Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs

Pull ext3, reiserfs, udf & isofs fixes from Jan Kara:
 "The contains a bunch of ext3 cleanups and minor improvements, major
  reiserfs locking changes which should hopefully fix deadlocks
  introduced by BKL removal, and udf/isofs changes to refuse mounting fs
  rw instead of mounting it ro automatically which makes eject button
  work as expected for all media (see the changelog for why userspace
  should be ok with this change)"

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
  jbd: use a single printk for jbd_debug()
  reiserfs: locking, release lock around quota operations
  reiserfs: locking, handle nested locks properly
  reiserfs: locking, push write lock out of xattr code
  jbd: relocate assert after state lock in journal_commit_transaction()
  udf: Refuse RW mount of the filesystem instead of making it RO
  udf: Standardize return values in mount sequence
  isofs: Refuse RW mount of the filesystem instead of making it RO
  ext3: allow specifying external journal by pathname mount option
  jbd: remove unneeded semicolon

15 files changed, 324 insertions, 285 deletions

diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c
index a98b7740a0f..dc9a6829f7c 100644
--- a/fs/reiserfs/bitmap.c
+++ b/fs/reiserfs/bitmap.c
@@ -423,8 +423,11 @@ static void _reiserfs_free_block(struct reiserfs_transaction_handle *th,
 	set_sb_free_blocks(rs, sb_free_blocks(rs) + 1);
 
 	journal_mark_dirty(th, s, sbh);
-	if (for_unformatted)
+	if (for_unformatted) {
+		int depth = reiserfs_write_unlock_nested(s);
 		dquot_free_block_nodirty(inode, 1);
+		reiserfs_write_lock_nested(s, depth);
+	}
 }
 
 void reiserfs_free_block(struct reiserfs_transaction_handle *th,
@@ -1128,6 +1131,7 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
 	b_blocknr_t finish = SB_BLOCK_COUNT(s) - 1;
 	int passno = 0;
 	int nr_allocated = 0;
+	int depth;
 
 	determine_prealloc_size(hint);
 	if (!hint->formatted_node) {
@@ -1137,10 +1141,13 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
 			       "reiserquota: allocating %d blocks id=%u",
 			       amount_needed, hint->inode->i_uid);
 #endif
+		depth = reiserfs_write_unlock_nested(s);
 		quota_ret =
 		    dquot_alloc_block_nodirty(hint->inode, amount_needed);
-		if (quota_ret)	/* Quota exceeded? */
+		if (quota_ret) {	/* Quota exceeded? */
+			reiserfs_write_lock_nested(s, depth);
 			return QUOTA_EXCEEDED;
+		}
 		if (hint->preallocate && hint->prealloc_size) {
 #ifdef REISERQUOTA_DEBUG
 			reiserfs_debug(s, REISERFS_DEBUG_CODE,
@@ -1153,6 +1160,7 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
 				hint->preallocate = hint->prealloc_size = 0;
 		}
 		/* for unformatted nodes, force large allocations */
+		reiserfs_write_lock_nested(s, depth);
 	}
 
 	do {
@@ -1181,9 +1189,11 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
 					       hint->inode->i_uid);
 #endif
 				/* Free not allocated blocks */
+				depth = reiserfs_write_unlock_nested(s);
 				dquot_free_block_nodirty(hint->inode,
 					amount_needed + hint->prealloc_size -
 					nr_allocated);
+				reiserfs_write_lock_nested(s, depth);
 			}
 			while (nr_allocated--)
 				reiserfs_free_block(hint->th, hint->inode,
@@ -1214,10 +1224,13 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
 			       REISERFS_I(hint->inode)->i_prealloc_count,
 			       hint->inode->i_uid);
 #endif
+
+		depth = reiserfs_write_unlock_nested(s);
 		dquot_free_block_nodirty(hint->inode, amount_needed +
 					 hint->prealloc_size - nr_allocated -
 					 REISERFS_I(hint->inode)->
 					 i_prealloc_count);
+		reiserfs_write_lock_nested(s, depth);
 	}
 
 	return CARRY_ON;
@@ -1340,10 +1353,11 @@ struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb,
 		                 "reading failed", __func__, block);
 	else {
 		if (buffer_locked(bh)) {
+			int depth;
 			PROC_INFO_INC(sb, scan_bitmap.wait);
-			reiserfs_write_unlock(sb);
+			depth = reiserfs_write_unlock_nested(sb);
 			__wait_on_buffer(bh);
-			reiserfs_write_lock(sb);
+			reiserfs_write_lock_nested(sb, depth);
 		}
 		BUG_ON(!buffer_uptodate(bh));
 		BUG_ON(atomic_read(&bh->b_count) == 0);
diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c
index 03e4ca5624d..1fd2051109a 100644
--- a/fs/reiserfs/dir.c
+++ b/fs/reiserfs/dir.c
@@ -71,6 +71,7 @@ int reiserfs_readdir_inode(struct inode *inode, struct dir_context *ctx)
 	char small_buf[32];	/* avoid kmalloc if we can */
 	struct reiserfs_dir_entry de;
 	int ret = 0;
+	int depth;
 
 	reiserfs_write_lock(inode->i_sb);
 
@@ -181,17 +182,17 @@ int reiserfs_readdir_inode(struct inode *inode, struct dir_context *ctx)
 				 * Since filldir might sleep, we can release
 				 * the write lock here for other waiters
 				 */
-				reiserfs_write_unlock(inode->i_sb);
+				depth = reiserfs_write_unlock_nested(inode->i_sb);
 				if (!dir_emit
 				    (ctx, local_buf, d_reclen, d_ino,
 				     DT_UNKNOWN)) {
-					reiserfs_write_lock(inode->i_sb);
+					reiserfs_write_lock_nested(inode->i_sb, depth);
 					if (local_buf != small_buf) {
 						kfree(local_buf);
 					}
 					goto end;
 				}
-				reiserfs_write_lock(inode->i_sb);
+				reiserfs_write_lock_nested(inode->i_sb, depth);
 				if (local_buf != small_buf) {
 					kfree(local_buf);
 				}
diff --git a/fs/reiserfs/fix_node.c b/fs/reiserfs/fix_node.c
index 430e0658704..dc4d4153031 100644
--- a/fs/reiserfs/fix_node.c
+++ b/fs/reiserfs/fix_node.c
@@ -1022,9 +1022,9 @@ static int get_far_parent(struct tree_balance *tb,
 	if (buffer_locked(*pcom_father)) {
 
 		/* Release the write lock while the buffer is busy */
-		reiserfs_write_unlock(tb->tb_sb);
+		int depth = reiserfs_write_unlock_nested(tb->tb_sb);
 		__wait_on_buffer(*pcom_father);
-		reiserfs_write_lock(tb->tb_sb);
+		reiserfs_write_lock_nested(tb->tb_sb, depth);
 		if (FILESYSTEM_CHANGED_TB(tb)) {
 			brelse(*pcom_father);
 			return REPEAT_SEARCH;
@@ -1929,9 +1929,9 @@ static int get_direct_parent(struct tree_balance *tb, int h)
 		return REPEAT_SEARCH;
 
 	if (buffer_locked(bh)) {
-		reiserfs_write_unlock(tb->tb_sb);
+		int depth = reiserfs_write_unlock_nested(tb->tb_sb);
 		__wait_on_buffer(bh);
-		reiserfs_write_lock(tb->tb_sb);
+		reiserfs_write_lock_nested(tb->tb_sb, depth);
 		if (FILESYSTEM_CHANGED_TB(tb))
 			return REPEAT_SEARCH;
 	}
@@ -1952,6 +1952,7 @@ static int get_neighbors(struct tree_balance *tb, int h)
 	unsigned long son_number;
 	struct super_block *sb = tb->tb_sb;
 	struct buffer_head *bh;
+	int depth;
 
 	PROC_INFO_INC(sb, get_neighbors[h]);
 
@@ -1969,9 +1970,9 @@ static int get_neighbors(struct tree_balance *tb, int h)
 		     tb->FL[h]) ? tb->lkey[h] : B_NR_ITEMS(tb->
 								       FL[h]);
 		son_number = B_N_CHILD_NUM(tb->FL[h], child_position);
-		reiserfs_write_unlock(sb);
+		depth = reiserfs_write_unlock_nested(tb->tb_sb);
 		bh = sb_bread(sb, son_number);
-		reiserfs_write_lock(sb);
+		reiserfs_write_lock_nested(tb->tb_sb, depth);
 		if (!bh)
 			return IO_ERROR;
 		if (FILESYSTEM_CHANGED_TB(tb)) {
@@ -2009,9 +2010,9 @@ static int get_neighbors(struct tree_balance *tb, int h)
 		child_position =
 		    (bh == tb->FR[h]) ? tb->rkey[h] + 1 : 0;
 		son_number = B_N_CHILD_NUM(tb->FR[h], child_position);
-		reiserfs_write_unlock(sb);
+		depth = reiserfs_write_unlock_nested(tb->tb_sb);
 		bh = sb_bread(sb, son_number);
-		reiserfs_write_lock(sb);
+		reiserfs_write_lock_nested(tb->tb_sb, depth);
 		if (!bh)
 			return IO_ERROR;
 		if (FILESYSTEM_CHANGED_TB(tb)) {
@@ -2272,6 +2273,7 @@ static int wait_tb_buffers_until_unlocked(struct tree_balance *tb)
 		}
 
 		if (locked) {
+			int depth;
 #ifdef CONFIG_REISERFS_CHECK
 			repeat_counter++;
 			if ((repeat_counter % 10000) == 0) {
@@ -2286,9 +2288,9 @@ static int wait_tb_buffers_until_unlocked(struct tree_balance *tb)
 				    REPEAT_SEARCH : CARRY_ON;
 			}
 #endif
-			reiserfs_write_unlock(tb->tb_sb);
+			depth = reiserfs_write_unlock_nested(tb->tb_sb);
 			__wait_on_buffer(locked);
-			reiserfs_write_lock(tb->tb_sb);
+			reiserfs_write_lock_nested(tb->tb_sb, depth);
 			if (FILESYSTEM_CHANGED_TB(tb))
 				return REPEAT_SEARCH;
 		}
@@ -2359,9 +2361,9 @@ int fix_nodes(int op_mode, struct tree_balance *tb,
 
 	/* if it possible in indirect_to_direct conversion */
 	if (buffer_locked(tbS0)) {
-		reiserfs_write_unlock(tb->tb_sb);
+		int depth = reiserfs_write_unlock_nested(tb->tb_sb);
 		__wait_on_buffer(tbS0);
-		reiserfs_write_lock(tb->tb_sb);
+		reiserfs_write_lock_nested(tb->tb_sb, depth);
 		if (FILESYSTEM_CHANGED_TB(tb))
 			return REPEAT_SEARCH;
 	}
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 0048cc16a6a..ad62bdbb451 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -30,7 +30,6 @@ void reiserfs_evict_inode(struct inode *inode)
 	    JOURNAL_PER_BALANCE_CNT * 2 +
 	    2 * REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb);
 	struct reiserfs_transaction_handle th;
-	int depth;
 	int err;
 
 	if (!inode->i_nlink && !is_bad_inode(inode))
@@ -40,12 +39,13 @@ void reiserfs_evict_inode(struct inode *inode)
 	if (inode->i_nlink)
 		goto no_delete;
 
-	depth = reiserfs_write_lock_once(inode->i_sb);
-
 	/* The = 0 happens when we abort creating a new inode for some reason like lack of space.. */
 	if (!(inode->i_state & I_NEW) && INODE_PKEY(inode)->k_objectid != 0) {	/* also handles bad_inode case */
+
 		reiserfs_delete_xattrs(inode);
 
+		reiserfs_write_lock(inode->i_sb);
+
 		if (journal_begin(&th, inode->i_sb, jbegin_count))
 			goto out;
 		reiserfs_update_inode_transaction(inode);
@@ -57,8 +57,11 @@ void reiserfs_evict_inode(struct inode *inode)
 		/* Do quota update inside a transaction for journaled quotas. We must do that
 		 * after delete_object so that quota updates go into the same transaction as
 		 * stat data deletion */
-		if (!err) 
+		if (!err) {
+			int depth = reiserfs_write_unlock_nested(inode->i_sb);
 			dquot_free_inode(inode);
+			reiserfs_write_lock_nested(inode->i_sb, depth);
+		}
 
 		if (journal_end(&th, inode->i_sb, jbegin_count))
 			goto out;
@@ -72,12 +75,12 @@ void reiserfs_evict_inode(struct inode *inode)
 		/* all items of file are deleted, so we can remove "save" link */
 		remove_save_link(inode, 0 /* not truncate */ );	/* we can't do anything
 								 * about an error here */
+out:
+		reiserfs_write_unlock(inode->i_sb);
 	} else {
 		/* no object items are in the tree */
 		;
 	}
-      out:
-	reiserfs_write_unlock_once(inode->i_sb, depth);
 	clear_inode(inode);	/* note this must go after the journal_end to prevent deadlock */
 	dquot_drop(inode);
 	inode->i_blocks = 0;
@@ -610,7 +613,6 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 	__le32 *item;
 	int done;
 	int fs_gen;
-	int lock_depth;
 	struct reiserfs_transaction_handle *th = NULL;
 	/* space reserved in transaction batch:
 	   . 3 balancings in direct->indirect conversion
@@ -626,11 +628,11 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 	loff_t new_offset =
 	    (((loff_t) block) << inode->i_sb->s_blocksize_bits) + 1;
 
-	lock_depth = reiserfs_write_lock_once(inode->i_sb);
+	reiserfs_write_lock(inode->i_sb);
 	version = get_inode_item_key_version(inode);
 
 	if (!file_capable(inode, block)) {
-		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+		reiserfs_write_unlock(inode->i_sb);
 		return -EFBIG;
 	}
 
@@ -642,7 +644,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 		/* find number of block-th logical block of the file */
 		ret = _get_block_create_0(inode, block, bh_result,
 					  create | GET_BLOCK_READ_DIRECT);
-		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+		reiserfs_write_unlock(inode->i_sb);
 		return ret;
 	}
 	/*
@@ -760,7 +762,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 		if (!dangle && th)
 			retval = reiserfs_end_persistent_transaction(th);
 
-		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+		reiserfs_write_unlock(inode->i_sb);
 
 		/* the item was found, so new blocks were not added to the file
 		 ** there is no need to make sure the inode is updated with this
@@ -1011,11 +1013,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 		 * long time.  reschedule if needed and also release the write
 		 * lock for others.
 		 */
-		if (need_resched()) {
-			reiserfs_write_unlock_once(inode->i_sb, lock_depth);
-			schedule();
-			lock_depth = reiserfs_write_lock_once(inode->i_sb);
-		}
+		reiserfs_cond_resched(inode->i_sb);
 
 		retval = search_for_position_by_key(inode->i_sb, &key, &path);
 		if (retval == IO_ERROR) {
@@ -1050,7 +1048,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 			retval = err;
 	}
 
-	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+	reiserfs_write_unlock(inode->i_sb);
 	reiserfs_check_path(&path);
 	return retval;
 }
@@ -1509,14 +1507,15 @@ struct inode *reiserfs_iget(struct super_block *s, const struct cpu_key *key)
 {
 	struct inode *inode;
 	struct reiserfs_iget_args args;
+	int depth;
 
 	args.objectid = key->on_disk_key.k_objectid;
 	args.dirid = key->on_disk_key.k_dir_id;
-	reiserfs_write_unlock(s);
+	depth = reiserfs_write_unlock_nested(s);
 	inode = iget5_locked(s, key->on_disk_key.k_objectid,
 			     reiserfs_find_actor, reiserfs_init_locked_inode,
 			     (void *)(&args));
-	reiserfs_write_lock(s);
+	reiserfs_write_lock_nested(s, depth);
 	if (!inode)
 		return ERR_PTR(-ENOMEM);
 
@@ -1772,7 +1771,7 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 		       struct inode *inode,
 		       struct reiserfs_security_handle *security)
 {
-	struct super_block *sb;
+	struct super_block *sb = dir->i_sb;
 	struct reiserfs_iget_args args;
 	INITIALIZE_PATH(path_to_key);
 	struct cpu_key key;
@@ -1780,12 +1779,13 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	struct stat_data sd;
 	int retval;
 	int err;
+	int depth;
 
 	BUG_ON(!th->t_trans_id);
 
-	reiserfs_write_unlock(inode->i_sb);
+	depth = reiserfs_write_unlock_nested(sb);
 	err = dquot_alloc_inode(inode);
-	reiserfs_write_lock(inode->i_sb);
+	reiserfs_write_lock_nested(sb, depth);
 	if (err)
 		goto out_end_trans;
 	if (!dir->i_nlink) {
@@ -1793,8 +1793,6 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 		goto out_bad_inode;
 	}
 
-	sb = dir->i_sb;
-
 	/* item head of new item */
 	ih.ih_key.k_dir_id = reiserfs_choose_packing(dir);
 	ih.ih_key.k_objectid = cpu_to_le32(reiserfs_get_unused_objectid(th));
@@ -1812,10 +1810,10 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	memcpy(INODE_PKEY(inode), &(ih.ih_key), KEY_SIZE);
 	args.dirid = le32_to_cpu(ih.ih_key.k_dir_id);
 
-	reiserfs_write_unlock(inode->i_sb);
+	depth = reiserfs_write_unlock_nested(inode->i_sb);
 	err = insert_inode_locked4(inode, args.objectid,
 			     reiserfs_find_actor, &args);
-	reiserfs_write_lock(inode->i_sb);
+	reiserfs_write_lock_nested(inode->i_sb, depth);
 	if (err) {
 		err = -EINVAL;
 		goto out_bad_inode;
@@ -1941,7 +1939,9 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	}
 
 	if (reiserfs_posixacl(inode->i_sb)) {
+		reiserfs_write_unlock(inode->i_sb);
 		retval = reiserfs_inherit_default_acl(th, dir, dentry, inode);
+		reiserfs_write_lock(inode->i_sb);
 		if (retval) {
 			err = retval;
 			reiserfs_check_path(&path_to_key);
@@ -1956,7 +1956,9 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 		inode->i_flags |= S_PRIVATE;
 
 	if (security->name) {
+		reiserfs_write_unlock(inode->i_sb);
 		retval = reiserfs_security_write(th, inode, security);
+		reiserfs_write_lock(inode->i_sb);
 		if (retval) {
 			err = retval;
 			reiserfs_check_path(&path_to_key);
@@ -1982,14 +1984,16 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	INODE_PKEY(inode)->k_objectid = 0;
 
 	/* Quota change must be inside a transaction for journaling */
+	depth = reiserfs_write_unlock_nested(inode->i_sb);
 	dquot_free_inode(inode);
+	reiserfs_write_lock_nested(inode->i_sb, depth);
 
       out_end_trans:
 	journal_end(th, th->t_super, th->t_blocks_allocated);
-	reiserfs_write_unlock(inode->i_sb);
 	/* Drop can be outside and it needs more credits so it's better to have it outside */
+	depth = reiserfs_write_unlock_nested(inode->i_sb);
 	dquot_drop(inode);
-	reiserfs_write_lock(inode->i_sb);
+	reiserfs_write_lock_nested(inode->i_sb, depth);
 	inode->i_flags |= S_NOQUOTA;
 	make_bad_inode(inode);
 
@@ -2103,9 +2107,8 @@ int reiserfs_truncate_file(struct inode *inode, int update_timestamps)
 	int error;
 	struct buffer_head *bh = NULL;
 	int err2;
-	int lock_depth;
 
-	lock_depth = reiserfs_write_lock_once(inode->i_sb);
+	reiserfs_write_lock(inode->i_sb);
 
 	if (inode->i_size > 0) {
 		error = grab_tail_page(inode, &page, &bh);
@@ -2174,7 +2177,7 @@ int reiserfs_truncate_file(struct inode *inode, int update_timestamps)
 		page_cache_release(page);
 	}
 
-	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+	reiserfs_write_unlock(inode->i_sb);
 
 	return 0;
       out:
@@ -2183,7 +2186,7 @@ int reiserfs_truncate_file(struct inode *inode, int update_timestamps)
 		page_cache_release(page);
 	}
 
-	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+	reiserfs_write_unlock(inode->i_sb);
 
 	return error;
 }
@@ -2648,10 +2651,11 @@ int __reiserfs_write_begin(struct page *page, unsigned from, unsigned len)
 	struct inode *inode = page->mapping->host;
 	int ret;
 	int old_ref = 0;
+	int depth;
 
-	reiserfs_write_unlock(inode->i_sb);
+	depth = reiserfs_write_unlock_nested(inode->i_sb);
 	reiserfs_wait_on_write_block(inode->i_sb);
-	reiserfs_write_lock(inode->i_sb);
+	reiserfs_write_lock_nested(inode->i_sb, depth);
 
 	fix_tail_page_for_writing(page);
 	if (reiserfs_transaction_running(inode->i_sb)) {
@@ -2708,7 +2712,6 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 	int update_sd = 0;
 	struct reiserfs_transaction_handle *th;
 	unsigned start;
-	int lock_depth = 0;
 	bool locked = false;
 
 	if ((unsigned long)fsdata & AOP_FLAG_CONT_EXPAND)
@@ -2737,7 +2740,7 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 	 */
 	if (pos + copied > inode->i_size) {
 		struct reiserfs_transaction_handle myth;
-		lock_depth = reiserfs_write_lock_once(inode->i_sb);
+		reiserfs_write_lock(inode->i_sb);
 		locked = true;
 		/* If the file have grown beyond the border where it
 		   can have a tail, unmark it as needing a tail
@@ -2768,7 +2771,7 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 	}
 	if (th) {
 		if (!locked) {
-			lock_depth = reiserfs_write_lock_once(inode->i_sb);
+			reiserfs_write_lock(inode->i_sb);
 			locked = true;
 		}
 		if (!update_sd)
@@ -2780,7 +2783,7 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 
       out:
 	if (locked)
-		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+		reiserfs_write_unlock(inode->i_sb);
 	unlock_page(page);
 	page_cache_release(page);
 
@@ -2790,7 +2793,7 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 	return ret == 0 ? copied : ret;
 
       journal_error:
-	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+	reiserfs_write_unlock(inode->i_sb);
 	locked = false;
 	if (th) {
 		if (!update_sd)
@@ -2808,10 +2811,11 @@ int reiserfs_commit_write(struct file *f, struct page *page,
 	int ret = 0;
 	int update_sd = 0;
 	struct reiserfs_transaction_handle *th = NULL;
+	int depth;
 
-	reiserfs_write_unlock(inode->i_sb);
+	depth = reiserfs_write_unlock_nested(inode->i_sb);
 	reiserfs_wait_on_write_block(inode->i_sb);
-	reiserfs_write_lock(inode->i_sb);
+	reiserfs_write_lock_nested(inode->i_sb, depth);
 
 	if (reiserfs_transaction_running(inode->i_sb)) {
 		th = current->journal_info;
@@ -3110,7 +3114,6 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = dentry->d_inode;
 	unsigned int ia_valid;
-	int depth;
 	int error;
 
 	error = inode_change_ok(inode, attr);
@@ -3122,13 +3125,14 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 
 	if (is_quota_modification(inode, attr))
 		dquot_initialize(inode);
-	depth = reiserfs_write_lock_once(inode->i_sb);
+	reiserfs_write_lock(inode->i_sb);
 	if (attr->ia_valid & ATTR_SIZE) {
 		/* version 2 items will be caught by the s_maxbytes check
 		 ** done for us in vmtruncate
 		 */
 		if (get_inode_item_key_version(inode) == KEY_FORMAT_3_5 &&
 		    attr->ia_size > MAX_NON_LFS) {
+			reiserfs_write_unlock(inode->i_sb);
 			error = -EFBIG;
 			goto out;
 		}
@@ -3150,8 +3154,10 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 				if (err)
 					error = err;
 			}
-			if (error)
+			if (error) {
+				reiserfs_write_unlock(inode->i_sb);
 				goto out;
+			}
 			/*
 			 * file size is changed, ctime and mtime are
 			 * to be updated
@@ -3159,6 +3165,7 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 			attr->ia_valid |= (ATTR_MTIME | ATTR_CTIME);
 		}
 	}
+	reiserfs_write_unlock(inode->i_sb);
 
 	if ((((attr->ia_valid & ATTR_UID) && (from_kuid(&init_user_ns, attr->ia_uid) & ~0xffff)) ||
 	     ((attr->ia_valid & ATTR_GID) && (from_kgid(&init_user_ns, attr->ia_gid) & ~0xffff))) &&
@@ -3183,14 +3190,16 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 			return error;
 
 		/* (user+group)*(old+new) structure - we count quota info and , inode write (sb, inode) */
+		reiserfs_write_lock(inode->i_sb);
 		error = journal_begin(&th, inode->i_sb, jbegin_count);
+		reiserfs_write_unlock(inode->i_sb);
 		if (error)
 			goto out;
-		reiserfs_write_unlock_once(inode->i_sb, depth);
 		error = dquot_transfer(inode, attr);
-		depth = reiserfs_write_lock_once(inode->i_sb);
+		reiserfs_write_lock(inode->i_sb);
 		if (error) {
 			journal_end(&th, inode->i_sb, jbegin_count);
+			reiserfs_write_unlock(inode->i_sb);
 			goto out;
 		}
 
@@ -3202,17 +3211,11 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 			inode->i_gid = attr->ia_gid;
 		mark_inode_dirty(inode);
 		error = journal_end(&th, inode->i_sb, jbegin_count);
+		reiserfs_write_unlock(inode->i_sb);
 		if (error)
 			goto out;
 	}
 
-	/*
-	 * Relax the lock here, as it might truncate the
-	 * inode pages and wait for inode pages locks.
-	 * To release such page lock, the owner needs the
-	 * reiserfs lock
-	 */
-	reiserfs_write_unlock_once(inode->i_sb, depth);
 	if ((attr->ia_valid & ATTR_SIZE) &&
 	    attr->ia_size != i_size_read(inode)) {
 		error = inode_newsize_ok(inode, attr->ia_size);
@@ -3226,16 +3229,13 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 		setattr_copy(inode, attr);
 		mark_inode_dirty(inode);
 	}
-	depth = reiserfs_write_lock_once(inode->i_sb);
 
 	if (!error && reiserfs_posixacl(inode->i_sb)) {
 		if (attr->ia_valid & ATTR_MODE)
 			error = reiserfs_acl_chmod(inode);
 	}
 
-      out:
-	reiserfs_write_unlock_once(inode->i_sb, depth);
-
+out:
 	return error;
 }
 
diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c
index 15cb5fe6b42..946ccbf5b5a 100644
--- a/fs/reiserfs/ioctl.c
+++ b/fs/reiserfs/ioctl.c
@@ -167,7 +167,6 @@ int reiserfs_commit_write(struct file *f, struct page *page,
 int reiserfs_unpack(struct inode *inode, struct file *filp)
 {
 	int retval = 0;
-	int depth;
 	int index;
 	struct page *page;
 	struct address_space *mapping;
@@ -183,11 +182,11 @@ int reiserfs_unpack(struct inode *inode, struct file *filp)
 		return 0;
 	}
 
-	depth = reiserfs_write_lock_once(inode->i_sb);
-
 	/* we need to make sure nobody is changing the file size beneath us */
 	reiserfs_mutex_lock_safe(&inode->i_mutex, inode->i_sb);
 
+	reiserfs_write_lock(inode->i_sb);
+
 	write_from = inode->i_size & (blocksize - 1);
 	/* if we are on a block boundary, we are already unpacked.  */
 	if (write_from == 0) {
@@ -221,6 +220,6 @@ int reiserfs_unpack(struct inode *inode, struct file *filp)
 
       out:
 	mutex_unlock(&inode->i_mutex);
-	reiserfs_write_unlock_once(inode->i_sb, depth);
+	reiserfs_write_unlock(inode->i_sb);
 	return retval;
 }
diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index 742fdd4c209..73feacc49b2 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -947,9 +947,11 @@ static int reiserfs_async_progress_wait(struct super_block *s)
 	struct reiserfs_journal *j = SB_JOURNAL(s);
 
 	if (atomic_read(&j->j_async_throttle)) {
-		reiserfs_write_unlock(s);
+		int depth;
+
+		depth = reiserfs_write_unlock_nested(s);
 		congestion_wait(BLK_RW_ASYNC, HZ / 10);
-		reiserfs_write_lock(s);
+		reiserfs_write_lock_nested(s, depth);
 	}
 
 	return 0;
@@ -972,6 +974,7 @@ static int flush_commit_list(struct super_block *s,
 	struct reiserfs_journal *journal = SB_JOURNAL(s);
 	int retval = 0;
 	int write_len;
+	int depth;
 
 	reiserfs_check_lock_depth(s, "flush_commit_list");
 
@@ -1018,12 +1021,12 @@ static int flush_commit_list(struct super_block *s,
 		 * We might sleep in numerous places inside
 		 * write_ordered_buffers. Relax the write lock.
 		 */
-		reiserfs_write_unlock(s);
+		depth = reiserfs_write_unlock_nested(s);
 		ret = write_ordered_buffers(&journal->j_dirty_buffers_lock,
 					    journal, jl, &jl->j_bh_list);
 		if (ret < 0 && retval == 0)
 			retval = ret;
-		reiserfs_write_lock(s);
+		reiserfs_write_lock_nested(s, depth);
 	}
 	BUG_ON(!list_empty(&jl->j_bh_list));
 	/*
@@ -1043,9 +1046,9 @@ static int flush_commit_list(struct super_block *s,
 		tbh = journal_find_get_block(s, bn);
 		if (tbh) {
 			if (buffer_dirty(tbh)) {
-		            reiserfs_write_unlock(s);
+		            depth = reiserfs_write_unlock_nested(s);
 			    ll_rw_block(WRITE, 1, &tbh);
-			    reiserfs_write_lock(s);
+			    reiserfs_write_lock_nested(s, depth);
 			}
 			put_bh(tbh) ;
 		}
@@ -1057,17 +1060,17 @@ static int flush_commit_list(struct super_block *s,
 		    (jl->j_start + i) % SB_ONDISK_JOURNAL_SIZE(s);
 		tbh = journal_find_get_block(s, bn);
 
-		reiserfs_write_unlock(s);
-		wait_on_buffer(tbh);
-		reiserfs_write_lock(s);
+		depth = reiserfs_write_unlock_nested(s);
+		__wait_on_buffer(tbh);
+		reiserfs_write_lock_nested(s, depth);
 		// since we're using ll_rw_blk above, it might have skipped over
 		// a locked buffer.  Double check here
 		//
 		/* redundant, sync_dirty_buffer() checks */
 		if (buffer_dirty(tbh)) {
-			reiserfs_write_unlock(s);
+			depth = reiserfs_write_unlock_nested(s);
 			sync_dirty_buffer(tbh);
-			reiserfs_write_lock(s);
+			reiserfs_write_lock_nested(s, depth);
 		}
 		if (unlikely(!buffer_uptodate(tbh))) {
 #ifdef CONFIG_REISERFS_CHECK
@@ -1091,12 +1094,12 @@ static int flush_commit_list(struct super_block *s,
 		if (buffer_dirty(jl->j_commit_bh))
 			BUG();
 		mark_buffer_dirty(jl->j_commit_bh) ;
-		reiserfs_write_unlock(s);
+		depth = reiserfs_write_unlock_nested(s);
 		if (reiserfs_barrier_flush(s))
 			__sync_dirty_buffer(jl->j_commit_bh, WRITE_FLUSH_FUA);
 		else
 			sync_dirty_buffer(jl->j_commit_bh);
-		reiserfs_write_lock(s);
+		reiserfs_write_lock_nested(s, depth);
 	}
 
 	/* If there was a write error in the journal - we can't commit this
@@ -1228,15 +1231,16 @@ static int _update_journal_header_block(struct super_block *sb,
 {
 	struct reiserfs_journal_header *jh;
 	struct reiserfs_journal *journal = SB_JOURNAL(sb);
+	int depth;
 
 	if (reiserfs_is_journal_aborted(journal))
 		return -EIO;
 
 	if (trans_id >= journal->j_last_flush_trans_id) {
 		if (buffer_locked((journal->j_header_bh))) {
-			reiserfs_write_unlock(sb);
-			wait_on_buffer((journal->j_header_bh));
-			reiserfs_write_lock(sb);
+			depth = reiserfs_write_unlock_nested(sb);
+			__wait_on_buffer(journal->j_header_bh);
+			reiserfs_write_lock_nested(sb, depth);
 			if (unlikely(!buffer_uptodate(journal->j_header_bh))) {
 #ifdef CONFIG_REISERFS_CHECK
 				reiserfs_warning(sb, "journal-699",
@@ -1254,14 +1258,14 @@ static int _update_journal_header_block(struct super_block *sb,
 		jh->j_mount_id = cpu_to_le32(journal->j_mount_id);
 
 		set_buffer_dirty(journal->j_header_bh);
-		reiserfs_write_unlock(sb);
+		depth = reiserfs_write_unlock_nested(sb);
 
 		if (reiserfs_barrier_flush(sb))
 			__sync_dirty_buffer(journal->j_header_bh, WRITE_FLUSH_FUA);
 		else
 			sync_dirty_buffer(journal->j_header_bh);
 
-		reiserfs_write_lock(sb);
+		reiserfs_write_lock_nested(sb, depth);
 		if (!buffer_uptodate(journal->j_header_bh)) {
 			reiserfs_warning(sb, "journal-837",
 					 "IO error during journal replay");
@@ -1341,6 +1345,7 @@ static int flush_journal_list(struct super_block *s,
 	unsigned long j_len_saved = jl->j_len;
 	struct reiserfs_journal *journal = SB_JOURNAL(s);
 	int err = 0;
+	int depth;
 
 	BUG_ON(j_len_saved <= 0);
 
@@ -1495,9 +1500,9 @@ static int flush_journal_list(struct super_block *s,
 						       "cn->bh is NULL");
 				}
 
-				reiserfs_write_unlock(s);
-				wait_on_buffer(cn->bh);
-				reiserfs_write_lock(s);
+				depth = reiserfs_write_unlock_nested(s);
+				__wait_on_buffer(cn->bh);
+				reiserfs_write_lock_nested(s, depth);
 
 				if (!cn->bh) {
 					reiserfs_panic(s, "journal-1012",
@@ -1974,6 +1979,7 @@ static int journal_compare_desc_commit(struct super_block *sb,
 /* returns 0 if it did not find a description block
 ** returns -1 if it found a corrupt commit block
 ** returns 1 if both desc and commit were valid
+** NOTE: only called during fs mount
 */
 static int journal_transaction_is_valid(struct super_block *sb,
 					struct buffer_head *d_bh,
@@ -2073,8 +2079,9 @@ static void brelse_array(struct buffer_head **heads, int num)
 
 /*
 ** given the start, and values for the oldest acceptable transactions,
-** this either reads in a replays a transaction, or returns because the transaction
-** is invalid, or too old.
+** this either reads in a replays a transaction, or returns because the
+** transaction is invalid, or too old.
+** NOTE: only called during fs mount
 */
 static int journal_read_transaction(struct super_block *sb,
 				    unsigned long cur_dblock,
@@ -2208,10 +2215,7 @@ static int journal_read_transaction(struct super_block *sb,
 	ll_rw_block(READ, get_desc_trans_len(desc), log_blocks);
 	for (i = 0; i < get_desc_trans_len(desc); i++) {
 
-		reiserfs_write_unlock(sb);
 		wait_on_buffer(log_blocks[i]);
-		reiserfs_write_lock(sb);
-
 		if (!buffer_uptodate(log_blocks[i])) {
 			reiserfs_warning(sb, "journal-1212",
 					 "REPLAY FAILURE fsck required! "
@@ -2318,12 +2322,13 @@ static struct buffer_head *reiserfs_breada(struct block_device *dev,
 
 /*
 ** read and replay the log
-** on a clean unmount, the journal header's next unflushed pointer will be to an invalid
-** transaction.  This tests that before finding all the transactions in the log, which makes normal mount times fast.
-**
-** After a crash, this starts with the next unflushed transaction, and replays until it finds one too old, or invalid.
-**
+** on a clean unmount, the journal header's next unflushed pointer will
+** be to an invalid transaction.  This tests that before finding all the
+** transactions in the log, which makes normal mount times fast.
+** After a crash, this starts with the next unflushed transaction, and
+** replays until it finds one too old, or invalid.
 ** On exit, it sets things up so the first transaction will work correctly.
+** NOTE: only called during fs mount
 */
 static int journal_read(struct super_block *sb)
 {
@@ -2501,14 +2506,18 @@ static int journal_read(struct super_block *sb)
 			      "replayed %d transactions in %lu seconds\n",
 			      replay_count, get_seconds() - start);
 	}
+	/* needed to satisfy the locking in _update_journal_header_block */
+	reiserfs_write_lock(sb);
 	if (!bdev_read_only(sb->s_bdev) &&
 	    _update_journal_header_block(sb, journal->j_start,
 					 journal->j_last_flush_trans_id)) {
+		reiserfs_write_unlock(sb);
 		/* replay failed, caller must call free_journal_ram and abort