Merge git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw: (32 commits)
  GFS2: Move all locking inside the inode creation function
  GFS2: Clean up symlink creation
  GFS2: Clean up mkdir
  GFS2: Use UUID field in generic superblock
  GFS2: Rename ops_inode.c to inode.c
  GFS2: Inode.c is empty now, remove it
  GFS2: Move final part of inode.c into super.c
  GFS2: Move most of the remaining inode.c into ops_inode.c
  GFS2: Move gfs2_refresh_inode() and friends into glops.c
  GFS2: Remove gfs2_dinode_print() function
  GFS2: When adding a new dir entry, inc link count if it is a subdir
  GFS2: Make gfs2_dir_del update link count when required
  GFS2: Don't use gfs2_change_nlink in link syscall
  GFS2: Don't use a try lock when promoting to a higher mode
  GFS2: Double check link count under glock
  GFS2: Improve bug trap code in ->releasepage()
  GFS2: Fix ail list traversal
  GFS2: make sure fallocate bytes is a multiple of blksize
  GFS2: Add an AIL writeback tracepoint
  GFS2: Make writeback more responsive to system conditions
  ...

fs/gfs2/Makefile

 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

fs/gfs2/aops.c

@@ -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;

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)
 {
@@ -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);
@@ -1767,95 +1769,21 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
 	return 0;
 }
 
-/**
- * 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;
@@ -1887,13 +1815,17 @@ 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) {
+		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);
+		if (blk != leaf_no)
 			brelse(bh);
 
 		gfs2_rlist_add(sdp, &rlist, blk);
@@ -1918,12 +1850,17 @@ 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) {
+		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);
+		if (blk != leaf_no)
 			brelse(bh);
 
 		gfs2_free_meta(dip, blk, 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;
+
+	while (index < hsize) {
+		lp_offset = index & (sdp->sd_hash_ptrs - 1);
+		ht_offset = index - lp_offset;
 
-	error = gfs2_trans_begin(sdp, RES_DINODE, 0);
+		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)
-		return error;
+				goto out;
+			leaf = (struct gfs2_leaf *)bh->b_data;
+			len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth));
 
-	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);
+			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;
 }

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,

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);

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;
 }
 
 /**
@@ -826,6 +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;
 
@@ -833,13 +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);
@@ -870,6 +863,9 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset,
 		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;

fs/gfs2/glock.c

@@ -143,14 +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);
@@ -158,17 +153,8 @@ static int demote_ok(const struct gfs2_glock *gl)
 }
 
 
-/**
- * __gfs2_glock_schedule_for_reclaim - Add a glock to the reclaim list
- * @gl: the glock
- *
- * If the glock is demotable, then we add it (or move it) to the end
- * of the glock LRU list.
- */
-
-static void __gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
+void gfs2_glock_add_to_lru(struct gfs2_glock *gl)
 {
-	if (demote_ok(gl)) {
 	spin_lock(&lru_lock);
 
 	if (!list_empty(&gl->gl_lru))
@@ -177,15 +163,33 @@ static void __gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
 		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);
 }
 
-void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
+/**
+ * __gfs2_glock_schedule_for_reclaim - Add a glock to the reclaim list
+ * @gl: the glock
+ *
+ * If the glock is demotable, then we add it (or move it) to the end
+ * of the glock LRU list.
+ */
+
+static 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);
 }
 
 /**
@@ -217,12 +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);
@@ -542,11 +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);
@@ -648,7 +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)) {
@@ -1025,6 +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) &&
@@ -1082,6 +1079,7 @@ void gfs2_glock_dq(struct gfs2_holder *gh)
 		    !test_bit(GLF_DEMOTE, &gl->gl_flags))
 			fast_path = 1;
 	}
+	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);
@@ -1365,6 +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 */
@@ -1387,6 +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);
@@ -1459,12 +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)
@@ -1599,9 +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))
@@ -1624,6 +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;
 }
@@ -1658,14 +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) {

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);

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)) {
@@ -76,7 +61,47 @@ 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);
 }
@@ -226,6 +251,119 @@ static int inode_go_demote_ok(const struct gfs2_glock *gl)
 	return 1;
 }
 
+/**
+ * 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

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__ */

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 */
 

fs/gfs2/inode.c

 /*
  * 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"
@@ -26,19 +28,14 @@
 #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"
-
-struct gfs2_inum_range_host {
-	u64 ir_start;
-	u64 ir_length;
-};
+#include "super.h"
+#include "glops.h"
 
 struct gfs2_skip_data {
 	u64 no_addr;
@@ -74,14 +71,14 @@ static int iget_set(struct inode *inode, void *opaque)
 	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;
 	struct gfs2_skip_data data;
 
 	data.no_addr = no_addr;
 	data.skipped = 0;
-	data.non_block = 0;
+	data.non_block = non_block;
 	return ilookup5(sb, hash, iget_test, &data);
 }
 
@@ -248,203 +245,6 @@ struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
 	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)
 {
@@ -543,7 +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) {
@@ -613,21 +413,44 @@ static int alloc_dinode(struct gfs2_inode *dip, u64 *no_addr, u64 *generation)
 	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;
@@ -646,7 +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));
@@ -654,16 +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;
@@ -678,6 +491,25 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl,
 	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;
@@ -685,7 +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;
@@ -707,7 +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);
 
@@ -761,14 +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);
@@ -815,27 +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
- *
- * @ghs[0] is an initialized holder for the directory
- * @ghs[1] is the holder for the inode lock
+ * 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)
  *
- * 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;
@@ -843,10 +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;
 
@@ -864,7 +696,7 @@ 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;
 
@@ -891,118 +723,1170 @@ 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;
 }
 
-static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr)
+/**
+ * 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 = &ip->i_inode;
-	struct buffer_head *dibh;
-	int error;
+	struct inode *inode;
+	int ret;
 
-	error = gfs2_meta_inode_buffer(ip, &dibh);
-	if (error)
-		return error;
+	for (;;) {
+		ret = gfs2_create_inode(dir, dentry, S_IFREG | mode, 0, NULL, 0);
+		if (ret != -EEXIST || (nd && (nd->flags & LOOKUP_EXCL)))
+			return ret;
 
-	setattr_copy(inode, attr);
-	mark_inode_dirty(inode);
-	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
-	gfs2_dinode_out(ip, dibh->b_data);
-	brelse(dibh);
+		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_setattr_simple -
- * @ip:
- * @attr:
+ * 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 with a reference on the vnode.
+ * Called by the VFS layer. Lock dir and call gfs2_lookupi()
  *
  * Returns: errno
  */
 
-int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr)
+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);
 
-	if (current->journal_info)
-		return __gfs2_setattr_simple(ip, attr);
+	return NULL;
+}
 
-	error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE, 0);
+/**
+ * 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)
-		return error;
+		goto out_parent;
 
-	error = __gfs2_setattr_simple(ip, attr);
-	gfs2_trans_end(GFS2_SB(&ip->i_inode));
+	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;
 }
 
-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);
-}
+/*
+ * 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)
+{
+	struct inode *inode = &ip->i_inode;
+	struct buffer_head *dibh;
+	int error;
+
+	error = gfs2_meta_inode_buffer(ip, &dibh);
+	if (error)
+		return error;
+
+	setattr_copy(inode, attr);
+	mark_inode_dirty(inode);
+	gfs2_trans_add_bh(ip->i_gl, dibh, 1);
+	gfs2_dinode_out(ip, dibh->b_data);
+	brelse(dibh);
+	return 0;
+}
+
+/**
+ * gfs2_setattr_simple -
+ * @ip:
+ * @attr:
+ *
+ * Returns: errno
+ */
+
+int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr)
+{
+	int error;
+
+	if (current->journal_info)
+		return __gfs2_setattr_simple(ip, attr);
+
+	error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE, 0);
+	if (error)
+		return error;
+
+	error = __gfs2_setattr_simple(ip, attr);
+	gfs2_trans_end(GFS2_SB(&ip->i_inode));
+	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_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,
+};
 

fs/gfs2/inode.h

@@ -102,22 +102,16 @@ extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type,
 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;

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"
@@ -83,23 +84,23 @@ 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) {
+	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);
@@ -113,25 +114,67 @@ __acquires(&sdp->sd_ail_lock)
 
 		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);
-
-			get_bh(bh);
+		mapping = bh->b_page->mapping;
+		if (!mapping)
+			continue;
 		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);
-			}
+		generic_writepages(mapping, wbc);
 		spin_lock(&sdp->sd_ail_lock);
+		if (wbc->nr_to_write <= 0)
+			break;
+		return 1;
+	}
 
-			retry = 1;
+	return 0;
+}
+
+
+/**
+ * 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;
+		if (gfs2_ail1_start_one(sdp, wbc, ai))
+			goto restart;
 	}
-	} while (retry);
+	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)
+		if (buffer_busy(bh))
 			continue;
-			else
-				break;
-		}
-
 		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);

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);

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,

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)

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) {

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__ */
 

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;
 }
 
@@ -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;
 	}

fs/gfs2/ops_inode.c

-/*
- * 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,
-};
-

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);
 		}
 
@@ -1365,7 +1377,7 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal,
 
 	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 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal,
 		    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;

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 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc)
 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 @@ static void gfs2_put_super(struct super_block *sb)
 
 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
@@ -1371,15 +1490,13 @@ static void gfs2_evict_inode(struct inode *inode)
 	}
 
 	error = gfs2_dinode_dealloc(ip);
-	if (error)
 	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:
@@ -1394,6 +1511,7 @@ static void gfs2_evict_inode(struct inode *inode)
 	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) {

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);

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: