno = fnode->u.external[0].disk_secno;
+		hpfs_inode->i_parent_dir = le32_to_cpu(fnode->up);
+		hpfs_inode->i_dno = le32_to_cpu(fnode->u.external[0].disk_secno);
 		if (hpfs_sb(sb)->sb_chk >= 2) {
 			struct buffer_head *bh0;
 			if (hpfs_map_fnode(sb, hpfs_inode->i_parent_dir, &bh0)) brelse(bh0);
@@ -133,7 +132,7 @@ void hpfs_read_inode(struct inode *i)
 		i->i_op = &hpfs_file_iops;
 		i->i_fop = &hpfs_file_ops;
 		i->i_nlink = 1;
-		i->i_size = fnode->file_size;
+		i->i_size = le32_to_cpu(fnode->file_size);
 		i->i_blocks = ((i->i_size + 511) >> 9) + 1;
 		i->i_data.a_ops = &hpfs_aops;
 		hpfs_i(i)->mmu_private = i->i_size;
@@ -144,7 +143,7 @@ void hpfs_read_inode(struct inode *i)
 static void hpfs_write_inode_ea(struct inode *i, struct fnode *fnode)
 {
 	struct hpfs_inode_info *hpfs_inode = hpfs_i(i);
-	/*if (fnode->acl_size_l || fnode->acl_size_s) {
+	/*if (le32_to_cpu(fnode->acl_size_l) || le16_to_cpu(fnode->acl_size_s)) {
 		   Some unknown structures like ACL may be in fnode,
 		   we'd better not overwrite them
 		hpfs_error(i->i_sb, "fnode %08x has some unknown HPFS386 stuctures", i->i_ino);
@@ -187,9 +186,7 @@ void hpfs_write_inode(struct inode *i)
 		kfree(hpfs_inode->i_rddir_off);
 		hpfs_inode->i_rddir_off = NULL;
 	}
-	mutex_lock(&hpfs_inode->i_parent_mutex);
 	if (!i->i_nlink) {
-		mutex_unlock(&hpfs_inode->i_parent_mutex);
 		return;
 	}
 	parent = iget_locked(i->i_sb, hpfs_inode->i_parent_dir);
@@ -200,14 +197,9 @@ void hpfs_write_inode(struct inode *i)
 			hpfs_read_inode(parent);
 			unlock_new_inode(parent);
 		}
-		mutex_lock(&hpfs_inode->i_mutex);
 		hpfs_write_inode_nolock(i);
-		mutex_unlock(&hpfs_inode->i_mutex);
 		iput(parent);
-	} else {
-		mark_inode_dirty(i);
 	}
-	mutex_unlock(&hpfs_inode->i_parent_mutex);
 }
 
 void hpfs_write_inode_nolock(struct inode *i)
@@ -226,30 +218,30 @@ void hpfs_write_inode_nolock(struct inode *i)
 		}
 	} else de = NULL;
 	if (S_ISREG(i->i_mode)) {
-		fnode->file_size = i->i_size;
-		if (de) de->file_size = i->i_size;
+		fnode->file_size = cpu_to_le32(i->i_size);
+		if (de) de->file_size = cpu_to_le32(i->i_size);
 	} else if (S_ISDIR(i->i_mode)) {
-		fnode->file_size = 0;
-		if (de) de->file_size = 0;
+		fnode->file_size = cpu_to_le32(0);
+		if (de) de->file_size = cpu_to_le32(0);
 	}
 	hpfs_write_inode_ea(i, fnode);
 	if (de) {
-		de->write_date = gmt_to_local(i->i_sb, i->i_mtime.tv_sec);
-		de->read_date = gmt_to_local(i->i_sb, i->i_atime.tv_sec);
-		de->creation_date = gmt_to_local(i->i_sb, i->i_ctime.tv_sec);
+		de->write_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_mtime.tv_sec));
+		de->read_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_atime.tv_sec));
+		de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_ctime.tv_sec));
 		de->read_only = !(i->i_mode & 0222);
-		de->ea_size = hpfs_inode->i_ea_size;
+		de->ea_size = cpu_to_le32(hpfs_inode->i_ea_size);
 		hpfs_mark_4buffers_dirty(&qbh);
 		hpfs_brelse4(&qbh);
 	}
 	if (S_ISDIR(i->i_mode)) {
 		if ((de = map_dirent(i, hpfs_inode->i_dno, "\001\001", 2, NULL, &qbh))) {
-			de->write_date = gmt_to_local(i->i_sb, i->i_mtime.tv_sec);
-			de->read_date = gmt_to_local(i->i_sb, i->i_atime.tv_sec);
-			de->creation_date = gmt_to_local(i->i_sb, i->i_ctime.tv_sec);
+			de->write_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_mtime.tv_sec));
+			de->read_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_atime.tv_sec));
+			de->creation_date = cpu_to_le32(gmt_to_local(i->i_sb, i->i_ctime.tv_sec));
 			de->read_only = !(i->i_mode & 0222);
-			de->ea_size = /*hpfs_inode->i_ea_size*/0;
-			de->file_size = 0;
+			de->ea_size = cpu_to_le32(/*hpfs_inode->i_ea_size*/0);
+			de->file_size = cpu_to_le32(0);
 			hpfs_mark_4buffers_dirty(&qbh);
 			hpfs_brelse4(&qbh);
 		} else
@@ -269,6 +261,10 @@ int hpfs_setattr(struct dentry *dentry, struct iattr *attr)
 	hpfs_lock(inode->i_sb);
 	if (inode->i_ino == hpfs_sb(inode->i_sb)->sb_root)
 		goto out_unlock;
+	if ((attr->ia_valid & ATTR_UID) && attr->ia_uid >= 0x10000)
+		goto out_unlock;
+	if ((attr->ia_valid & ATTR_GID) && attr->ia_gid >= 0x10000)
+		goto out_unlock;
 	if ((attr->ia_valid & ATTR_SIZE) && attr->ia_size > inode->i_size)
 		goto out_unlock;
 
@@ -284,7 +280,6 @@ int hpfs_setattr(struct dentry *dentry, struct iattr *attr)
 	}
 
 	setattr_copy(inode, attr);
-	mark_inode_dirty(inode);
 
 	hpfs_write_inode(inode);
 
diff --git a/fs/hpfs/map.c b/fs/hpfs/map.c
index 840d033ecee..a790821366a 100644
--- a/fs/hpfs/map.c
+++ b/fs/hpfs/map.c
@@ -21,7 +21,7 @@ unsigned int *hpfs_map_bitmap(struct super_block *s, unsigned bmp_block,
 		hpfs_error(s, "hpfs_map_bitmap called with bad parameter: %08x at %s", bmp_block, id);
 		return NULL;
 	}
-	sec = hpfs_sb(s)->sb_bmp_dir[bmp_block];
+	sec = le32_to_cpu(hpfs_sb(s)->sb_bmp_dir[bmp_block]);
 	if (!sec || sec > hpfs_sb(s)->sb_fs_size-4) {
 		hpfs_error(s, "invalid bitmap block pointer %08x -> %08x at %s", bmp_block, sec, id);
 		return NULL;
@@ -46,18 +46,18 @@ unsigned char *hpfs_load_code_page(struct super_block *s, secno cps)
 	struct code_page_data *cpd;
 	struct code_page_directory *cp = hpfs_map_sector(s, cps, &bh, 0);
 	if (!cp) return NULL;
-	if (cp->magic != CP_DIR_MAGIC) {
-		printk("HPFS: Code page directory magic doesn't match (magic = %08x)\n", cp->magic);
+	if (le32_to_cpu(cp->magic) != CP_DIR_MAGIC) {
+		printk("HPFS: Code page directory magic doesn't match (magic = %08x)\n", le32_to_cpu(cp->magic));
 		brelse(bh);
 		return NULL;
 	}
-	if (!cp->n_code_pages) {
+	if (!le32_to_cpu(cp->n_code_pages)) {
 		printk("HPFS: n_code_pages == 0\n");
 		brelse(bh);
 		return NULL;
 	}
-	cpds = cp->array[0].code_page_data;
-	cpi = cp->array[0].index;
+	cpds = le32_to_cpu(cp->array[0].code_page_data);
+	cpi = le16_to_cpu(cp->array[0].index);
 	brelse(bh);
 
 	if (cpi >= 3) {
@@ -66,12 +66,12 @@ unsigned char *hpfs_load_code_page(struct super_block *s, secno cps)
 	}
 	
 	if (!(cpd = hpfs_map_sector(s, cpds, &bh, 0))) return NULL;
-	if ((unsigned)cpd->offs[cpi] > 0x178) {
+	if (le16_to_cpu(cpd->offs[cpi]) > 0x178) {
 		printk("HPFS: Code page index out of sector\n");
 		brelse(bh);
 		return NULL;
 	}
-	ptr = (unsigned char *)cpd + cpd->offs[cpi] + 6;
+	ptr = (unsigned char *)cpd + le16_to_cpu(cpd->offs[cpi]) + 6;
 	if (!(cp_table = kmalloc(256, GFP_KERNEL))) {
 		printk("HPFS: out of memory for code page table\n");
 		brelse(bh);
@@ -125,7 +125,7 @@ struct fnode *hpfs_map_fnode(struct super_block *s, ino_t ino, struct buffer_hea
 		if (hpfs_sb(s)->sb_chk) {
 			struct extended_attribute *ea;
 			struct extended_attribute *ea_end;
-			if (fnode->magic != FNODE_MAGIC) {
+			if (le32_to_cpu(fnode->magic) != FNODE_MAGIC) {
 				hpfs_error(s, "bad magic on fnode %08lx",
 					(unsigned long)ino);
 				goto bail;
@@ -138,7 +138,7 @@ struct fnode *hpfs_map_fnode(struct super_block *s, ino_t ino, struct buffer_hea
 					    (unsigned long)ino);
 					goto bail;
 				}
-				if (fnode->btree.first_free !=
+				if (le16_to_cpu(fnode->btree.first_free) !=
 				    8 + fnode->btree.n_used_nodes * (fnode->btree.internal ? 8 : 12)) {
 					hpfs_error(s,
 					    "bad first_free pointer in fnode %08lx",
@@ -146,12 +146,12 @@ struct fnode *hpfs_map_fnode(struct super_block *s, ino_t ino, struct buffer_hea
 					goto bail;
 				}
 			}
-			if (fnode->ea_size_s && ((signed int)fnode->ea_offs < 0xc4 ||
-			   (signed int)fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s > 0x200)) {
+			if (le16_to_cpu(fnode->ea_size_s) && (le16_to_cpu(fnode->ea_offs) < 0xc4 ||
+			   le16_to_cpu(fnode->ea_offs) + le16_to_cpu(fnode->acl_size_s) + le16_to_cpu(fnode->ea_size_s) > 0x200)) {
 				hpfs_error(s,
 					"bad EA info in fnode %08lx: ea_offs == %04x ea_size_s == %04x",
 					(unsigned long)ino,
-					fnode->ea_offs, fnode->ea_size_s);
+					le16_to_cpu(fnode->ea_offs), le16_to_cpu(fnode->ea_size_s));
 				goto bail;
 			}
 			ea = fnode_ea(fnode);
@@ -178,16 +178,20 @@ struct anode *hpfs_map_anode(struct super_block *s, anode_secno ano, struct buff
 	if (hpfs_sb(s)->sb_chk) if (hpfs_chk_sectors(s, ano, 1, "anode")) return NULL;
 	if ((anode = hpfs_map_sector(s, ano, bhp, ANODE_RD_AHEAD)))
 		if (hpfs_sb(s)->sb_chk) {
-			if (anode->magic != ANODE_MAGIC || anode->self != ano) {
+			if (le32_to_cpu(anode->magic) != ANODE_MAGIC) {
 				hpfs_error(s, "bad magic on anode %08x", ano);
 				goto bail;
 			}
+			if (le32_to_cpu(anode->self) != ano) {
+				hpfs_error(s, "self pointer invalid on anode %08x", ano);
+				goto bail;
+			}
 			if ((unsigned)anode->btree.n_used_nodes + (unsigned)anode->btree.n_free_nodes !=
 			    (anode->btree.internal ? 60 : 40)) {
 				hpfs_error(s, "bad number of nodes in anode %08x", ano);
 				goto bail;
 			}
-			if (anode->btree.first_free !=
+			if (le16_to_cpu(anode->btree.first_free) !=
 			    8 + anode->btree.n_used_nodes * (anode->btree.internal ? 8 : 12)) {
 				hpfs_error(s, "bad first_free pointer in anode %08x", ano);
 				goto bail;
@@ -219,26 +223,26 @@ struct dnode *hpfs_map_dnode(struct super_block *s, unsigned secno,
 			unsigned p, pp = 0;
 			unsigned char *d = (unsigned char *)dnode;
 			int b = 0;
-			if (dnode->magic != DNODE_MAGIC) {
+			if (le32_to_cpu(dnode->magic) != DNODE_MAGIC) {
 				hpfs_error(s, "bad magic on dnode %08x", secno);
 				goto bail;
 			}
-			if (dnode->self != secno)
-				hpfs_error(s, "bad self pointer on dnode %08x self = %08x", secno, dnode->self);
+			if (le32_to_cpu(dnode->self) != secno)
+				hpfs_error(s, "bad self pointer on dnode %08x self = %08x", secno, le32_to_cpu(dnode->self));
 			/* Check dirents - bad dirents would cause infinite
 			   loops or shooting to memory */
-			if (dnode->first_free > 2048/* || dnode->first_free < 84*/) {
-				hpfs_error(s, "dnode %08x has first_free == %08x", secno, dnode->first_free);
+			if (le32_to_cpu(dnode->first_free) > 2048) {
+				hpfs_error(s, "dnode %08x has first_free == %08x", secno, le32_to_cpu(dnode->first_free));
 				goto bail;
 			}
-			for (p = 20; p < dnode->first_free; p += d[p] + (d[p+1] << 8)) {
+			for (p = 20; p < le32_to_cpu(dnode->first_free); p += d[p] + (d[p+1] << 8)) {
 				struct hpfs_dirent *de = (struct hpfs_dirent *)((char *)dnode + p);
-				if (de->length > 292 || (de->length < 32) || (de->length & 3) || p + de->length > 2048) {
+				if (le16_to_cpu(de->length) > 292 || (le16_to_cpu(de->length) < 32) || (le16_to_cpu(de->length) & 3) || p + le16_to_cpu(de->length) > 2048) {
 					hpfs_error(s, "bad dirent size in dnode %08x, dirent %03x, last %03x", secno, p, pp);
 					goto bail;
 				}
-				if (((31 + de->namelen + de->down*4 + 3) & ~3) != de->length) {
-					if (((31 + de->namelen + de->down*4 + 3) & ~3) < de->length && s->s_flags & MS_RDONLY) goto ok;
+				if (((31 + de->namelen + de->down*4 + 3) & ~3) != le16_to_cpu(de->length)) {
+					if (((31 + de->namelen + de->down*4 + 3) & ~3) < le16_to_cpu(de->length) && s->s_flags & MS_RDONLY) goto ok;
 					hpfs_error(s, "namelen does not match dirent size in dnode %08x, dirent %03x, last %03x", secno, p, pp);
 					goto bail;
 				}
@@ -251,7 +255,7 @@ struct dnode *hpfs_map_dnode(struct super_block *s, unsigned secno,
 				pp = p;
 				
 			}
-			if (p != dnode->first_free) {
+			if (p != le32_to_cpu(dnode->first_free)) {
 				hpfs_error(s, "size on last dirent does not match first_free; dnode %08x", secno);
 				goto bail;
 			}
@@ -277,7 +281,7 @@ dnode_secno hpfs_fnode_dno(struct super_block *s, ino_t ino)
 	if (!fnode)
 		return 0;
 
-	dno = fnode->u.external[0].disk_secno;
+	dno = le32_to_cpu(fnode->u.external[0].disk_secno);
 	brelse(bh);
 	return dno;
 }
diff --git a/fs/hpfs/name.c b/fs/hpfs/name.c
index f24736d7a43..9acdf338def 100644
--- a/fs/hpfs/name.c
+++ b/fs/hpfs/name.c
@@ -8,39 +8,6 @@
 
 #include "hpfs_fn.h"
 
-static const char *text_postfix[]={
-".ASM", ".BAS", ".BAT", ".C", ".CC", ".CFG", ".CMD", ".CON", ".CPP", ".DEF",
-".DOC", ".DPR", ".ERX", ".H", ".HPP", ".HTM", ".HTML", ".JAVA", ".LOG", ".PAS",
-".RC", ".TEX", ".TXT", ".Y", ""};
-
-static const char *text_prefix[]={
-"AUTOEXEC.", "CHANGES", "COPYING", "CONFIG.", "CREDITS", "FAQ", "FILE_ID.DIZ",
-"MAKEFILE", "READ.ME", "README", "TERMCAP", ""};
-
-void hpfs_decide_conv(struct inode *inode, const unsigned char *name, unsigned len)
-{
-	struct hpfs_inode_info *hpfs_inode = hpfs_i(inode);
-	int i;
-	if (hpfs_inode->i_conv != CONV_AUTO) return;
-	for (i = 0; *text_postfix[i]; i++) {
-		int l = strlen(text_postfix[i]);
-		if (l <= len)
-			if (!hpfs_compare_names(inode->i_sb, text_postfix[i], l, name + len - l, l, 0))
-				goto text;
-	}
-	for (i = 0; *text_prefix[i]; i++) {
-		int l = strlen(text_prefix[i]);
-		if (l <= len)
-			if (!hpfs_compare_names(inode->i_sb, text_prefix[i], l, name, l, 0))
-				goto text;
-	}
-	hpfs_inode->i_conv = CONV_BINARY;
-	return;
-	text:
-	hpfs_inode->i_conv = CONV_TEXT;
-	return;
-}
-
 static inline int not_allowed_char(unsigned char c)
 {
 	return c<' ' || c=='"' || c=='*' || c=='/' || c==':' || c=='<' ||
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index d5f8c8a1902..1f05839c27a 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -29,7 +29,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	fnode = hpfs_alloc_fnode(dir->i_sb, hpfs_i(dir)->i_dno, &fno, &bh);
 	if (!fnode)
 		goto bail;
-	dnode = hpfs_alloc_dnode(dir->i_sb, fno, &dno, &qbh0, 1);
+	dnode = hpfs_alloc_dnode(dir->i_sb, fno, &dno, &qbh0);
 	if (!dnode)
 		goto bail1;
 	memset(&dee, 0, sizeof dee);
@@ -37,8 +37,8 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (!(mode & 0222)) dee.read_only = 1;
 	/*dee.archive = 0;*/
 	dee.hidden = name[0] == '.';
-	dee.fnode = fno;
-	dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds());
+	dee.fnode = cpu_to_le32(fno);
+	dee.creation_date = dee.write_date = dee.read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds()));
 	result = new_inode(dir->i_sb);
 	if (!result)
 		goto bail2;
@@ -46,7 +46,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	result->i_ino = fno;
 	hpfs_i(result)->i_parent_dir = dir->i_ino;
 	hpfs_i(result)->i_dno = dno;
-	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date);
+	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
 	result->i_ctime.tv_nsec = 0; 
 	result->i_mtime.tv_nsec = 0; 
 	result->i_atime.tv_nsec = 0; 
@@ -60,8 +60,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (dee.read_only)
 		result->i_mode &= ~0222;
 
-	mutex_lock(&hpfs_i(dir)->i_mutex);
-	r = hpfs_add_dirent(dir, name, len, &dee, 0);
+	r = hpfs_add_dirent(dir, name, len, &dee);
 	if (r == 1)
 		goto bail3;
 	if (r == -1) {
@@ -70,21 +69,21 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	}
 	fnode->len = len;
 	memcpy(fnode->name, name, len > 15 ? 15 : len);
-	fnode->up = dir->i_ino;
+	fnode->up = cpu_to_le32(dir->i_ino);
 	fnode->dirflag = 1;
 	fnode->btree.n_free_nodes = 7;
 	fnode->btree.n_used_nodes = 1;
-	fnode->btree.first_free = 0x14;
-	fnode->u.external[0].disk_secno = dno;
-	fnode->u.external[0].file_secno = -1;
+	fnode->btree.first_free = cpu_to_le16(0x14);
+	fnode->u.external[0].disk_secno = cpu_to_le32(dno);
+	fnode->u.external[0].file_secno = cpu_to_le32(-1);
 	dnode->root_dnode = 1;
-	dnode->up = fno;
+	dnode->up = cpu_to_le32(fno);
 	de = hpfs_add_de(dir->i_sb, dnode, "\001\001", 2, 0);
-	de->creation_date = de->write_date = de->read_date = gmt_to_local(dir->i_sb, get_seconds());
+	de->creation_date = de->write_date = de->read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds()));
 	if (!(mode & 0222)) de->read_only = 1;
 	de->first = de->directory = 1;
 	/*de->hidden = de->system = 0;*/
-	de->fnode = fno;
+	de->fnode = cpu_to_le32(fno);
 	mark_buffer_dirty(bh);
 	brelse(bh);
 	hpfs_mark_4buffers_dirty(&qbh0);
@@ -101,11 +100,9 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 		hpfs_write_inode_nolock(result);
 	}
 	d_instantiate(dentry, result);
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	hpfs_unlock(dir->i_sb);
 	return 0;
 bail3:
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	iput(result);
 bail2:
 	hpfs_brelse4(&qbh0);
@@ -140,8 +137,8 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 	if (!(mode & 0222)) dee.read_only = 1;
 	dee.archive = 1;
 	dee.hidden = name[0] == '.';
-	dee.fnode = fno;
-	dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds());
+	dee.fnode = cpu_to_le32(fno);
+	dee.creation_date = dee.write_date = dee.read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds()));
 
 	result = new_inode(dir->i_sb);
 	if (!result)
@@ -154,9 +151,8 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 	result->i_op = &hpfs_file_iops;
 	result->i_fop = &hpfs_file_ops;
 	result->i_nlink = 1;
-	hpfs_decide_conv(result, name, len);
 	hpfs_i(result)->i_parent_dir = dir->i_ino;
-	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date);
+	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
 	result->i_ctime.tv_nsec = 0;
 	result->i_mtime.tv_nsec = 0;
 	result->i_atime.tv_nsec = 0;
@@ -168,8 +164,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 	result->i_data.a_ops = &hpfs_aops;
 	hpfs_i(result)->mmu_private = 0;
 
-	mutex_lock(&hpfs_i(dir)->i_mutex);
-	r = hpfs_add_dirent(dir, name, len, &dee, 0);
+	r = hpfs_add_dirent(dir, name, len, &dee);
 	if (r == 1)
 		goto bail2;
 	if (r == -1) {
@@ -178,7 +173,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 	}
 	fnode->len = len;
 	memcpy(fnode->name, name, len > 15 ? 15 : len);
-	fnode->up = dir->i_ino;
+	fnode->up = cpu_to_le32(dir->i_ino);
 	mark_buffer_dirty(bh);
 	brelse(bh);
 
@@ -193,12 +188,10 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
 		hpfs_write_inode_nolock(result);
 	}
 	d_instantiate(dentry, result);
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	hpfs_unlock(dir->i_sb);
 	return 0;
 
 bail2:
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	iput(result);
 bail1:
 	brelse(bh);
@@ -232,8 +225,8 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
 	if (!(mode & 0222)) dee.read_only = 1;
 	dee.archive = 1;
 	dee.hidden = name[0] == '.';
-	dee.fnode = fno;
-	dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds());
+	dee.fnode = cpu_to_le32(fno);
+	dee.creation_date = dee.write_date = dee.read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds()));
 
 	result = new_inode(dir->i_sb);
 	if (!result)
@@ -242,7 +235,7 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
 	hpfs_init_inode(result);
 	result->i_ino = fno;
 	hpfs_i(result)->i_parent_dir = dir->i_ino;
-	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date);
+	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
 	result->i_ctime.tv_nsec = 0;
 	result->i_mtime.tv_nsec = 0;
 	result->i_atime.tv_nsec = 0;
@@ -254,8 +247,7 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
 	result->i_blocks = 1;
 	init_special_inode(result, mode, rdev);
 
-	mutex_lock(&hpfs_i(dir)->i_mutex);
-	r = hpfs_add_dirent(dir, name, len, &dee, 0);
+	r = hpfs_add_dirent(dir, name, len, &dee);
 	if (r == 1)
 		goto bail2;
 	if (r == -1) {
@@ -264,19 +256,17 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
 	}
 	fnode->len = len;
 	memcpy(fnode->name, name, len > 15 ? 15 : len);
-	fnode->up = dir->i_ino;
+	fnode->up = cpu_to_le32(dir->i_ino);
 	mark_buffer_dirty(bh);
 
 	insert_inode_hash(result);
 
 	hpfs_write_inode_nolock(result);
 	d_instantiate(dentry, result);
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	brelse(bh);
 	hpfs_unlock(dir->i_sb);
 	return 0;
 bail2:
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	iput(result);
 bail1:
 	brelse(bh);
@@ -310,8 +300,8 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 	memset(&dee, 0, sizeof dee);
 	dee.archive = 1;
 	dee.hidden = name[0] == '.';
-	dee.fnode = fno;
-	dee.creation_date = dee.write_date = dee.read_date = gmt_to_local(dir->i_sb, get_seconds());
+	dee.fnode = cpu_to_le32(fno);
+	dee.creation_date = dee.write_date = dee.read_date = cpu_to_le32(gmt_to_local(dir->i_sb, get_seconds()));
 
 	result = new_inode(dir->i_sb);
 	if (!result)
@@ -319,7 +309,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 	result->i_ino = fno;
 	hpfs_init_inode(result);
 	hpfs_i(result)->i_parent_dir = dir->i_ino;
-	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, dee.creation_date);
+	result->i_ctime.tv_sec = result->i_mtime.tv_sec = result->i_atime.tv_sec = local_to_gmt(dir->i_sb, le32_to_cpu(dee.creation_date));
 	result->i_ctime.tv_nsec = 0;
 	result->i_mtime.tv_nsec = 0;
 	result->i_atime.tv_nsec = 0;
@@ -333,8 +323,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 	result->i_op = &page_symlink_inode_operations;
 	result->i_data.a_ops = &hpfs_symlink_aops;
 
-	mutex_lock(&hpfs_i(dir)->i_mutex);
-	r = hpfs_add_dirent(dir, name, len, &dee, 0);
+	r = hpfs_add_dirent(dir, name, len, &dee);
 	if (r == 1)
 		goto bail2;
 	if (r == -1) {
@@ -343,7 +332,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 	}
 	fnode->len = len;
 	memcpy(fnode->name, name, len > 15 ? 15 : len);
-	fnode->up = dir->i_ino;
+	fnode->up = cpu_to_le32(dir->i_ino);
 	hpfs_set_ea(result, fnode, "SYMLINK", symlink, strlen(symlink));
 	mark_buffer_dirty(bh);
 	brelse(bh);
@@ -352,11 +341,9 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
 
 	hpfs_write_inode_nolock(result);
 	d_instantiate(dentry, result);
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	hpfs_unlock(dir->i_sb);
 	return 0;
 bail2:
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
 	iput(result);
 bail1:
 	brelse(bh);
@@ -374,7 +361,6 @@ static int hpfs_unlink(struct inode *dir, struct dentry *dentry)
 	struct hpfs_dirent *de;
 	struct inode *inode = dentry->d_inode;
 	dnode_secno dno;
-	fnode_secno fno;
 	int r;
 	int rep = 0;
 	int err;
@@ -382,8 +368,6 @@ static int hpfs_unlink(struct inode *dir, struct dentry *dentry)
 	hpfs_lock(dir->i_sb);
 	hpfs_adjust_length(name, &len);
 again:
-	mutex_lock(&hpfs_i(inode)->i_parent_mutex);
-	mutex_lock(&hpfs_i(dir)->i_mutex);
 	err = -ENOENT;
 	de = map_dirent(dir, hpfs_i(dir)->i_dno, name, len, &dno, &qbh);
 	if (!de)
@@ -397,7 +381,6 @@ again:
 	if (de->directory)
 		goto out1;
 
-	fno = de->fnode;
 	r = hpfs_remove_dirent(dir, dno, de, &qbh, 1);
 	switch (r) {
 	case 1:
@@ -410,8 +393,6 @@ again:
 		if (rep++)
 			break;
 
-		mutex_unlock(&hpfs_i(dir)->i_mutex);
-		mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
 		dentry_unhash(dentry);
 		if (!d_unhashed(dentry)) {
 			dput(dentry);
@@ -445,8 +426,6 @@ again:
 out1:
 	hpfs_brelse4(&qbh);
 out:
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
-	mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
 	hpfs_unlock(dir->i_sb);
 	return err;
 }
@@ -459,15 +438,12 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 	struct hpfs_dirent *de;
 	struct inode *inode = dentry->d_inode;
 	dnode_secno dno;
-	fnode_secno fno;
 	int n_items = 0;
 	int err;
 	int r;
 
 	hpfs_adjust_length(name, &len);
 	hpfs_lock(dir->i_sb);
-	mutex_lock(&hpfs_i(inode)->i_parent_mutex);
-	mutex_lock(&hpfs_i(dir)->i_mutex);
 	err = -ENOENT;
 	de = map_dirent(dir, hpfs_i(dir)->i_dno, name, len, &dno, &qbh);
 	if (!de)
@@ -486,7 +462,6 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 	if (n_items)
 		goto out1;
 
-	fno = de->fnode;
 	r = hpfs_remove_dirent(dir, dno, de, &qbh, 1);
 	switch (r) {
 	case 1:
@@ -505,8 +480,6 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
 out1:
 	hpfs_brelse4(&qbh);
 out:
-	mutex_unlock(&hpfs_i(dir)->i_mutex);
-	mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
 	hpfs_unlock(dir->i_sb);
 	return err;
 }
@@ -568,12 +541,6 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	hpfs_lock(i->i_sb);
 	/* order doesn't matter, due to VFS exclusion */
-	mutex_lock(&hpfs_i(i)->i_parent_mutex);
-	if (new_inode)
-		mutex_lock(&hpfs_i(new_inode)->i_parent_mutex);
-	mutex_lock(&hpfs_i(old_dir)->i_mutex);
-	if (new_dir != old_dir)
-		mutex_lock(&hpfs_i(new_dir)->i_mutex);
 	
 	/* Erm? Moving over the empty non-busy directory is perfectly legal */
 	if (new_inode && S_ISDIR(new_inode->i_mode)) {
@@ -610,9 +577,7 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	if (new_dir == old_dir) hpfs_brelse4(&qbh);
 
-	hpfs_lock_creation(i->i_sb);
-	if ((r = hpfs_add_dirent(new_dir, new_name, new_len, &de, 1))) {
-		hpfs_unlock_creation(i->i_sb);
+	if ((r = hpfs_add_dirent(new_dir, new_name, new_len, &de))) {
 		if (r == -1) hpfs_error(new_dir->i_sb, "hpfs_rename: dirent already exists!");
 		err = r == 1 ? -ENOSPC : -EFSERROR;
 		if (new_dir != old_dir) hpfs_brelse4(&qbh);
@@ -621,20 +586,17 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	
 	if (new_dir == old_dir)
 		if (!(dep = map_dirent(old_dir, hpfs_i(old_dir)->i_dno, old_name, old_len, &dno, &qbh))) {
-			hpfs_unlock_creation(i->i_sb);
 			hpfs_error(i->i_sb, "lookup succeeded but map dirent failed at #2");
 			err = -ENOENT;
 			goto end1;
 		}
 
 	if ((r = hpfs_remove_dirent(old_dir, dno, dep, &qbh, 0))) {
-		hpfs_unlock_creation(i->i_sb);
 		hpfs_error(i->i_sb, "hpfs_rename: could not remove dirent");
 		err = r == 2 ? -ENOSPC : -EFSERROR;
 		goto end1;
 	}
-	hpfs_unlock_creation(i->i_sb);
-	
+
 	end:
 	hpfs_i(i)->i_parent_dir = new_dir->i_ino;
 	if (S_ISDIR(i->i_mode)) {
@@ -642,22 +604,14 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		drop_nlink(old_dir);
 	}
 	if ((fnode = hpfs_map_fnode(i->i_sb, i->i_ino, &bh))) {
-		fnode->up = new_dir->i_ino;
+		fnode->up = cpu_to_le32(new_dir->i_ino);
 		fnode->len = new_len;
 		memcpy(fnode->name, new_name, new_len>15?15:new_len);
 		if (new_len < 15) memset(&fnode->name[new_len], 0, 15 - new_len);
 		mark_buffer_dirty(bh);
 		brelse(bh);
 	}
-	hpfs_i(i)->i_conv = hpfs_sb(i->i_sb)->sb_conv;
-	hpfs_decide_conv(i, new_name, new_len);
 end1:
-	if (old_dir != new_dir)
-		mutex_unlock(&hpfs_i(new_dir)->i_mutex);
-	mutex_unlock(&hpfs_i(old_dir)->i_mutex);
-	mutex_unlock(&hpfs_i(i)->i_parent_mutex);
-	if (new_inode)
-		mutex_unlock(&hpfs_i(new_inode)->i_parent_mutex);
 	hpfs_unlock(i->i_sb);
 	return err;
 }
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index c89b4080858..98580a3b500 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -18,15 +18,16 @@
 
 /* Mark the filesystem dirty, so that chkdsk checks it when os/2 booted */
 
-static void mark_dirty(struct super_block *s)
+static void mark_dirty(struct super_block *s, int remount)
 {
-	if (hpfs_sb(s)->sb_chkdsk && !(s->s_flags & MS_RDONLY)) {
+	if (hpfs_sb(s)->sb_chkdsk && (remount || !(s->s_flags & MS_RDONLY))) {
 		struct buffer_head *bh;
 		struct hpfs_spare_block *sb;
 		if ((sb = hpfs_map_sector(s, 17, &bh, 0))) {
 			sb->dirty = 1;
 			sb->old_wrote = 0;
 			mark_buffer_dirty(bh);
+			sync_dirty_buffer(bh);
 			brelse(bh);
 		}
 	}
@@ -40,10 +41,12 @@ static void unmark_dirty(struct super_block *s)
 	struct buffer_head *bh;
 	struct hpfs_spare_block *sb;
 	if (s->s_flags & MS_RDONLY) return;
+	sync_blockdev(s->s_bdev);
 	if ((sb = hpfs_map_sector(s, 17, &bh, 0))) {
 		sb->dirty = hpfs_sb(s)->sb_chkdsk > 1 - hpfs_sb(s)->sb_was_error;
 		sb->old_wrote = hpfs_sb(s)->sb_chkdsk >= 2 && !hpfs_sb(s)->sb_was_error;
 		mark_buffer_dirty(bh);
+		sync_dirty_buffer(bh);
 		brelse(bh);
 	}
 }
@@ -63,13 +66,13 @@ void hpfs_error(struct super_block *s, const char *fmt, ...)
 	if (!hpfs_sb(s)->sb_was_error) {
 		if (hpfs_sb(s)->sb_err == 2) {
 			printk("; crashing the system because you wanted it\n");
-			mark_dirty(s);
+			mark_dirty(s, 0);
 			panic("HPFS panic");
 		} else if (hpfs_sb(s)->sb_err == 1) {
 			if (s->s_flags & MS_RDONLY) printk("; already mounted read-only\n");
 			else {
 				printk("; remounting read-only\n");
-				mark_dirty(s);
+				mark_dirty(s, 0);
 				s->s_flags |= MS_RDONLY;
 			}
 		} else if (s->s_flags & MS_RDONLY) printk("; going on - but anything won't be destroyed because it's read-only\n");
@@ -102,9 +105,12 @@ static void hpfs_put_super(struct super_block *s)
 {
 	struct hpfs_sb_info *sbi = hpfs_sb(s);
 
+	hpfs_lock(s);
+	unmark_dirty(s);
+	hpfs_unlock(s);
+
 	kfree(sbi->sb_cp_table);
 	kfree(sbi->sb_bmp_dir);
-	unmark_dirty(s);
 	s->s_fs_info = NULL;
 	kfree(sbi);
 }
@@ -129,7 +135,7 @@ static unsigned count_bitmaps(struct super_block *s)
 	n_bands = (hpfs_sb(s)->sb_fs_size + 0x3fff) >> 14;
 	count = 0;
 	for (n = 0; n < n_bands; n++)
-		count += hpfs_count_one_bitmap(s, hpfs_sb(s)->sb_bmp_dir[n]);
+		count += hpfs_count_one_bitmap(s, le32_to_cpu(hpfs_sb(s)->sb_bmp_dir[n]));
 	return count;
 }
 
@@ -188,8 +194,6 @@ static void init_once(void *foo)
 {
 	struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo;
 
-	mutex_init(&ei->i_mutex);
-	mutex_init(&ei->i_parent_mutex);
 	inode_init_once(&ei->vfs_inode);
 }
 
@@ -218,7 +222,6 @@ static void destroy_inodecache(void)
 
 enum {
 	Opt_help, Opt_uid, Opt_gid, Opt_umask, Opt_case_lower, Opt_case_asis,
-	Opt_conv_binary, Opt_conv_text, Opt_conv_auto,
 	Opt_check_none, Opt_check_normal, Opt_check_strict,
 	Opt_err_cont, Opt_err_ro, Opt_err_panic,
 	Opt_eas_no, Opt_eas_ro, Opt_eas_rw,
@@ -233,9 +236,6 @@ static const match_table_t tokens = {
 	{Opt_umask, "umask=%o"},
 	{Opt_case_lower, "case=lower"},
 	{Opt_case_asis, "case=asis"},
-	{Opt_conv_binary, "conv=binary"},
-	{Opt_conv_text, "conv=text"},
-	{Opt_conv_auto, "conv=auto"},
 	{Opt_check_none, "check=none"},
 	{Opt_check_normal, "check=normal"},
 	{Opt_check_strict, "check=strict"},
@@ -253,7 +253,7 @@ static const match_table_t tokens = {
 };
 
 static int parse_opts(char *opts, uid_t *uid, gid_t *gid, umode_t *umask,
-		      int *lowercase, int *conv, int *eas, int *chk, int *errs,
+		      int *lowercase, int *eas, int *chk, int *errs,
 		      int *chkdsk, int *timeshift)
 {
 	char *p;
@@ -295,15 +295,6 @@ static int parse_opts(char *opts, uid_t *uid, gid_t *gid, umode_t *umask,
 		case Opt_case_asis:
 			*lowercase = 0;
 			break;
-		case Opt_conv_binary:
-			*conv = CONV_BINARY;
-			break;
-		case Opt_conv_text:
-			*conv = CONV_TEXT;
-			break;
-		case Opt_conv_auto:
-			*conv = CONV_AUTO;
-			break;
 		case Opt_check_none:
 			*chk = 0;
 			break;
@@ -370,9 +361,6 @@ HPFS filesystem options:\n\
       umask=xxx         set mode of files that don't have mode specified in eas\n\
       case=lower        lowercase all files\n\
       case=asis         do not lowercase files (default)\n\
-      conv=binary       do not convert CR/LF -> LF (default)\n\
-      conv=auto         convert only files with known text extensions\n\
-      conv=text         convert all files\n\
       check=none        no fs checks - kernel may crash on corrupted filesystem\n\
       check=normal      do some checks - it should not crash (default)\n\
       check=strict      do extra time-consuming checks, used for debugging\n\
@@ -394,7 +382,7 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data)
 	uid_t uid;
 	gid_t gid;
 	umode_t umask;
-	int lowercase, conv, eas, chk, errs, chkdsk, timeshift;
+	int lowercase, eas, chk, errs, chkdsk, timeshift;
 	int o;
 	struct hpfs_sb_info *sbi = hpfs_sb(s);
 	char *new_opts = kstrdup(data, GFP_KERNEL);
@@ -405,11 +393,11 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data)
 	lock_super(s);
 	uid = sbi->sb_uid; gid = sbi->sb_gid;
 	umask = 0777 & ~sbi->sb_mode;
-	lowercase = sbi->sb_lowercase; conv = sbi->sb_conv;
+	lowercase = sbi->sb_lowercase;
 	eas = sbi->sb_eas; chk = sbi->sb_chk; chkdsk = sbi->sb_chkdsk;
 	errs = sbi->sb_err; timeshift = sbi->sb_timeshift;
 
-	if (!(o = parse_opts(data, &uid, &gid, &umask, &lowercase, &conv,
+	if (!(o = parse_opts(data, &uid, &gid, &umask, &lowercase,
 	    &eas, &chk, &errs, &chkdsk, &timeshift))) {
 		printk("HPFS: bad mount options.\n");
 		goto out_err;
@@ -427,11 +415,11 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data)
 
 	sbi->sb_uid = uid; sbi->sb_gid = gid;
 	sbi->sb_mode = 0777 & ~umask;
-	sbi->sb_lowercase = lowercase; sbi->sb_conv = conv;
+	sbi->sb_lowercase = lowercase;
 	sbi->sb_eas = eas; sbi->sb_chk = chk; sbi->sb_chkdsk = chkdsk;
 	sbi->sb_err = errs; sbi->sb_timeshift = timeshift;
 
-	if (!(*flags & MS_RDONLY)) mark_dirty(s);
+	if (!(*flags & MS_RDONLY)) mark_dirty(s, 1);
 
 	replace_mount_options(s, new_opts);
 
@@ -471,7 +459,7 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 	uid_t uid;
 	gid_t gid;
 	umode_t umask;
-	int lowercase, conv, eas, chk, errs, chkdsk, timeshift;
+	int lowercase, eas, chk, errs, chkdsk, timeshift;
 
 	dnode_secno root_dno;
 	struct hpfs_dirent *de = NULL;
@@ -479,11 +467,6 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 
 	int o;
 
-	if (num_possible_cpus() > 1) {
-		printk(KERN_ERR "HPFS is not SMP safe\n");
-		return -EINVAL;
-	}
-
 	save_mount_options(s, options);
 
 	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
@@ -495,20 +478,20 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 	sbi->sb_bmp_dir = NULL;
 	sbi->sb_cp_table = NULL;
 
-	mutex_init(&sbi->hpfs_creation_de);
+	mutex_init(&sbi->hpfs_mutex);
+	hpfs_lock(s);
 
 	uid = current_uid();
 	gid = current_gid();
 	umask = current_umask();
 	lowercase = 0;
-	conv = CONV_BINARY;
 	eas = 2;
 	chk = 1;
 	errs = 1;
 	chkdsk = 1;
 	timeshift = 0;
 
-	if (!(o = parse_opts(options, &uid, &gid, &umask, &lowercase, &conv,
+	if (!(o = parse_opts(options, &uid, &gid, &umask, &lowercase,
 	    &eas, &chk, &errs, &chkdsk, &timeshift))) {
 		printk("HPFS: bad mount options.\n");
 		goto bail0;
@@ -526,9 +509,9 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 	if (!(spareblock = hpfs_map_sector(s, 17, &bh2, 0))) goto bail3;
 
 	/* Check magics */
-	if (/*bootblock->magic != BB_MAGIC
-	    ||*/ superblock->magic != SB_MAGIC
-	    || spareblock->magic != SP_MAGIC) {
+	if (/*le16_to_cpu(bootblock->magic) != BB_MAGIC
+	    ||*/ le32_to_cpu(superblock->magic) != SB_MAGIC
+	    || le32_to_cpu(spareblock->magic) != SP_MAGIC) {
 		if (!silent) printk("HPFS: Bad magic ... probably not HPFS\n");
 		goto bail4;
 	}
@@ -549,19 +532,18 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 	s->s_op = &hpfs_sops;
 	s->s_d_op = &hpfs_dentry_operations;
 
-	sbi->sb_root = superblock->root;
-	sbi->sb_fs_size = superblock->n_sectors;
-	sbi->sb_bitmaps = superblock->bitmaps;
-	sbi->sb_dirband_start = superblock->dir_band_start;
-	sbi->sb_dirband_size = superblock->n_dir_band;
-	sbi->sb_dmap = superblock->dir_band_bitmap;
+	sbi->sb_root = le32_to_cpu(superblock->root);
+	sbi->sb_fs_size = le32_to_cpu(superblock->n_sectors);
+	sbi->sb_bitmaps = le32_to_cpu(superblock->bitmaps);
+	sbi->sb_dirband_start = le32_to_cpu(superblock->dir_band_start);
+	sbi->sb_dirband_size = le32_to_cpu(superblock->n_dir_band);
+	sbi->sb_dmap = le32_to_cpu(superblock->dir_band_bitmap);
 	sbi->sb_uid = uid;
 	sbi->sb_gid = gid;
 	sbi->sb_mode = 0777 & ~umask;
 	sbi->sb_n_free = -1;
 	sbi->sb_n_free_dnodes = -1;
 	sbi->sb_lowercase = lowercase;
-	sbi->sb_conv = conv;
 	sbi->sb_eas = eas;
 	sbi->sb_chk = chk;
 	sbi->sb_chkdsk = chkdsk;
@@ -573,7 +555,7 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 	sbi->sb_max_fwd_alloc = 0xffffff;
 	
 	/* Load bitmap directory */
-	if (!(sbi->sb_bmp_dir = hpfs_load_bitmap_directory(s, superblock->bitmaps)))
+	if (!(sbi->sb_bmp_dir = hpfs_load_bitmap_directory(s, le32_to_cpu(superblock->bitmaps))))
 		goto bail4;
 	
 	/* Check for general fs errors*/
@@ -591,20 +573,20 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 		mark_buffer_dirty(bh2);
 	}
 
-	if (spareblock->hotfixes_used || spareblock->n_spares_used) {
+	if (le32_to_cpu(spareblock->hotfixes_used) || le32_to_cpu(spareblock->n_spares_used)) {
 		if (errs >= 2) {
 			printk("HPFS: Hotfixes not supported here, try chkdsk\n");
-			mark_dirty(s);
+			mark_dirty(s, 0);
 			goto bail4;
 		}
 		hpfs_error(s, "hotfixes not supported here, try chkdsk");
 		if (errs == 0) printk("HPFS: Proceeding, but your filesystem will be probably corrupted by this driver...\n");
 		else printk("HPFS: This driver may read bad files or crash when operating on disk with hotfixes.\n");
 	}
-	if (spareblock->n_dnode_spares != spareblock->n_dnode_spares_free) {
+	if (le32_to_cpu(spareblock->n_dnode_spares) != le32_to_cpu(spareblock->n_dnode_spares_free)) {
 		if (errs >= 2) {
 			printk("HPFS: Spare dnodes used, try chkdsk\n");
-			mark_dirty(s);
+			mark_dirty(s, 0);
 			goto bail4;
 		}
 		hpfs_error(s, "warning: spare dnodes used, try chkdsk");
@@ -612,26 +594,26 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 	}
 	if (chk) {
 		unsigned a;
-		if (superblock->dir_band_end - superblock->dir_band_start + 1 != superblock->n_dir_band ||
-		    superblock->dir_band_end < superblock->dir_band_start || superblock->n_dir_band > 0x4000) {
+		if (le32_to_cpu(superblock->dir_band_end) - le32_to_cpu(superblock->dir_band_start) + 1 != le32_to_cpu(superblock->n_dir_band) ||
+		    le32_to_cpu(superblock->dir_band_end) < le32_to_cpu(superblock->dir_band_start) || le32_to_cpu(superblock->n_dir_band) > 0x4000) {
 			hpfs_error(s, "dir band size mismatch: dir_band_start==%08x, dir_band_end==%08x, n_dir_band==%08x",
-				superblock->dir_band_start, superblock->dir_band_end, superblock->n_dir_band);
+				le32_to_cpu(superblock->dir_band_start), le32_to_cpu(superblock->dir_band_end), le32_to_cpu(superblock->n_dir_band));
 			goto bail4;
 		}
 		a = sbi->sb_dirband_size;
 		sbi->sb_dirband_size = 0;
-		if (hpfs_chk_sectors(s, superblock->dir_band_start, superblock->n_dir_band, "dir_band") ||
-		    hpfs_chk_sectors(s, superblock->dir_band_bitmap, 4, "dir_band_bitmap") ||
-		    hpfs_chk_sectors(s, superblock->bitmaps, 4, "bitmaps")) {
-			mark_dirty(s);
+		if (hpfs_chk_sectors(s, le32_to_cpu(superblock->dir_band_start), le32_to_cpu(superblock->n_dir_band), "dir_band") ||
+		    hpfs_chk_sectors(s, le32_to_cpu(superblock->dir_band_bitmap), 4, "dir_band_bitmap") ||
+		    hpfs_chk_sectors(s, le32_to_cpu(superblock->bitmaps), 4, "bitmaps")) {
+			mark_dirty(s, 0);
 			goto bail4;
 		}
 		sbi->sb_dirband_size = a;
 	} else printk("HPFS: You really don't want any checks? You are crazy...\n");
 
 	/* Load code page table */
-	if (spareblock->n_code_pages)
-		if (!(sbi->sb_cp_table = hpfs_load_code_page(s, spareblock->code_page_dir)))
+	if (le32_to_cpu(spareblock->n_code_pages))
+		if (!(sbi->sb_cp_table = hpfs_load_code_page(s, le32_to_cpu(spareblock->code_page_dir))))
 			printk("HPFS: Warning: code page support is disabled\n");
 
 	brelse(bh2);
@@ -660,13 +642,13 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 	if (!de)
 		hpfs_error(s, "unable to find root dir");
 	else {
-		root->i_atime.tv_sec = local_to_gmt(s, de->read_date);
+		root->i_atime.tv_sec = local_to_gmt(s, le32_to_cpu(de->read_date));
 		root->i_atime.tv_nsec = 0;
-		root->i_mtime.tv_sec = local_to_gmt(s, de->write_date);
+		root->i_mtime.tv_sec = local_to_gmt(s, le32_to_cpu(de->write_date));
 		root->i_mtime.tv_nsec = 0;
-		root->i_ctime.tv_sec = local_to_gmt(s, de->creation_date);
+		root->i_ctime.tv_sec = local_to_gmt(s, le32_to_cpu(de->creation_date));
 		root->i_ctime.tv_nsec = 0;
-		hpfs_i(root)->i_ea_size = de->ea_size;
+		hpfs_i(root)->i_ea_size = le16_to_cpu(de->ea_size);
 		hpfs_i(root)->i_parent_dir = root->i_ino;
 		if (root->i_size == -1)
 			root->i_size = 2048;
@@ -674,6 +656,7 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 			root->i_blocks = 5;
 		hpfs_brelse4(&qbh);
 	}
+	hpfs_unlock(s);
 	return 0;
 
 bail4:	brelse(bh2);
@@ -681,6 +664,7 @@ bail3:	brelse(bh1);
 bail2:	brelse(bh0);
 bail1:
 bail0:
+	hpfs_unlock(s);
 	kfree(sbi->sb_bmp_dir);
 	kfree(sbi->sb_cp_table);
 	s->s_fs_info = NULL;
diff --git a/fs/inode.c b/fs/inode.c
index 33c963d08ab..05f4fa52132 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -24,6 +24,7 @@
 #include <linux/mount.h>
 #include <linux/async.h>
 #include <linux/posix_acl.h>
+#include <linux/prefetch.h>
 #include <linux/ima.h>
 #include <linux/cred.h>
 #include "internal.h"
diff --git a/fs/logfs/dev_bdev.c b/fs/logfs/dev_bdev.c
index 1adc8d455f0..df0de27c273 100644
--- a/fs/logfs/dev_bdev.c
+++ b/fs/logfs/dev_bdev.c
@@ -10,6 +10,7 @@
 #include <linux/blkdev.h>
 #include <linux/buffer_head.h>
 #include <linux/gfp.h>
+#include <linux/prefetch.h>
 
 #define PAGE_OFS(ofs) ((ofs) & (PAGE_SIZE-1))
 
diff --git a/fs/logfs/super.c b/fs/logfs/super.c
index 33435e4b14d..ce03a182c77 100644
--- a/fs/logfs/super.c
+++ b/fs/logfs/super.c
@@ -480,10 +480,6 @@ static int logfs_read_sb(struct super_block *sb, int read_only)
 			!read_only)
 		return -EIO;
 
-	mutex_init(&super->s_dirop_mutex);
-	mutex_init(&super->s_object_alias_mutex);
-	INIT_LIST_HEAD(&super->s_freeing_list);
-
 	ret = logfs_init_rw(sb);
 	if (ret)
 		return ret;
@@ -601,6 +597,10 @@ static struct dentry *logfs_mount(struct file_system_type *type, int flags,
 	if (!super)
 		return ERR_PTR(-ENOMEM);
 
+	mutex_init(&super->s_dirop_mutex);
+	mutex_init(&super->s_object_alias_mutex);
+	INIT_LIST_HEAD(&super->s_freeing_list);
+
 	if (!devname)
 		err = logfs_get_sb_bdev(super, type, devname);
 	else if (strncmp(devname, "mtd", 3))
diff --git a/fs/namei.c b/fs/namei.c
index 54fc993e302..6ff858c049c 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -179,7 +179,7 @@ EXPORT_SYMBOL(putname);
 static int acl_permission_check(struct inode *inode, int mask, unsigned int flags,
 		int (*check_acl)(struct inode *inode, int mask, unsigned int flags))
 {
-	umode_t			mode = inode->i_mode;
+	unsigned int mode = inode->i_mode;
 
 	mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
 
@@ -1378,12 +1378,12 @@ 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;
 	}
+	BUG_ON(nd->depth >= MAX_NESTED_LINKS);
 
 	nd->depth++;
 	current->link_count++;
diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index 89fc160fd5b..1f063bacd28 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -119,7 +119,7 @@ Elong:
 }
 
 #ifdef CONFIG_NFS_V4
-static rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors, struct inode *inode)
+static rpc_authflavor_t nfs_find_best_sec(struct nfs4_secinfo_flavors *flavors)
 {
 	struct gss_api_mech *mech;
 	struct xdr_netobj oid;
@@ -166,7 +166,7 @@ static int nfs_negotiate_security(const struct dentry *parent,
 		}
 		flavors = page_address(page);
 		ret = secinfo(parent->d_inode, &dentry->d_name, flavors);
-		*flavor = nfs_find_best_sec(flavors, dentry->d_inode);
+		*flavor = nfs_find_best_sec(flavors);
 		put_page(page);
 	}
 
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index e1c261ddd65..c4a69833dd0 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -47,6 +47,7 @@ enum nfs4_client_state {
 	NFS4CLNT_LAYOUTRECALL,
 	NFS4CLNT_SESSION_RESET,
 	NFS4CLNT_RECALL_SLOT,
+	NFS4CLNT_LEASE_CONFIRM,
 };
 
 enum nfs4_session_state {
diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c
index 6f8192f4cfc..be79dc9f386 100644
--- a/fs/nfs/nfs4filelayout.c
+++ b/fs/nfs/nfs4filelayout.c
@@ -117,6 +117,8 @@ static int filelayout_async_handle_error(struct rpc_task *task,
 	case -EKEYEXPIRED:
 		rpc_delay(task, FILELAYOUT_POLL_RETRY_MAX);
 		break;
+	case -NFS4ERR_RETRY_UNCACHED_REP:
+		break;
 	default:
 		dprintk("%s DS error. Retry through MDS %d\n", __func__,
 			task->tk_status);
@@ -416,7 +418,8 @@ static int
 filelayout_check_layout(struct pnfs_layout_hdr *lo,
 			struct nfs4_filelayout_segment *fl,
 			struct nfs4_layoutget_res *lgr,
-			struct nfs4_deviceid *id)
+			struct nfs4_deviceid *id,
+			gfp_t gfp_flags)
 {
 	struct nfs4_file_layout_dsaddr *dsaddr;
 	int status = -EINVAL;
@@ -439,7 +442,7 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo,
 	/* find and reference the deviceid */
 	dsaddr = nfs4_fl_find_get_deviceid(id);
 	if (dsaddr == NULL) {
-		dsaddr = get_device_info(lo->plh_inode, id);
+		dsaddr = get_device_info(lo->plh_inode, id, gfp_flags);
 		if (dsaddr == NULL)
 			goto out;
 	}
@@ -500,7 +503,8 @@ static int
 filelayout_decode_layout(struct pnfs_layout_hdr *flo,
 			 struct nfs4_filelayout_segment *fl,
 			 struct nfs4_layoutget_res *lgr,
-			 struct nfs4_deviceid *id)
+			 struct nfs4_deviceid *id,
+			 gfp_t gfp_flags)
 {
 	struct xdr_stream stream;
 	struct xdr_buf buf = {
@@ -516,7 +520,7 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo,
 
 	dprintk("%s: set_layout_map Begin\n", __func__);
 
-	scratch = alloc_page(GFP_KERNEL);
+	scratch = alloc_page(gfp_flags);
 	if (!scratch)
 		return -ENOMEM;
 
@@ -554,13 +558,13 @@ filelayout_decode_layout(struct pnfs_layout_hdr *flo,
 		goto out_err;
 
 	fl->fh_array = kzalloc(fl->num_fh * sizeof(struct nfs_fh *),
-			       GFP_KERNEL);
+			       gfp_flags);
 	if (!fl->fh_array)
 		goto out_err;
 
 	for (i = 0; i < fl->num_fh; i++) {
 		/* Do we want to use a mempool here? */
-		fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), GFP_KERNEL);
+		fl->fh_array[i] = kmalloc(sizeof(struct nfs_fh), gfp_flags);
 		if (!fl->fh_array[i])
 			goto out_err_free;
 
@@ -605,19 +609,20 @@ filelayout_free_lseg(struct pnfs_layout_segment *lseg)
 
 static struct pnfs_layout_segment *
 filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
-		      struct nfs4_layoutget_res *lgr)
+		      struct nfs4_layoutget_res *lgr,
+		      gfp_t gfp_flags)
 {
 	struct nfs4_filelayout_segment *fl;
 	int rc;
 	struct nfs4_deviceid id;
 
 	dprintk("--> %s\n", __func__);
-	fl = kzalloc(sizeof(*fl), GFP_KERNEL);
+	fl = kzalloc(sizeof(*fl), gfp_flags);
 	if (!fl)
 		return NULL;
 
-	rc = filelayout_decode_layout(layoutid, fl, lgr, &id);
-	if (rc != 0 || filelayout_check_layout(layoutid, fl, lgr, &id)) {
+	rc = filelayout_decode_layout(layoutid, fl, lgr, &id, gfp_flags);
+	if (rc != 0 || filelayout_check_layout(layoutid, fl, lgr, &id, gfp_flags)) {
 		_filelayout_free_lseg(fl);
 		return NULL;
 	}
@@ -633,7 +638,7 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid,
 		int size = (fl->stripe_type == STRIPE_SPARSE) ?
 			fl->dsaddr->ds_num : fl->dsaddr->stripe_count;
 
-		fl->commit_buckets = kcalloc(size, sizeof(struct list_head), GFP_KERNEL);
+		fl->commit_buckets = kcalloc(size, sizeof(struct list_head), gfp_flags);
 		if (!fl->commit_buckets) {
 			filelayout_free_lseg(&fl->generic_hdr);
 			return NULL;
diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/nfs4filelayout.h
index 7c44579f583..2b461d77b43 100644
--- a/fs/nfs/nfs4filelayout.h
+++ b/fs/nfs/nfs4filelayout.h
@@ -104,6 +104,6 @@ extern struct nfs4_file_layout_dsaddr *
 nfs4_fl_find_get_deviceid(struct nfs4_deviceid *dev_id);
 extern void nfs4_fl_put_deviceid(struct nfs4_file_layout_dsaddr *dsaddr);
 struct nfs4_file_layout_dsaddr *
-get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id);
+get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags);
 
 #endif /* FS_NFS_NFS4FILELAYOUT_H */
diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/nfs4filelayoutdev.c
index de5350f2b24..db07c7af139 100644
--- a/fs/nfs/nfs4filelayoutdev.c
+++ b/fs/nfs/nfs4filelayoutdev.c
@@ -225,11 +225,11 @@ nfs4_fl_free_deviceid(struct nfs4_file_layout_dsaddr *dsaddr)
 }
 
 static struct nfs4_pnfs_ds *
-nfs4_pnfs_ds_add(struct inode *inode, u32 ip_addr, u32 port)
+nfs4_pnfs_ds_add(struct inode *inode, u32 ip_addr, u32 port, gfp_t gfp_flags)
 {
 	struct nfs4_pnfs_ds *tmp_ds, *ds;
 
-	ds = kzalloc(sizeof(*tmp_ds), GFP_KERNEL);
+	ds = kzalloc(sizeof(*tmp_ds), gfp_flags);
 	if (!ds)
 		goto out;
 
@@ -261,7 +261,7 @@ out:
  * Currently only support ipv4, and one multi-path address.
  */
 static struct nfs4_pnfs_ds *
-decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode)
+decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode, gfp_t gfp_flags)
 {
 	struct nfs4_pnfs_ds *ds = NULL;
 	char *buf;
@@ -303,7 +303,7 @@ decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode)
 			rlen);
 		goto out_err;
 	}
-	buf = kmalloc(rlen + 1, GFP_KERNEL);
+	buf = kmalloc(rlen + 1, gfp_flags);
 	if (!buf) {
 		dprintk("%s: Not enough memory\n", __func__);
 		goto out_err;
@@ -333,7 +333,7 @@ decode_and_add_ds(struct xdr_stream *streamp, struct inode *inode)
 	sscanf(pstr, "-%d-%d", &tmp[0], &tmp[1]);
 	port = htons((tmp[0] << 8) | (tmp[1]));
 
-	ds = nfs4_pnfs_ds_add(inode, ip_addr, port);
+	ds = nfs4_pnfs_ds_add(inode, ip_addr, port, gfp_flags);
 	dprintk("%s: Decoded address and port %s\n", __func__, buf);
 out_free:
 	kfree(buf);
@@ -343,7 +343,7 @@ out_err:
 
 /* Decode opaque device data and return the result */
 static struct nfs4_file_layout_dsaddr*
-decode_device(struct inode *ino, struct pnfs_device *pdev)
+decode_device(struct inode *ino, struct pnfs_device *pdev, gfp_t gfp_flags)
 {
 	int i;
 	u32 cnt, num;
@@ -362,7 +362,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev)
 	struct page *scratch;
 
 	/* set up xdr stream */
-	scratch = alloc_page(GFP_KERNEL);
+	scratch = alloc_page(gfp_flags);
 	if (!scratch)
 		goto out_err;
 
@@ -384,7 +384,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev)
 	}
 
 	/* read stripe indices */
-	stripe_indices = kcalloc(cnt, sizeof(u8), GFP_KERNEL);
+	stripe_indices = kcalloc(cnt, sizeof(u8), gfp_flags);
 	if (!stripe_indices)
 		goto out_err_free_scratch;
 
@@ -423,7 +423,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev)
 
 	dsaddr = kzalloc(sizeof(*dsaddr) +
 			(sizeof(struct nfs4_pnfs_ds *) * (num - 1)),
-			GFP_KERNEL);
+			gfp_flags);
 	if (!dsaddr)
 		goto out_err_free_stripe_indices;
 
@@ -452,7 +452,7 @@ decode_device(struct inode *ino, struct pnfs_device *pdev)
 		for (j = 0; j < mp_count; j++) {
 			if (j == 0) {
 				dsaddr->ds_list[i] = decode_and_add_ds(&stream,
-					ino);
+					ino, gfp_flags);
 				if (dsaddr->ds_list[i] == NULL)
 					goto out_err_free_deviceid;
 			} else {
@@ -503,12 +503,12 @@ out_err:
  * available devices.
  */
 static struct nfs4_file_layout_dsaddr *
-decode_and_add_device(struct inode *inode, struct pnfs_device *dev)
+decode_and_add_device(struct inode *inode, struct pnfs_device *dev, gfp_t gfp_flags)
 {
 	struct nfs4_file_layout_dsaddr *d, *new;
 	long hash;
 
-	new = decode_device(inode, dev);
+	new = decode_device(inode, dev, gfp_flags);
 	if (!new) {
 		printk(KERN_WARNING "%s: Could not decode or add device\n",
 			__func__);
@@ -537,7 +537,7 @@ decode_and_add_device(struct inode *inode, struct pnfs_device *dev)
  * of available devices, and return it.
  */
 struct nfs4_file_layout_dsaddr *
-get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id)
+get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id, gfp_t gfp_flags)
 {
 	struct pnfs_device *pdev = NULL;
 	u32 max_resp_sz;
@@ -556,17 +556,17 @@ get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id)
 	dprintk("%s inode %p max_resp_sz %u max_pages %d\n",
 		__func__, inode, max_resp_sz, max_pages);
 
-	pdev = kzalloc(sizeof(struct pnfs_device), GFP_KERNEL);
+	pdev = kzalloc(sizeof(struct pnfs_device), gfp_flags);
 	if (pdev == NULL)
 		return NULL;
 
-	pages = kzalloc(max_pages * sizeof(struct page *), GFP_KERNEL);
+	pages = kzalloc(max_pages * sizeof(struct page *), gfp_flags);
 	if (pages == NULL) {
 		kfree(pdev);
 		return NULL;
 	}
 	for (i = 0; i < max_pages; i++) {
-		pages[i] = alloc_page(GFP_KERNEL);
+		pages[i] = alloc_page(gfp_flags);
 		if (!pages[i])
 			goto out_free;
 	}
@@ -587,7 +587,7 @@ get_device_info(struct inode *inode, struct nfs4_deviceid *dev_id)
 	 * Found new device, need to decode it and then add it to the
 	 * list of known devices for this mountpoint.
 	 */
-	dsaddr = decode_and_add_device(inode, pdev);
+	dsaddr = decode_and_add_device(inode, pdev, gfp_flags);
 out_free:
 	for (i = 0; i < max_pages; i++)
 		__free_page(pages[i]);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 9bf41eab3e4..cf1b339c393 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -46,6 +46,7 @@
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_page.h>
+#include <linux/nfs_mount.h>
 #include <linux/namei.h>
 #include <linux/mount.h>
 #include <linux/module.h>
@@ -299,6 +300,7 @@ static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struc
 			ret = nfs4_delay(server->client, &exception->timeout);
 			if (ret != 0)
 				break;
+		case -NFS4ERR_RETRY_UNCACHED_REP:
 		case -NFS4ERR_OLD_STATEID:
 			exception->retry = 1;
 			break;
@@ -443,8 +445,8 @@ static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *
 	if (res->sr_status == 1)
 		res->sr_status = NFS_OK;
 
-	/* -ERESTARTSYS can result in skipping nfs41_sequence_setup */
-	if (!res->sr_slot)
+	/* don't increment the sequence number if the task wasn't sent */
+	if (!RPC_WAS_SENT(task))
 		goto out;
 
 	/* Check the SEQUENCE operation status */
@@ -2185,9 +2187,14 @@ static int nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle,
 	struct nfs4_exception exception = { };
 	int err;
 	do {
-		err = nfs4_handle_exception(server,
-				_nfs4_lookup_root(server, fhandle, info),
-				&exception);
+		err = _nfs4_lookup_root(server, fhandle, info);
+		switch (err) {
+		case 0:
+		case -NFS4ERR_WRONGSEC:
+			break;
+		default:
+			err = nfs4_handle_exception(server, err, &exception);
+		}
 	} while (exception.retry);
 	return err;
 }
@@ -2208,25 +2215,47 @@ out:
 	return ret;
 }
 
-/*
- * get the file handle for the "/" directory on the server
- */
-static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
+static int nfs4_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle,
 			      struct nfs_fsinfo *info)
 {
 	int i, len, status = 0;
-	rpc_authflavor_t flav_array[NFS_MAX_SECFLAVORS + 2];
+	rpc_authflavor_t flav_array[NFS_MAX_SECFLAVORS];
 
-	flav_array[0] = RPC_AUTH_UNIX;
-	len = gss_mech_list_pseudoflavors(&flav_array[1]);
-	flav_array[1+len] = RPC_AUTH_NULL;
-	len += 2;
+	len = gss_mech_list_pseudoflavors(&flav_array[0]);
+	flav_array[len] = RPC_AUTH_NULL;
+	len += 1;
 
 	for (i = 0; i < len; i++) {
 		status = nfs4_lookup_root_sec(server, fhandle, info, flav_array[i]);
-		if (status != -EPERM)
-			break;
+		if (status == -NFS4ERR_WRONGSEC || status == -EACCES)
+			continue;
+		break;
 	}
+	/*
+	 * -EACCESS could mean that the user doesn't have correct permissions
+	 * to access the mount.  It could also mean that we tried to mount
+	 * with a gss auth flavor, but rpc.gssd isn't running.  Either way,
+	 * existing mount programs don't handle -EACCES very well so it should
+	 * be mapped to -EPERM instead.
+	 */
+	if (status == -EACCES)
+		status = -EPERM;
+	return status;
+}
+
+/*
+ * get the file handle for the "/" directory on the server
+ */
+static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle,
+			      struct nfs_fsinfo *info)
+{
+	int status = nfs4_lookup_root(server, fhandle, info);
+	if ((status == -NFS4ERR_WRONGSEC) && !(server->flags & NFS_MOUNT_SECFLAVOUR))
+		/*
+		 * A status of -NFS4ERR_WRONGSEC will be mapped to -EPERM
+		 * by nfs4_map_errors() as this function exits.
+		 */
+		status = nfs4_find_root_sec(server, fhandle, info);
 	if (status == 0)
 		status = nfs4_server_capabilities(server, fhandle);
 	if (status == 0)
@@ -3667,6 +3696,7 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 			rpc_delay(task, NFS4_POLL_RETRY_MAX);
 			task->tk_status = 0;
 			return -EAGAIN;
+		case -NFS4ERR_RETRY_UNCACHED_REP:
 		case -NFS4ERR_OLD_STATEID:
 			task->tk_status = 0;
 			return -EAGAIN;
@@ -3723,21 +3753,20 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program,
 				sizeof(setclientid.sc_uaddr), "%s.%u.%u",
 				clp->cl_ipaddr, port >> 8, port & 255);
 
-		status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
+		status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 		if (status != -NFS4ERR_CLID_INUSE)
 			break;
-		if (signalled())
+		if (loop != 0) {
+			++clp->cl_id_uniquifier;
 			break;
-		if (loop++ & 1)
-			ssleep(clp->cl_lease_time / HZ + 1);
-		else
-			if (++clp->cl_id_uniquifier == 0)
-				break;
+		}
+		++loop;
+		ssleep(clp->cl_lease_time / HZ + 1);
 	}
 	return status;
 }
 
-static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp,
+int nfs4_proc_setclientid_confirm(struct nfs_client *clp,
 		struct nfs4_setclientid_res *arg,
 		struct rpc_cred *cred)
 {
@@ -3752,7 +3781,7 @@ static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp,
 	int status;
 
 	now = jiffies;
-	status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
+	status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 	if (status == 0) {
 		spin_lock(&clp->cl_lock);
 		clp->cl_lease_time = fsinfo.lease_time * HZ;
@@ -3762,26 +3791,6 @@ static int _nfs4_proc_setclientid_confirm(struct nfs_client *clp,
 	return status;
 }
 
-int nfs4_proc_setclientid_confirm(struct nfs_client *clp,
-		struct nfs4_setclientid_res *arg,
-		struct rpc_cred *cred)
-{
-	long timeout = 0;
-	int err;
-	do {
-		err = _nfs4_proc_setclientid_confirm(clp, arg, cred);
-		switch (err) {
-			case 0:
-				return err;
-			case -NFS4ERR_RESOURCE:
-				/* The IBM lawyers misread another document! */
-			case -NFS4ERR_DELAY:
-				err = nfs4_delay(clp->cl_rpcclient, &timeout);
-		}
-	} while (err == 0);
-	return err;
-}
-
 struct nfs4_delegreturndata {
 	struct nfs4_delegreturnargs args;
 	struct nfs4_delegreturnres res;
@@ -4786,7 +4795,7 @@ int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 				init_utsname()->domainname,
 				clp->cl_rpcclient->cl_auth->au_flavor);
 
-	status = rpc_call_sync(clp->cl_rpcclient, &msg, 0);
+	status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 	if (!status)
 		status = nfs4_check_cl_exchange_flags(clp->cl_exchange_flags);
 	dprintk("<-- %s status= %d\n", __func__, status);
@@ -4837,6 +4846,8 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
 		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
 		rpc_delay(task, NFS4_POLL_RETRY_MIN);
 		task->tk_status = 0;
+		/* fall through */
+	case -NFS4ERR_RETRY_UNCACHED_REP:
 		nfs_restart_rpc(task, data->clp);
 		return;
 	}
@@ -4869,7 +4880,8 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo)
 		.rpc_client = clp->cl_rpcclient,
 		.rpc_message = &msg,
 		.callback_ops = &nfs4_get_lease_time_ops,
-		.callback_data = &data
+		.callback_data = &data,
+		.flags = RPC_TASK_TIMEOUT,
 	};
 	int status;
 
@@ -5171,7 +5183,7 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
 	nfs4_init_channel_attrs(&args);
 	args.flags = (SESSION4_PERSIST | SESSION4_BACK_CHAN);
 
-	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
+	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 
 	if (!status)
 		/* Verify the session's negotiated channel_attrs values */
@@ -5194,20 +5206,10 @@ 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);
 
-	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);
-
+	status = _nfs4_proc_create_session(clp);
 	if (status)
 		goto out;
 
@@ -5248,7 +5250,7 @@ int nfs4_proc_destroy_session(struct nfs4_session *session)
 	msg.rpc_argp = session;
 	msg.rpc_resp = NULL;
 	msg.rpc_cred = NULL;
-	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, 0);
+	status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
 
 	if (status)
 		printk(KERN_WARNING
@@ -5481,6 +5483,8 @@ static int nfs41_reclaim_complete_handle_errors(struct rpc_task *task, struct nf
 		break;
 	case -NFS4ERR_DELAY:
 		rpc_delay(task, NFS4_POLL_RETRY_MAX);
+		/* fall through */
+	case -NFS4ERR_RETRY_UNCACHED_REP:
 		return -EAGAIN;
 	default:
 		nfs4_schedule_lease_recovery(clp);
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index a6804f704d9..036f5adc9e1 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -64,10 +64,15 @@ static LIST_HEAD(nfs4_clientid_list);
 
 int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 {
-	struct nfs4_setclientid_res clid;
+	struct nfs4_setclientid_res clid = {
+		.clientid = clp->cl_clientid,
+		.confirm = clp->cl_confirm,
+	};
 	unsigned short port;
 	int status;
 
+	if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state))
+		goto do_confirm;
 	port = nfs_callback_tcpport;
 	if (clp->cl_addr.ss_family == AF_INET6)
 		port = nfs_callback_tcpport6;
@@ -75,10 +80,14 @@ int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 	status = nfs4_proc_setclientid(clp, NFS4_CALLBACK, port, cred, &clid);
 	if (status != 0)
 		goto out;
+	clp->cl_clientid = clid.clientid;
+	clp->cl_confirm = clid.confirm;
+	set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
+do_confirm:
 	status = nfs4_proc_setclientid_confirm(clp, &clid, cred);
 	if (status != 0)
 		goto out;
-	clp->cl_clientid = clid.clientid;
+	clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
 	nfs4_schedule_state_renewal(clp);
 out:
 	return status;
@@ -230,13 +239,18 @@ int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	int status;
 
+	if (test_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state))
+		goto do_confirm;
 	nfs4_begin_drain_session(clp);
 	status = nfs4_proc_exchange_id(clp, cred);
 	if (status != 0)
 		goto out;
+	set_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
+do_confirm:
 	status = nfs4_proc_create_session(clp);
 	if (status != 0)
 		goto out;
+	clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
 	nfs41_setup_state_renewal(clp);
 	nfs_mark_client_ready(clp, NFS_CS_READY);
 out:
@@ -1584,20 +1598,23 @@ static int nfs4_recall_slot(struct nfs_client *clp) { return 0; }
  */
 static void nfs4_set_lease_expired(struct nfs_client *clp, int status)
 {
-	if (nfs4_has_session(clp)) {
-		switch (status) {
-		case -NFS4ERR_DELAY:
-		case -NFS4ERR_CLID_INUSE:
-		case -EAGAIN:
-			break;
+	switch (status) {
+	case -NFS4ERR_CLID_INUSE:
+	case -NFS4ERR_STALE_CLIENTID:
+		clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
+		break;
+	case -NFS4ERR_DELAY:
+	case -ETIMEDOUT:
+	case -EAGAIN:
+		ssleep(1);
+		break;
 
-		case -EKEYEXPIRED:
-			nfs4_warn_keyexpired(clp->cl_hostname);
-		case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery
-					 * in nfs4_exchange_id */
-		default:
-			return;
-		}
+	case -EKEYEXPIRED:
+		nfs4_warn_keyexpired(clp->cl_hostname);
+	case -NFS4ERR_NOT_SAME: /* FixMe: implement recovery
+				 * in nfs4_exchange_id */
+	default:
+		return;
 	}
 	set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 }
@@ -1607,7 +1624,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
 	int status = 0;
 
 	/* Ensure exclusive access to NFSv4 state */
-	for(;;) {
+	do {
 		if (test_and_clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state)) {
 			/* We're going to have to re-establish a clientid */
 			status = nfs4_reclaim_lease(clp);
@@ -1691,7 +1708,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			break;
 		if (test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) != 0)
 			break;
-	}
+	} while (atomic_read(&clp->cl_count) > 1);
 	return;
 out_error:
 	printk(KERN_WARNING "Error: state manager failed on NFSv4 server %s"
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index dddfb5795d7..c3ccd2c4683 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -1452,26 +1452,25 @@ static void encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args,
 
 static void encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg *readdir, struct rpc_rqst *req, struct compound_hdr *hdr)
 {
-	uint32_t attrs[2] = {0, 0};
+	uint32_t attrs[2] = {
+		FATTR4_WORD0_RDATTR_ERROR,
+		FATTR4_WORD1_MOUNTED_ON_FILEID,
+	};
 	uint32_t dircount = readdir->count >> 1;
 	__be32 *p;
 
 	if (readdir->plus) {
 		attrs[0] |= FATTR4_WORD0_TYPE|FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE|
-			FATTR4_WORD0_FSID|FATTR4_WORD0_FILEHANDLE;
+			FATTR4_WORD0_FSID|FATTR4_WORD0_FILEHANDLE|FATTR4_WORD0_FILEID;
 		attrs[1] |= FATTR4_WORD1_MODE|FATTR4_WORD1_NUMLINKS|FATTR4_WORD1_OWNER|
 			FATTR4_WORD1_OWNER_GROUP|FATTR4_WORD1_RAWDEV|
 			FATTR4_WORD1_SPACE_USED|FATTR4_WORD1_TIME_ACCESS|
 			FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY;
 		dircount >>= 1;
 	}
-	attrs[0] |= FATTR4_WORD0_RDATTR_ERROR|FATTR4_WORD0_FILEID;
-	attrs[1] |= FATTR4_WORD1_MOUNTED_ON_FILEID;
-	/* Switch to mounted_on_fileid if the server supports it */
-	if (readdir->bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID)
-		attrs[0] &= ~FATTR4_WORD0_FILEID;
-	else
-		attrs[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID;
+	/* Use mounted_on_fileid only if the server supports it */
+	if (!(readdir->bitmask[1] & FATTR4_WORD1_MOUNTED_ON_FILEID))
+		attrs[0] |= FATTR4_WORD0_FILEID;
 
 	p = reserve_space(xdr, 12+NFS4_VERIFIER_SIZE+20);
 	*p++ = cpu_to_be32(OP_READDIR);
@@ -3140,7 +3139,7 @@ static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitma
 			goto out_overflow;
 		xdr_decode_hyper(p, fileid);
 		bitmap[1] &= ~FATTR4_WORD1_MOUNTED_ON_FILEID;
-		ret = NFS_ATTR_FATTR_FILEID;
+		ret = NFS_ATTR_FATTR_MOUNTED_ON_FILEID;
 	}
 	dprintk("%s: fileid=%Lu\n", __func__, (unsigned long long)*fileid);
 	return ret;
@@ -4002,7 +4001,6 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap,
 {
 	int status;
 	umode_t fmode = 0;
-	uint64_t fileid;
 	uint32_t type;
 
 	status = decode_attr_type(xdr, bitmap, &type);
@@ -4101,13 +4099,10 @@ static int decode_getfattr_attrs(struct xdr_stream *xdr, uint32_t *bitmap,
 		goto xdr_error;
 	fattr->valid |= status;
 
-	status = decode_attr_mounted_on_fileid(xdr, bitmap, &fileid);
+	status = decode_attr_mounted_on_fileid(xdr, bitmap, &fattr->mounted_on_fileid);
 	if (status < 0)
 		goto xdr_error;
-	if (status != 0 && !(fattr->valid & status)) {
-		fattr->fileid = fileid;
-		fattr->valid |= status;
-	}
+	fattr->valid |= status;
 
 xdr_error:
 	dprintk("%s: xdr returned %d\n", __func__, -status);
@@ -4838,17 +4833,21 @@ static int decode_secinfo(struct xdr_stream *xdr, struct nfs4_secinfo_res *res)
 	struct nfs4_secinfo_flavor *sec_flavor;
 	int status;
 	__be32 *p;
-	int i;
+	int i, num_flavors;
 
 	status = decode_op_hdr(xdr, OP_SECINFO);
+	if (status)
+		goto out;
 	p = xdr_inline_decode(xdr, 4);
 	if (unlikely(!p))
 		goto out_overflow;
-	res->flavors->num_flavors = be32_to_cpup(p);
 
-	for (i = 0; i < res->flavors->num_flavors; i++) {
+	res->flavors->num_flavors = 0;
+	num_flavors = be32_to_cpup(p);
+
+	for (i = 0; i < num_flavors; i++) {
 		sec_flavor = &res->flavors->flavors[i];
-		if ((char *)&sec_flavor[1] - (char *)res > PAGE_SIZE)
+		if ((char *)&sec_flavor[1] - (char *)res->flavors > PAGE_SIZE)
 			break;
 
 		p = xdr_inline_decode(xdr, 4);
@@ -4857,13 +4856,15 @@ static int decode_secinfo(struct xdr_stream *xdr, struct nfs4_secinfo_res *res)
 		sec_flavor->flavor = be32_to_cpup(p);
 
 		if (sec_flavor->flavor == RPC_AUTH_GSS) {
-			if (decode_secinfo_gss(xdr, sec_flavor))
-				break;
+			status = decode_secinfo_gss(xdr, sec_flavor);
+			if (status)
+				goto out;
 		}
+		res->flavors->num_flavors++;
 	}
 
-	return 0;
-
+out:
+	return status;
 out_overflow:
 	print_overflow_msg(__func__, xdr);
 	return -EIO;
@@ -6408,7 +6409,9 @@ int nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry,
 	if (decode_getfattr_attrs(xdr, bitmap, entry->fattr, entry->fh,
 					entry->server, 1) < 0)
 		goto out_overflow;
-	if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID)
+	if (entry->fattr->valid & NFS_ATTR_FATTR_MOUNTED_ON_FILEID)
+		entry->ino = entry->fattr->mounted_on_fileid;
+	else if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID)
 		entry->ino = entry->fattr->fileid;
 
 	entry->d_type = DT_UNKNOWN;
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index d9ab97269ce..f57f5281a52 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -383,6 +383,7 @@ pnfs_destroy_all_layouts(struct nfs_client *clp)
 				plh_layouts);
 		dprintk("%s freeing layout for inode %lu\n", __func__,
 			lo->plh_inode->i_ino);
+		list_del_init(&lo->plh_layouts);
 		pnfs_destroy_layout(NFS_I(lo->plh_inode));
 	}
 }
@@ -466,7 +467,8 @@ pnfs_choose_layoutget_stateid(nfs4_stateid *dst, struct pnfs_layout_hdr *lo,
 static struct pnfs_layout_segment *
 send_layoutget(struct pnfs_layout_hdr *lo,
 	   struct nfs_open_context *ctx,
-	   u32 iomode)
+	   u32 iomode,
+	   gfp_t gfp_flags)
 {
 	struct inode *ino = lo->plh_inode;
 	struct nfs_server *server = NFS_SERVER(ino);
@@ -479,7 +481,7 @@ send_layoutget(struct pnfs_layout_hdr *lo,
 	dprintk("--> %s\n", __func__);
 
 	BUG_ON(ctx == NULL);
-	lgp = kzalloc(sizeof(*lgp), GFP_KERNEL);
+	lgp = kzalloc(sizeof(*lgp), gfp_flags);
 	if (lgp == NULL)
 		return NULL;
 
@@ -487,12 +489,12 @@ send_layoutget(struct pnfs_layout_hdr *lo,
 	max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
 	max_pages = max_resp_sz >> PAGE_SHIFT;
 
-	pages = kzalloc(max_pages * sizeof(struct page *), GFP_KERNEL);
+	pages = kzalloc(max_pages * sizeof(struct page *), gfp_flags);
 	if (!pages)
 		goto out_err_free;
 
 	for (i = 0; i < max_pages; i++) {
-		pages[i] = alloc_page(GFP_KERNEL);
+		pages[i] = alloc_page(gfp_flags);
 		if (!pages[i])
 			goto out_err_free;
 	}
@@ -508,6 +510,7 @@ send_layoutget(struct pnfs_layout_hdr *lo,
 	lgp->args.layout.pages = pages;
 	lgp->args.layout.pglen = max_pages * PAGE_SIZE;
 	lgp->lsegpp = &lseg;
+	lgp->gfp_flags = gfp_flags;
 
 	/* Synchronously retrieve layout information from server and
 	 * store in lseg.
@@ -665,11 +668,11 @@ pnfs_insert_layout(struct pnfs_layout_hdr *lo,
 }
 
 static struct pnfs_layout_hdr *
-alloc_init_layout_hdr(struct inode *ino)
+alloc_init_layout_hdr(struct inode *ino, gfp_t gfp_flags)
 {
 	struct pnfs_layout_hdr *lo;
 
-	lo = kzalloc(sizeof(struct pnfs_layout_hdr), GFP_KERNEL);
+	lo = kzalloc(sizeof(struct pnfs_layout_hdr), gfp_flags);
 	if (!lo)
 		return NULL;
 	atomic_set(&lo->plh_refcount, 1);
@@ -681,7 +684,7 @@ alloc_init_layout_hdr(struct inode *ino)
 }
 
 static struct pnfs_layout_hdr *
-pnfs_find_alloc_layout(struct inode *ino)
+pnfs_find_alloc_layout(struct inode *ino, gfp_t gfp_flags)
 {
 	struct nfs_inode *nfsi = NFS_I(ino);
 	struct pnfs_layout_hdr *new = NULL;
@@ -696,7 +699,7 @@ pnfs_find_alloc_layout(struct inode *ino)
 			return nfsi->layout;
 	}
 	spin_unlock(&ino->i_lock);
-	new = alloc_init_layout_hdr(ino);
+	new = alloc_init_layout_hdr(ino, gfp_flags);
 	spin_lock(&ino->i_lock);
 
 	if (likely(nfsi->layout == NULL))	/* Won the race? */
@@ -756,7 +759,8 @@ pnfs_find_lseg(struct pnfs_layout_hdr *lo, u32 iomode)
 struct pnfs_layout_segment *
 pnfs_update_layout(struct inode *ino,
 		   struct nfs_open_context *ctx,
-		   enum pnfs_iomode iomode)
+		   enum pnfs_iomode iomode,
+		   gfp_t gfp_flags)
 {
 	struct nfs_inode *nfsi = NFS_I(ino);
 	struct nfs_client *clp = NFS_SERVER(ino)->nfs_client;
@@ -767,7 +771,7 @@ pnfs_update_layout(struct inode *ino,
 	if (!pnfs_enabled_sb(NFS_SERVER(ino)))
 		return NULL;
 	spin_lock(&ino->i_lock);
-	lo = pnfs_find_alloc_layout(ino);
+	lo = pnfs_find_alloc_layout(ino, gfp_flags);
 	if (lo == NULL) {
 		dprintk("%s ERROR: can't get pnfs_layout_hdr\n", __func__);
 		goto out_unlock;
@@ -807,7 +811,7 @@ pnfs_update_layout(struct inode *ino,
 		spin_unlock(&clp->cl_lock);
 	}
 
-	lseg = send_layoutget(lo, ctx, iomode);
+	lseg = send_layoutget(lo, ctx, iomode, gfp_flags);
 	if (!lseg && first) {
 		spin_lock(&clp->cl_lock);
 		list_del_init(&lo->plh_layouts);
@@ -846,7 +850,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp)
 		goto out;
 	}
 	/* Inject layout blob into I/O device driver */
-	lseg = NFS_SERVER(ino)->pnfs_curr_ld->alloc_lseg(lo, res);
+	lseg = NFS_SERVER(ino)->pnfs_curr_ld->alloc_lseg(lo, res, lgp->gfp_flags);
 	if (!lseg || IS_ERR(lseg)) {
 		if (!lseg)
 			status = -ENOMEM;
@@ -899,7 +903,8 @@ static int pnfs_read_pg_test(struct nfs_pageio_descriptor *pgio,
 		/* This is first coelesce call for a series of nfs_pages */
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   prev->wb_context,
-						   IOMODE_READ);
+						   IOMODE_READ,
+						   GFP_KERNEL);
 	}
 	return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req);
 }
@@ -921,7 +926,8 @@ static int pnfs_write_pg_test(struct nfs_pageio_descriptor *pgio,
 		/* This is first coelesce call for a series of nfs_pages */
 		pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode,
 						   prev->wb_context,
-						   IOMODE_RW);
+						   IOMODE_RW,
+						   GFP_NOFS);
 	}
 	return NFS_SERVER(pgio->pg_inode)->pnfs_curr_ld->pg_test(pgio, prev, req);
 }
@@ -1004,6 +1010,7 @@ pnfs_set_layoutcommit(struct nfs_write_data *wdata)
 {
 	struct nfs_inode *nfsi = NFS_I(wdata->inode);
 	loff_t end_pos = wdata->args.offset + wdata->res.count;
+	bool mark_as_dirty = false;
 
 	spin_lock(&nfsi->vfs_inode.i_lock);
 	if (!test_and_set_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) {
@@ -1011,13 +1018,18 @@ pnfs_set_layoutcommit(struct nfs_write_data *wdata)
 		get_lseg(wdata->lseg);
 		wdata->lseg->pls_lc_cred =
 			get_rpccred(wdata->args.context->state->owner->so_cred);
-		mark_inode_dirty_sync(wdata->inode);
+		mark_as_dirty = true;
 		dprintk("%s: Set layoutcommit for inode %lu ",
 			__func__, wdata->inode->i_ino);
 	}
 	if (end_pos > wdata->lseg->pls_end_pos)
 		wdata->lseg->pls_end_pos = end_pos;
 	spin_unlock(&nfsi->vfs_inode.i_lock);
+
+	/* if pnfs_layoutcommit_inode() runs between inode locks, the next one
+	 * will be a noop because NFS_INO_LAYOUTCOMMIT will not be set */
+	if (mark_as_dirty)
+		mark_inode_dirty_sync(wdata->inode);
 }
 EXPORT_SYMBOL_GPL(pnfs_set_layoutcommit);
 
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index bc4827202e7..0c015bad9e7 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -70,7 +70,7 @@ struct pnfs_layoutdriver_type {
 	const u32 id;
 	const char *name;
 	struct module *owner;
-	struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr);
+	struct pnfs_layout_segment * (*alloc_lseg) (struct pnfs_layout_hdr *layoutid, struct nfs4_layoutget_res *lgr, gfp_t gfp_flags);
 	void (*free_lseg) (struct pnfs_layout_segment *lseg);
 
 	/* test for nfs page cache coalescing */
@@ -126,7 +126,7 @@ void get_layout_hdr(struct pnfs_layout_hdr *lo);
 void put_lseg(struct pnfs_layout_segment *lseg);
 struct pnfs_layout_segment *
 pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx,
-		   enum pnfs_iomode access_type);
+		   enum pnfs_iomode access_type, gfp_t gfp_flags);
 void set_pnfs_layoutdriver(struct nfs_server *, u32 id);
 void unset_pnfs_layoutdriver(struct nfs_server *);
 enum pnfs_try_status pnfs_try_to_write_data(struct nfs_write_data *,
@@ -245,7 +245,7 @@ static inline void put_lseg(struct pnfs_layout_segment *lseg)
 
 static inline struct pnfs_layout_segment *
 pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx,
-		   enum pnfs_iomode access_type)
+		   enum pnfs_iomode access_type, gfp_t gfp_flags)
 {
 	return NULL;
 }
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 7cded2b12a0..2bcf0dc306a 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -288,7 +288,7 @@ static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc)
 	atomic_set(&req->wb_complete, requests);
 
 	BUG_ON(desc->pg_lseg != NULL);
-	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ);
+	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ, GFP_KERNEL);
 	ClearPageError(page);
 	offset = 0;
 	nbytes = desc->pg_count;
@@ -351,7 +351,7 @@ static int nfs_pagein_one(struct nfs_pageio_descriptor *desc)
 	}
 	req = nfs_list_entry(data->pages.next);
 	if ((!lseg) && list_is_singular(&data->pages))
-		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ);
+		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_READ, GFP_KERNEL);
 
 	ret = nfs_read_rpcsetup(req, data, &nfs_read_full_ops, desc->pg_count,
 				0, lseg);
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 2b8e9a5e366..e288f06d3fa 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -1004,6 +1004,7 @@ static int nfs_parse_security_flavors(char *value,
 		return 0;
 	}
 
+	mnt->flags |= NFS_MOUNT_SECFLAVOUR;
 	mnt->auth_flavor_len = 1;
 	return 1;
 }
@@ -1976,6 +1977,15 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
 	if (error < 0)
 		goto out;
 
+	/*
+	 * noac is a special case. It implies -o sync, but that's not
+	 * necessarily reflected in the mtab options. do_remount_sb
+	 * will clear MS_SYNCHRONOUS if -o sync wasn't specified in the
+	 * remount options, so we have to explicitly reset it.
+	 */
+	if (data->flags & NFS_MOUNT_NOAC)
+		*flags |= MS_SYNCHRONOUS;
+
 	/* compare new mount options with old ones */
 	error = nfs_compare_remount_data(nfss, data);
 out:
@@ -2235,8 +2245,7 @@ static struct dentry *nfs_fs_mount(struct file_system_type *fs_type,
 	if (!s->s_root) {
 		/* initial superblock/root creation */
 		nfs_fill_super(s, data);
-		nfs_fscache_get_super_cookie(
-			s, data ? data->fscache_uniq : NULL, NULL);
+		nfs_fscache_get_super_cookie(s, data->fscache_uniq, NULL);
 	}
 
 	mntroot = nfs_get_root(s, mntfh, dev_name);
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index e4cbc11a74a..49c715b4ac9 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -680,7 +680,6 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page,
 	req = nfs_setup_write_request(ctx, page, offset, count);
 	if (IS_ERR(req))
 		return PTR_ERR(req);
-	nfs_mark_request_dirty(req);
 	/* Update file length */
 	nfs_grow_file(page, offset, count);
 	nfs_mark_uptodate(page, req->wb_pgbase, req->wb_bytes);
@@ -940,7 +939,7 @@ static int nfs_flush_multi(struct nfs_pageio_descriptor *desc)
 	atomic_set(&req->wb_complete, requests);
 
 	BUG_ON(desc->pg_lseg);
-	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW);
+	lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW, GFP_NOFS);
 	ClearPageError(page);
 	offset = 0;
 	nbytes = desc->pg_count;
@@ -1014,7 +1013,7 @@ static int nfs_flush_one(struct nfs_pageio_descriptor *desc)
 	}
 	req = nfs_list_entry(data->pages.next);
 	if ((!lseg) && list_is_singular(&data->pages))
-		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW);
+		lseg = pnfs_update_layout(desc->pg_inode, req->wb_context, IOMODE_RW, GFP_NOFS);
 
 	if ((desc->pg_ioflags & FLUSH_COND_STABLE) &&
 	    (desc->pg_moreio || NFS_I(desc->pg_inode)->ncommit))
@@ -1418,8 +1417,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata)
                                 task->tk_pid, task->tk_status);
 
 	/* Call the NFS version-specific code */
-	if (NFS_PROTO(data->inode)->commit_done(task, data) != 0)
-		return;
+	NFS_PROTO(data->inode)->commit_done(task, data);
 }
 
 void nfs_commit_release_pages(struct nfs_write_data *data)
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index aa309aa93fe..4cf04e11c66 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -258,6 +258,7 @@ static void nfs4_put_deleg_lease(struct nfs4_file *fp)
 	if (atomic_dec_and_test(&fp->fi_delegees)) {
 		vfs_setlease(fp->fi_deleg_file, F_UNLCK, &fp->fi_lease);
 		fp->fi_lease = NULL;
+		fput(fp->fi_deleg_file);
 		fp->fi_deleg_file = NULL;
 	}
 }
@@ -402,8 +403,8 @@ static void free_generic_stateid(struct nfs4_stateid *stp)
 	if (stp->st_access_bmap) {
 		oflag = nfs4_access_bmap_to_omode(stp);
 		nfs4_file_put_access(stp->st_file, oflag);
-		put_nfs4_file(stp->st_file);
 	}
+	put_nfs4_file(stp->st_file);
 	kmem_cache_free(stateid_slab, stp);
 }
 
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 2e1cebde90d..129f3c9f62d 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1363,7 +1363,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		goto out;
 	if (!(iap->ia_valid & ATTR_MODE))
 		iap->ia_mode = 0;
-	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
+	err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_EXEC);
 	if (err)
 		goto out;
 
@@ -1385,6 +1385,13 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (IS_ERR(dchild))
 		goto out_nfserr;
 
+	/* If file doesn't exist, check for permissions to create one */
+	if (!dchild->d_inode) {
+		err = fh_verify(rqstp, fhp, S_IFDIR, NFSD_MAY_CREATE);
+		if (err)
+			goto out;
+	}
+
 	err = fh_compose(resfhp, fhp->fh_export, dchild, fhp);
 	if (err)
 		goto out;
diff --git a/fs/nilfs2/alloc.c b/fs/nilfs2/alloc.c
index 0a0a66d98cc..f7684483785 100644
--- a/fs/nilfs2/alloc.c
+++ b/fs/nilfs2/alloc.c
@@ -646,7 +646,7 @@ int nilfs_palloc_freev(struct inode *inode, __u64 *entry_nrs, size_t nitems)
 	unsigned long group, group_offset;
 	int i, j, n, ret;
 
-	for (i = 0; i < nitems; i += n) {
+	for (i = 0; i < nitems; i = j) {
 		group = nilfs_palloc_group(inode, entry_nrs[i], &group_offset);
 		ret = nilfs_palloc_get_desc_block(inode, group, 0, &desc_bh);
 		if (ret < 0)
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index 643720209a9..9a3e6bbff27 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -539,25 +539,41 @@ static int o2hb_verify_crc(struct o2hb_region *reg,
 
 /* We want to make sure that nobody is heartbeating on top of us --
  * this will help detect an invalid configuration. */
-static int o2hb_check_last_timestamp(struct o2hb_region *reg)
+static void o2hb_check_last_timestamp(struct o2hb_region *reg)
 {
-	int node_num, ret;
 	struct o2hb_disk_slot *slot;
 	struct o2hb_disk_heartbeat_block *hb_block;
+	char *errstr;
 
-	node_num = o2nm_this_node();
-
-	ret = 1;
-	slot = &reg->hr_slots[node_num];
+	slot = &reg->hr_slots[o2nm_this_node()];
 	/* Don't check on our 1st timestamp */
-	if (slot->ds_last_time) {
-		hb_block = slot->ds_raw_block;
+	if (!slot->ds_last_time)
+		return;
 
-		if (le64_to_cpu(hb_block->hb_seq) != slot->ds_last_time)
-			ret = 0;
-	}
+	hb_block = slot->ds_raw_block;
+	if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time &&
+	    le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation &&
+	    hb_block->hb_node == slot->ds_node_num)
+		return;
 
-	return ret;
+#define ERRSTR1		"Another node is heartbeating on device"
+#define ERRSTR2		"Heartbeat generation mismatch on device"
+#define ERRSTR3		"Heartbeat sequence mismatch on device"
+
+	if (hb_block->hb_node != slot->ds_node_num)
+		errstr = ERRSTR1;
+	else if (le64_to_cpu(hb_block->hb_generation) !=
+		 slot->ds_last_generation)
+		errstr = ERRSTR2;
+	else
+		errstr = ERRSTR3;
+
+	mlog(ML_ERROR, "%s (%s): expected(%u:0x%llx, 0x%llx), "
+	     "ondisk(%u:0x%llx, 0x%llx)\n", errstr, reg->hr_dev_name,
+	     slot->ds_node_num, (unsigned long long)slot->ds_last_generation,
+	     (unsigned long long)slot->ds_last_time, hb_block->hb_node,
+	     (unsigned long long)le64_to_cpu(hb_block->hb_generation),
+	     (unsigned long long)le64_to_cpu(hb_block->hb_seq));
 }
 
 static inline void o2hb_prepare_block(struct o2hb_region *reg,
@@ -983,9 +999,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 	/* With an up to date view of the slots, we can check that no
 	 * other node has been improperly configured to heartbeat in
 	 * our slot. */
-	if (!o2hb_check_last_timestamp(reg))
-		mlog(ML_ERROR, "Device \"%s\": another node is heartbeating "
-		     "in our slot!\n", reg->hr_dev_name);
+	o2hb_check_last_timestamp(reg);
 
 	/* fill in the proper info for our next heartbeat */
 	o2hb_prepare_block(reg, reg->hr_generation);
@@ -999,8 +1013,8 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg)
 	}
 
 	i = -1;
-	while((i = find_next_bit(configured_nodes, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) {
-
+	while((i = find_next_bit(configured_nodes,
+				 O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) {
 		change |= o2hb_check_slot(reg, &reg->hr_slots[i]);
 	}
 
@@ -1690,6 +1704,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg,
 	struct file *filp = NULL;
 	struct inode *inode = NULL;
 	ssize_t ret = -EINVAL;
+	int live_threshold;
 
 	if (reg->hr_bdev)
 		goto out;
@@ -1766,8 +1781,18 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg,
 	 * A node is considered live after it has beat LIVE_THRESHOLD
 	 * times.  We're not steady until we've given them a chance
 	 * _after_ our first read.
+	 * The default threshold is bare minimum so as to limit the delay
+	 * during mounts. For global heartbeat, the threshold doubled for the
+	 * first region.
 	 */
-	atomic_set(&reg->hr_steady_iterations, O2HB_LIVE_THRESHOLD + 1);
+	live_threshold = O2HB_LIVE_THRESHOLD;
+	if (o2hb_global_heartbeat_active()) {
+		spin_lock(&o2hb_live_lock);
+		if (o2hb_pop_count(&o2hb_region_bitmap, O2NM_MAX_REGIONS) == 1)
+			live_threshold <<= 1;
+		spin_unlock(&o2hb_live_lock);
+	}
+	atomic_set(&reg->hr_steady_iterations, live_threshold + 1);
 
 	hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s",
 			      reg->hr_item.ci_name);
diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c
index 9fe5b8fd658..8582e3f4f12 100644
--- a/fs/ocfs2/dir.c
+++ b/fs/ocfs2/dir.c
@@ -2868,7 +2868,7 @@ static int ocfs2_expand_inline_dir(struct inode *dir, struct buffer_head *di_bh,
 		bytes = blocks_wanted << sb->s_blocksize_bits;
 	struct ocfs2_super *osb = OCFS2_SB(dir->i_sb);
 	struct ocfs2_inode_info *oi = OCFS2_I(dir);
-	struct ocfs2_alloc_context *data_ac;
+	struct ocfs2_alloc_context *data_ac = NULL;
 	struct ocfs2_alloc_context *meta_ac = NULL;
 	struct buffer_head *dirdata_bh = NULL;
 	struct buffer_head *dx_root_bh = NULL;
diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index 7540a492eab..3b179d6cbde 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -1614,7 +1614,8 @@ static int dlm_try_to_join_domain(struct dlm_ctxt *dlm)
 	spin_unlock(&dlm->spinlock);
 
 	/* Support for global heartbeat and node info was added in 1.1 */
-	if (dlm_protocol.pv_major > 1 || dlm_protocol.pv_minor > 0) {
+	if (dlm->dlm_locking_proto.pv_major > 1 ||
+	    dlm->dlm_locking_proto.pv_minor > 0) {
 		status = dlm_send_nodeinfo(dlm, ctxt->yes_resp_map);
 		if (status) {
 			mlog_errno(status);
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index fede57ed005..84d166328cf 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -2574,6 +2574,9 @@ fail:
 		res->state &= ~DLM_LOCK_RES_MIGRATING;
 		wake = 1;
 		spin_unlock(&res->spinlock);
+		if (dlm_is_host_down(ret))
+			dlm_wait_for_node_death(dlm, target,
+						DLM_NODE_DEATH_WAIT_MAX);
 		goto leave;
 	}
 
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 41565ae5285..89659d6dc20 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1607,6 +1607,9 @@ static void ocfs2_calc_trunc_pos(struct inode *inode,
 	range = le32_to_cpu(rec->e_cpos) + ocfs2_rec_clusters(el, rec);
 
 	if (le32_to_cpu(rec->e_cpos) >= trunc_start) {
+		/*
+		 * remove an entire extent record.
+		 */
 		*trunc_cpos = le32_to_cpu(rec->e_cpos);
 		/*
 		 * Skip holes if any.
@@ -1617,7 +1620,16 @@ static void ocfs2_calc_trunc_pos(struct inode *inode,
 		*blkno = le64_to_cpu(rec->e_blkno);
 		*trunc_end = le32_to_cpu(rec->e_cpos);
 	} else if (range > trunc_start) {
+		/*
+		 * remove a partial extent record, which means we're
+		 * removing the last extent record.
+		 */
 		*trunc_cpos = trunc_start;
+		/*
+		 * skip hole if any.
+		 */
+		if (range < *trunc_end)
+			*trunc_end = range;
 		*trunc_len = *trunc_end - trunc_start;
 		coff = trunc_start - le32_to_cpu(rec->e_cpos);
 		*blkno = le64_to_cpu(rec->e_blkno) +
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index b141a44605c..295d56454e8 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1260,6 +1260,9 @@ void ocfs2_complete_mount_recovery(struct ocfs2_super *osb)
 {
 	struct ocfs2_journal *journal = osb->journal;
 
+	if (ocfs2_is_hard_readonly(osb))
+		return;
+
 	/* No need to queue up our truncate_log as regular cleanup will catch
 	 * that */
 	ocfs2_queue_recovery_completion(journal, osb->slot_num,
diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h
index b68f87a8392..938387a10d5 100644
--- a/fs/ocfs2/ocfs2_fs.h
+++ b/fs/ocfs2/ocfs2_fs.h
@@ -1019,7 +1019,7 @@ struct ocfs2_xattr_entry {
 	__le16	xe_name_offset;  /* byte offset from the 1st entry in the
 				    local xattr storage(inode, xattr block or
 				    xattr bucket). */
-	__u8	xe_name_len;	 /* xattr name len, does't include prefix. */
+	__u8	xe_name_len;	 /* xattr name len, doesn't include prefix. */
 	__u8	xe_type;         /* the low 7 bits indicate the name prefix
 				  * type and the highest bit indicates whether
 				  * the EA is stored in the local storage. */
diff --git a/fs/partitions/efi.c b/fs/partitions/efi.c
index ac0ccb5026a..19d6750d1d6 100644
--- a/fs/partitions/efi.c
+++ b/fs/partitions/efi.c
@@ -348,6 +348,12 @@ static int is_gpt_valid(struct parsed_partitions *state, u64 lba,
 		goto fail;
 	}
 
+	/* Check that sizeof_partition_entry has the correct value */
+	if (le32_to_cpu((*gpt)->sizeof_partition_entry) != sizeof(gpt_entry)) {
+		pr_debug("GUID Partitition Entry Size check failed.\n");
+		goto fail;
+	}
+
 	if (!(*ptes = alloc_read_gpt_entries(state, *gpt)))
 		goto fail;
 
diff --git a/fs/partitions/ldm.c b/fs/partitions/ldm.c
index ce4f6244042..a29d5ccf3d5 100644
--- a/fs/partitions/ldm.c
+++ b/fs/partitions/ldm.c
@@ -1335,6 +1335,11 @@ static bool ldm_frag_add (const u8 *data, int size, struct list_head *frags)
 
 	list_add_tail (&f->list, frags);
 found:
+	if (rec >= f->num) {
+		ldm_error("REC value (%d) exceeds NUM value (%d)", rec, f->num);
+		return false;
+	}
+
 	if (f->map & (1 << rec)) {
 		ldm_error ("Duplicate VBLK, part %d.", rec);
 		f->map &= 0x7F;			/* Mark the group as broken */
diff --git a/fs/proc/base.c b/fs/proc/base.c
index dd6628d3ba4..dfa532730e5 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -3124,11 +3124,16 @@ static int proc_pid_fill_cache(struct file *filp, void *dirent, filldir_t filldi
 /* for the /proc/ directory itself, after non-process stuff has been done */
 int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir)
 {
-	unsigned int nr = filp->f_pos - FIRST_PROCESS_ENTRY;
-	struct task_struct *reaper = get_proc_task(filp->f_path.dentry->d_inode);
+	unsigned int nr;
+	struct task_struct *reaper;
 	struct tgid_iter iter;
 	struct pid_namespace *ns;
 
+	if (filp->f_pos >= PID_MAX_LIMIT + TGID_OFFSET)
+		goto out_no_task;
+	nr = filp->f_pos - FIRST_PROCESS_ENTRY;
+
+	reaper = get_proc_task(filp->f_path.dentry->d_inode);
 	if (!reaper)
 		goto out_no_task;
 
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 2e7addfd980..318d8654989 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -214,7 +214,7 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma)
 	int flags = vma->vm_flags;
 	unsigned long ino = 0;
 	unsigned long long pgoff = 0;
-	unsigned long start;
+	unsigned long start, end;
 	dev_t dev = 0;
 	int len;
 
@@ -227,13 +227,15 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma)
 
 	/* We don't show the stack guard page in /proc/maps */
 	start = vma->vm_start;
-	if (vma->vm_flags & VM_GROWSDOWN)
-		if (!vma_stack_continue(vma->vm_prev, vma->vm_start))
-			start += PAGE_SIZE;
+	if (stack_guard_page_start(vma, start))
+		start += PAGE_SIZE;
+	end = vma->vm_end;
+	if (stack_guard_page_end(vma, end))
+		end -= PAGE_SIZE;
 
 	seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n",
 			start,
-			vma->vm_end,
+			end,
 			flags & VM_READ ? 'r' : '-',
 			flags & VM_WRITE ? 'w' : '-',
 			flags & VM_EXEC ? 'x' : '-',
diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c
index f835a25625f..f2c3ff20ea6 100644
--- a/fs/pstore/platform.c
+++ b/fs/pstore/platform.c
@@ -152,21 +152,27 @@ EXPORT_SYMBOL_GPL(pstore_register);
 void pstore_get_records(void)
 {
 	struct pstore_info *psi = psinfo;
-	size_t			size;
+	ssize_t			size;
 	u64			id;
 	enum pstore_type_id	type;
 	struct timespec		time;
-	int			failed = 0;
+	int			failed = 0, rc;
 
 	if (!psi)
 		return;
 
 	mutex_lock(&psinfo->buf_mutex);
+	rc = psi->open(psi);
+	if (rc)
+		goto out;
+
 	while ((size = psi->read(&id, &type, &time)) > 0) {
-		if (pstore_mkfile(type, psi->name, id, psi->buf, size,
+		if (pstore_mkfile(type, psi->name, id, psi->buf, (size_t)size,
 				  time, psi->erase))
 			failed++;
 	}
+	psi->close(psi);
+out:
 	mutex_unlock(&psinfo->buf_mutex);
 
 	if (failed)
diff --git a/fs/super.c b/fs/super.c
index 8a06881b192..c04f7e0b7ed 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -948,8 +948,7 @@ mount_fs(struct file_system_type *type, int flags, const char *name, void *data)
 	 * filesystems should never set s_maxbytes larger than MAX_LFS_FILESIZE
 	 * but s_maxbytes was an unsigned long long for many releases. Throw
 	 * this warning for a little while to try and catch filesystems that
-	 * violate this rule. This warning should be either removed or
-	 * converted to a BUG() in 2.6.34.
+	 * violate this rule.
 	 */
 	WARN((sb->s_maxbytes < 0), "%s set sb->s_maxbytes to "
 		"negative value (%lld)\n", type->name, sb->s_maxbytes);
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index da3fefe91a8..1ad8c93c1b8 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -24,13 +24,6 @@
 
 #include "sysfs.h"
 
-/* used in crash dumps to help with debugging */
-static char last_sysfs_file[PATH_MAX];
-void sysfs_printk_last_file(void)
-{
-	printk(KERN_EMERG "last sysfs file: %s\n", last_sysfs_file);
-}
-
 /*
  * There's one sysfs_buffer for each open file and one
  * sysfs_open_dirent for each sysfs_dirent with one or more open
@@ -337,11 +330,6 @@ static int sysfs_open_file(struct inode *inode, struct file *file)
 	struct sysfs_buffer *buffer;
 	const struct sysfs_ops *ops;
 	int error = -EACCES;
-	char *p;
-
-	p = d_path(&file->f_path, last_sysfs_file, sizeof(last_sysfs_file));
-	if (!IS_ERR(p))
-		memmove(last_sysfs_file, p, strlen(p) + 1);
 
 	/* need attr_sd for attr and ops, its parent for kobj */
 	if (!sysfs_get_active(attr_sd))
diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c
index c8769dc222d..194414f8298 100644
--- a/fs/sysfs/group.c
+++ b/fs/sysfs/group.c
@@ -101,9 +101,9 @@ int sysfs_create_group(struct kobject *kobj,
 }
 
 /**
- * sysfs_update_group - given a directory kobject, create an attribute group
- * @kobj:	The kobject to create the group on
- * @grp:	The attribute group to create
+ * sysfs_update_group - given a directory kobject, update an attribute group
+ * @kobj:	The kobject to update the group on
+ * @grp:	The attribute group to update
  *
  * This function updates an attribute group.  Unlike
  * sysfs_create_group(), it will explicitly not warn or error if any
diff --git a/fs/ubifs/log.c b/fs/ubifs/log.c
index 4d0cb124146..40fa780ebea 100644
--- a/fs/ubifs/log.c
+++ b/fs/ubifs/log.c
@@ -175,26 +175,6 @@ void ubifs_add_bud(struct ubifs_info *c, struct ubifs_bud *bud)
 }
 
 /**
- * ubifs_create_buds_lists - create journal head buds lists for remount rw.
- * @c: UBIFS file-system description object
- */
-void ubifs_create_buds_lists(struct ubifs_info *c)
-{
-	struct rb_node *p;
-
-	spin_lock(&c->buds_lock);
-	p = rb_first(&c->buds);
-	while (p) {
-		struct ubifs_bud *bud = rb_entry(p, struct ubifs_bud, rb);
-		struct ubifs_jhead *jhead = &c->jheads[bud->jhead];
-
-		list_add_tail(&bud->list, &jhead->buds_list);
-		p = rb_next(p);
-	}
-	spin_unlock(&c->buds_lock);
-}
-
-/**
  * ubifs_add_bud_to_log - add a new bud to the log.
  * @c: UBIFS file-system description object
  * @jhead: journal head the bud belongs to
diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c
index 936f2cbfe6b..3dbad6fbd1e 100644
--- a/fs/ubifs/recovery.c
+++ b/fs/ubifs/recovery.c
@@ -317,6 +317,32 @@ int ubifs_recover_master_node(struct ubifs_info *c)
 			goto out_free;
 		}
 		memcpy(c->rcvrd_mst_node, c->mst_node, UBIFS_MST_NODE_SZ);
+
+		/*
+		 * We had to recover the master node, which means there was an
+		 * unclean reboot. However, it is possible that the master node
+		 * is clean at this point, i.e., %UBIFS_MST_DIRTY is not set.
+		 * E.g., consider the following chain of events:
+		 *
+		 * 1. UBIFS was cleanly unmounted, so the master node is clean
+		 * 2. UBIFS is being mounted R/W and starts changing the master
+		 *    node in the first (%UBIFS_MST_LNUM). A power cut happens,
+		 *    so this LEB ends up with some amount of garbage at the
+		 *    end.
+		 * 3. UBIFS is being mounted R/O. We reach this place and
+		 *    recover the master node from the second LEB
+		 *    (%UBIFS_MST_LNUM + 1). But we cannot update the media
+		 *    because we are being mounted R/O. We have to defer the
+		 *    operation.
+		 * 4. However, this master node (@c->mst_node) is marked as
+		 *    clean (since the step 1). And if we just return, the
+		 *    mount code will be confused and won't recover the master
+		 *    node when it is re-mounter R/W later.
+		 *
+		 *    Thus, to force the recovery by marking the master node as
+		 *    dirty.
+		 */
+		c->mst_node->flags |= cpu_to_le32(UBIFS_MST_DIRTY);
 	} else {
 		/* Write the recovered master node */
 		c->max_sqnum = le64_to_cpu(mst->ch.sqnum) - 1;
diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c
index eed0fcff8d7..d3d6d365bfc 100644
--- a/fs/ubifs/replay.c
+++ b/fs/ubifs/replay.c
@@ -59,6 +59,7 @@ enum {
  * @new_size: truncation new size
  * @free: amount of free space in a bud
  * @dirty: amount of dirty space in a bud from padding and deletion nodes
+ * @jhead: journal head number of the bud
  *
  * UBIFS journal replay must compare node sequence numbers, which means it must
  * build a tree of node information to insert into the TNC.
@@ -80,6 +81,7 @@ struct replay_entry {
 		struct {
 			int free;
 			int dirty;
+			int jhead;
 		};
 	};
 };
@@ -159,6 +161,11 @@ static int set_bud_lprops(struct ubifs_info *c, struct replay_entry *r)
 		err = PTR_ERR(lp);
 		goto out;
 	}
+
+	/* Make sure the journal head points to the latest bud */
+	err = ubifs_wbuf_seek_nolock(&c->jheads[r->jhead].wbuf, r->lnum,
+				     c->leb_size - r->free, UBI_SHORTTERM);
+
 out:
 	ubifs_release_lprops(c);
 	return err;
@@ -627,10 +634,6 @@ static int replay_bud(struct ubifs_info *c, int lnum, int offs, int jhead,
 	ubifs_assert(sleb->endpt - offs >= used);
 	ubifs_assert(sleb->endpt % c->min_io_size == 0);
 
-	if (sleb->endpt + c->min_io_size <= c->leb_size && !c->ro_mount)
-		err = ubifs_wbuf_seek_nolock(&c->jheads[jhead].wbuf, lnum,
-					     sleb->endpt, UBI_SHORTTERM);
-
 	*dirty = sleb->endpt - offs - used;
 	*free = c->leb_size - sleb->endpt;
 
@@ -653,12 +656,14 @@ out_dump:
  * @sqnum: sequence number
  * @free: amount of free space in bud
  * @dirty: amount of dirty space from padding and deletion nodes
+ * @jhead: journal head number for the bud
  *
  * This function inserts a reference node to the replay tree and returns zero
  * in case of success or a negative error code in case of failure.
  */
 static int insert_ref_node(struct ubifs_info *c, int lnum, int offs,
-			   unsigned long long sqnum, int free, int dirty)
+			   unsigned long long sqnum, int free, int dirty,
+			   int jhead)
 {
 	struct rb_node **p = &c->replay_tree.rb_node, *parent = NULL;
 	struct replay_entry *r;
@@ -688,6 +693,7 @@ static int insert_ref_node(struct ubifs_info *c, int lnum, int offs,
 	r->flags = REPLAY_REF;
 	r->free = free;
 	r->dirty = dirty;
+	r->jhead = jhead;
 
 	rb_link_node(&r->rb, parent, p);
 	rb_insert_color(&r->rb, &c->replay_tree);
@@ -712,7 +718,7 @@ static int replay_buds(struct ubifs_info *c)
 		if (err)
 			return err;
 		err = insert_ref_node(c, b->bud->lnum, b->bud->start, b->sqnum,
-				      free, dirty);
+				      free, dirty, b->bud->jhead);
 		if (err)
 			return err;
 	}
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index c75f6133206..04ad07f4fcc 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -1257,12 +1257,12 @@ static int mount_ubifs(struct ubifs_info *c)
 		goto out_free;
 	}
 
+	err = alloc_wbufs(c);
+	if (err)
+		goto out_cbuf;
+
 	sprintf(c->bgt_name, BGT_NAME_PATTERN, c->vi.ubi_num, c->vi.vol_id);
 	if (!c->ro_mount) {
-		err = alloc_wbufs(c);
-		if (err)
-			goto out_cbuf;
-
 		/* Create background thread */
 		c->bgt = kthread_create(ubifs_bg_thread, c, "%s", c->bgt_name);
 		if (IS_ERR(c->bgt)) {
@@ -1631,12 +1631,6 @@ static int ubifs_remount_rw(struct ubifs_info *c)
 	if (err)
 		goto out;
 
-	err = alloc_wbufs(c);
-	if (err)
-		goto out;
-
-	ubifs_create_buds_lists(c);
-
 	/* Create background thread */
 	c->bgt = kthread_create(ubifs_bg_thread, c, "%s", c->bgt_name);
 	if (IS_ERR(c->bgt)) {
@@ -1671,14 +1665,25 @@ static int ubifs_remount_rw(struct ubifs_info *c)
 	if (err)
 		goto out;
 
+	dbg_gen("re-mounted read-write");
+	c->remounting_rw = 0;
+
 	if (c->need_recovery) {
 		c->need_recovery = 0;
 		ubifs_msg("deferred recovery completed");
+	} else {
+		/*
+		 * Do not run the debugging space check if the were doing
+		 * recovery, because when we saved the information we had the
+		 * file-system in a state where the TNC and lprops has been
+		 * modified in memory, but all the I/O operations (including a
+		 * commit) were deferred. So the file-system was in
+		 * "non-committed" state. Now the file-system is in committed
+		 * state, and of course the amount of free space will change
+		 * because, for example, the old index size was imprecise.
+		 */
+		err = dbg_check_space_info(c);
 	}
-
-	dbg_gen("re-mounted read-write");
-	c->remounting_rw = 0;
-	err = dbg_check_space_info(c);
 	mutex_unlock(&c->umount_mutex);
 	return err;
 
@@ -1733,7 +1738,6 @@ static void ubifs_remount_ro(struct ubifs_info *c)
 	if (err)
 		ubifs_ro_mode(c, err);
 
-	free_wbufs(c);
 	vfree(c->orph_buf);
 	c->orph_buf = NULL;
 	kfree(c->write_reserve_buf);
@@ -1761,10 +1765,12 @@ static void ubifs_put_super(struct super_block *sb)
 	 * of the media. For example, there will be dirty inodes if we failed
 	 * to write them back because of I/O errors.
 	 */
-	ubifs_assert(atomic_long_read(&c->dirty_pg_cnt) == 0);
-	ubifs_assert(c->budg_idx_growth == 0);
-	ubifs_assert(c->budg_dd_growth == 0);
-	ubifs_assert(c->budg_data_growth == 0);
+	if (!c->ro_error) {
+		ubifs_assert(atomic_long_read(&c->dirty_pg_cnt) == 0);
+		ubifs_assert(c->budg_idx_growth == 0);
+		ubifs_assert(c->budg_dd_growth == 0);
+		ubifs_assert(c->budg_data_growth == 0);
+	}
 
 	/*
 	 * The 'c->umount_lock' prevents races between UBIFS memory shrinker
diff --git a/fs/xattr.c b/fs/xattr.c
index a19acdb81cd..f1ef94974de 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -666,7 +666,7 @@ generic_setxattr(struct dentry *dentry, const char *name, const void *value, siz
 	handler = xattr_resolve_name(dentry->d_sb->s_xattr, &name);
 	if (!handler)
 		return -EOPNOTSUPP;
-	return handler->set(dentry, name, value, size, 0, handler->flags);
+	return handler->set(dentry, name, value, size, flags, handler->flags);
 }
 
 /*
diff --git a/fs/xfs/linux-2.6/xfs_message.c b/fs/xfs/linux-2.6/xfs_message.c
index 3ca79560911..9f76cceb678 100644
--- a/fs/xfs/linux-2.6/xfs_message.c
+++ b/fs/xfs/linux-2.6/xfs_message.c
@@ -34,8 +34,10 @@ __xfs_printk(
 	const struct xfs_mount	*mp,
 	struct va_format	*vaf)
 {
-	if (mp && mp->m_fsname)
+	if (mp && mp->m_fsname) {
 		printk("%sXFS (%s): %pV\n", level, mp->m_fsname, vaf);
+		return;
+	}
 	printk("%sXFS: %pV\n", level, vaf);
 }
 
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index e4f9c1b0836..3e898a48122 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -926,6 +926,7 @@ restart:
 					XFS_LOOKUP_BATCH,
 					XFS_ICI_RECLAIM_TAG);
 			if (!nr_found) {
+				done = 1;
 				rcu_read_unlock();
 				break;
 			}
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index acdb92f14d5..5fc2380092c 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -346,20 +346,23 @@ xfs_ail_delete(
  */
 STATIC void
 xfs_ail_worker(
-	struct work_struct *work)
+	struct work_struct	*work)
 {
-	struct xfs_ail	*ailp = container_of(to_delayed_work(work),
+	struct xfs_ail		*ailp = container_of(to_delayed_work(work),
 					struct xfs_ail, xa_work);
-	long		tout;
-	xfs_lsn_t	target =  ailp->xa_target;
-	xfs_lsn_t	lsn;
-	xfs_log_item_t	*lip;
-	int		flush_log, count, stuck;
-	xfs_mount_t	*mp = ailp->xa_mount;
+	xfs_mount_t		*mp = ailp->xa_mount;
 	struct xfs_ail_cursor	*cur = &ailp->xa_cursors;
-	int		push_xfsbufd = 0;
+	xfs_log_item_t		*lip;
+	xfs_lsn_t		lsn;
+	xfs_lsn_t		target;
+	long			tout = 10;
+	int			flush_log = 0;
+	int			stuck = 0;
+	int			count = 0;
+	int			push_xfsbufd = 0;
 
 	spin_lock(&ailp->xa_lock);
+	target = ailp->xa_target;
 	xfs_trans_ail_cursor_init(ailp, cur);
 	lip = xfs_trans_ail_cursor_first(ailp, cur, ailp->xa_last_pushed_lsn);
 	if (!lip || XFS_FORCED_SHUTDOWN(mp)) {
@@ -368,8 +371,7 @@ xfs_ail_worker(
 		 */
 		xfs_trans_ail_cursor_done(ailp, cur);
 		spin_unlock(&ailp->xa_lock);
-		ailp->xa_last_pushed_lsn = 0;
-		return;
+		goto out_done;
 	}
 
 	XFS_STATS_INC(xs_push_ail);
@@ -386,8 +388,7 @@ xfs_ail_worker(
 	 * lots of contention on the AIL lists.
 	 */
 	lsn = lip->li_lsn;
-	flush_log = stuck = count = 0;
-	while ((XFS_LSN_CMP(lip->li_lsn, target) < 0)) {
+	while ((XFS_LSN_CMP(lip->li_lsn, target) <= 0)) {
 		int	lock_result;
 		/*
 		 * If we can lock the item without sleeping, unlock the AIL
@@ -480,21 +481,25 @@ xfs_ail_worker(
 	}
 
 	/* assume we have more work to do in a short while */
-	tout = 10;
+out_done:
 	if (!count) {
 		/* We're past our target or empty, so idle */
 		ailp->xa_last_pushed_lsn = 0;
 
 		/*
-		 * Check for an updated push target before clearing the
-		 * XFS_AIL_PUSHING_BIT. If the target changed, we've got more
-		 * work to do. Wait a bit longer before starting that work.
+		 * We clear the XFS_AIL_PUSHING_BIT first before checking
+		 * whether the target has changed. If the target has changed,
+		 * this pushes the requeue race directly onto the result of the
+		 * atomic test/set bit, so we are guaranteed that either the
+		 * the pusher that changed the target or ourselves will requeue
+		 * the work (but not both).
 		 */
+		clear_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags);
 		smp_rmb();
-		if (ailp->xa_target == target) {
-			clear_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags);
+		if (XFS_LSN_CMP(ailp->xa_target, target) == 0 ||
+		    test_and_set_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags))
 			return;
-		}
+
 		tout = 50;
 	} else if (XFS_LSN_CMP(lsn, target) >= 0) {
 		/*
@@ -553,7 +558,7 @@ xfs_ail_push(
 	 * the XFS_AIL_PUSHING_BIT.
 	 */
 	smp_wmb();
-	ailp->xa_target = threshold_lsn;
+	xfs_trans_ail_copy_lsn(ailp, &ailp->xa_target, &threshold_lsn);
 	if (!test_and_set_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags))
 		queue_delayed_work(xfs_syncd_wq, &ailp->xa_work, 0);
 }