Merge remote-tracking branch 'gfs2/master'

Conflicts:
	fs/gfs2/log.c

13 files changed, 247 insertions, 187 deletions

diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c
index c71995b111bf..0f5c4f9d5d62 100644
--- a/fs/gfs2/aops.c
+++ b/fs/gfs2/aops.c
@@ -884,8 +884,8 @@ static int gfs2_write_end(struct file *file, struct address_space *mapping,
 	}
 
 	brelse(dibh);
-	gfs2_trans_end(sdp);
 failed:
+	gfs2_trans_end(sdp);
 	if (al) {
 		gfs2_inplace_release(ip);
 		gfs2_quota_unlock(ip);
diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c
index 5c356d09c321..f7a31374ff82 100644
--- a/fs/gfs2/dir.c
+++ b/fs/gfs2/dir.c
@@ -82,12 +82,9 @@
 struct qstr gfs2_qdot __read_mostly;
 struct qstr gfs2_qdotdot __read_mostly;
 
-typedef int (*leaf_call_t) (struct gfs2_inode *dip, u32 index, u32 len,
-			    u64 leaf_no, void *data);
 typedef int (*gfs2_dscan_t)(const struct gfs2_dirent *dent,
 			    const struct qstr *name, void *opaque);
 
-
 int gfs2_dir_get_new_buffer(struct gfs2_inode *ip, u64 block,
 			    struct buffer_head **bhp)
 {
@@ -1506,7 +1503,7 @@ struct inode *gfs2_dir_search(struct inode *dir, const struct qstr *name)
 		inode = gfs2_inode_lookup(dir->i_sb, 
 				be16_to_cpu(dent->de_type),
 				be64_to_cpu(dent->de_inum.no_addr),
-				be64_to_cpu(dent->de_inum.no_formal_ino));
+				be64_to_cpu(dent->de_inum.no_formal_ino), 0);
 		brelse(bh);
 		return inode;
 	}
@@ -1768,94 +1765,20 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename,
 }
 
 /**
- * foreach_leaf - call a function for each leaf in a directory
- * @dip: the directory
- * @lc: the function to call for each each
- * @data: private data to pass to it
- *
- * Returns: errno
- */
-
-static int foreach_leaf(struct gfs2_inode *dip, leaf_call_t lc, void *data)
-{
-	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
-	struct buffer_head *bh;
-	struct gfs2_leaf *leaf;
-	u32 hsize, len;
-	u32 ht_offset, lp_offset, ht_offset_cur = -1;
-	u32 index = 0;
-	__be64 *lp;
-	u64 leaf_no;
-	int error = 0;
-
-	hsize = 1 << dip->i_depth;
-	if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) {
-		gfs2_consist_inode(dip);
-		return -EIO;
-	}
-
-	lp = kmalloc(sdp->sd_hash_bsize, GFP_NOFS);
-	if (!lp)
-		return -ENOMEM;
-
-	while (index < hsize) {
-		lp_offset = index & (sdp->sd_hash_ptrs - 1);
-		ht_offset = index - lp_offset;
-
-		if (ht_offset_cur != ht_offset) {
-			error = gfs2_dir_read_data(dip, (char *)lp,
-						ht_offset * sizeof(__be64),
-						sdp->sd_hash_bsize, 1);
-			if (error != sdp->sd_hash_bsize) {
-				if (error >= 0)
-					error = -EIO;
-				goto out;
-			}
-			ht_offset_cur = ht_offset;
-		}
-
-		leaf_no = be64_to_cpu(lp[lp_offset]);
-		if (leaf_no) {
-			error = get_leaf(dip, leaf_no, &bh);
-			if (error)
-				goto out;
-			leaf = (struct gfs2_leaf *)bh->b_data;
-			len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth));
-			brelse(bh);
-
-			error = lc(dip, index, len, leaf_no, data);
-			if (error)
-				goto out;
-
-			index = (index & ~(len - 1)) + len;
-		} else
-			index++;
-	}
-
-	if (index != hsize) {
-		gfs2_consist_inode(dip);
-		error = -EIO;
-	}
-
-out:
-	kfree(lp);
-
-	return error;
-}
-
-/**
  * leaf_dealloc - Deallocate a directory leaf
  * @dip: the directory
  * @index: the hash table offset in the directory
  * @len: the number of pointers to this leaf
  * @leaf_no: the leaf number
- * @data: not used
+ * @leaf_bh: buffer_head for the starting leaf
+ * last_dealloc: 1 if this is the final dealloc for the leaf, else 0
  *
  * Returns: errno
  */
 
 static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
-			u64 leaf_no, void *data)
+			u64 leaf_no, struct buffer_head *leaf_bh,
+			int last_dealloc)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
 	struct gfs2_leaf *tmp_leaf;
@@ -1887,14 +1810,18 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
 		goto out_qs;
 
 	/*  Count the number of leaves  */
+	bh = leaf_bh;
 
 	for (blk = leaf_no; blk; blk = nblk) {
-		error = get_leaf(dip, blk, &bh);
-		if (error)
-			goto out_rlist;
+		if (blk != leaf_no) {
+			error = get_leaf(dip, blk, &bh);
+			if (error)
+				goto out_rlist;
+		}
 		tmp_leaf = (struct gfs2_leaf *)bh->b_data;
 		nblk = be64_to_cpu(tmp_leaf->lf_next);
-		brelse(bh);
+		if (blk != leaf_no)
+			brelse(bh);
 
 		gfs2_rlist_add(sdp, &rlist, blk);
 		l_blocks++;
@@ -1918,13 +1845,18 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len,
 	if (error)
 		goto out_rg_gunlock;
 
+	bh = leaf_bh;
+
 	for (blk = leaf_no; blk; blk = nblk) {
-		error = get_leaf(dip, blk, &bh);
-		if (error)
-			goto out_end_trans;
+		if (blk != leaf_no) {
+			error = get_leaf(dip, blk, &bh);
+			if (error)
+				goto out_end_trans;
+		}
 		tmp_leaf = (struct gfs2_leaf *)bh->b_data;
 		nblk = be64_to_cpu(tmp_leaf->lf_next);
-		brelse(bh);
+		if (blk != leaf_no)
+			brelse(bh);
 
 		gfs2_free_meta(dip, blk, 1);
 		gfs2_add_inode_blocks(&dip->i_inode, -1);
@@ -1942,6 +1874,10 @@ 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 +1911,67 @@ int gfs2_dir_exhash_dealloc(struct gfs2_inode *dip)
 {
 	struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode);
 	struct buffer_head *bh;
-	int error;
+	struct gfs2_leaf *leaf;
+	u32 hsize, len;
+	u32 ht_offset, lp_offset, ht_offset_cur = -1;
+	u32 index = 0, next_index;
+	__be64 *lp;
+	u64 leaf_no;
+	int error = 0, last;
 
-	/* Dealloc on-disk leaves to FREEMETA state */
-	error = foreach_leaf(dip, leaf_dealloc, NULL);
-	if (error)
-		return error;
+	hsize = 1 << dip->i_depth;
+	if (hsize * sizeof(u64) != i_size_read(&dip->i_inode)) {
+		gfs2_consist_inode(dip);
+		return -EIO;
+	}
 
-	/* Make this a regular file in case we crash.
-	   (We don't want to free these blocks a second time.)  */
+	lp = kmalloc(sdp->sd_hash_bsize, GFP_NOFS);
+	if (!lp)
+		return -ENOMEM;
 
-	error = gfs2_trans_begin(sdp, RES_DINODE, 0);
-	if (error)
-		return error;
+	while (index < hsize) {
+		lp_offset = index & (sdp->sd_hash_ptrs - 1);
+		ht_offset = index - lp_offset;
 
-	error = gfs2_meta_inode_buffer(dip, &bh);
-	if (!error) {
-		gfs2_trans_add_bh(dip->i_gl, bh, 1);
-		((struct gfs2_dinode *)bh->b_data)->di_mode =
-						cpu_to_be32(S_IFREG);
-		brelse(bh);
+		if (ht_offset_cur != ht_offset) {
+			error = gfs2_dir_read_data(dip, (char *)lp,
+						ht_offset * sizeof(__be64),
+						sdp->sd_hash_bsize, 1);
+			if (error != sdp->sd_hash_bsize) {
+				if (error >= 0)
+					error = -EIO;
+				goto out;
+			}
+			ht_offset_cur = ht_offset;
+		}
+
+		leaf_no = be64_to_cpu(lp[lp_offset]);
+		if (leaf_no) {
+			error = get_leaf(dip, leaf_no, &bh);
+			if (error)
+				goto out;
+			leaf = (struct gfs2_leaf *)bh->b_data;
+			len = 1 << (dip->i_depth - be16_to_cpu(leaf->lf_depth));
+
+			next_index = (index & ~(len - 1)) + len;
+			last = ((next_index >= hsize) ? 1 : 0);
+			error = leaf_dealloc(dip, index, len, leaf_no, bh,
+					     last);
+			brelse(bh);
+			if (error)
+				goto out;
+			index = next_index;
+		} else
+			index++;
 	}
 
-	gfs2_trans_end(sdp);
+	if (index != hsize) {
+		gfs2_consist_inode(dip);
+		error = -EIO;
+	}
+
+out:
+	kfree(lp);
 
 	return error;
 }
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index b2682e073eee..e48310885c48 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -617,18 +617,51 @@ static ssize_t gfs2_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 	return generic_file_aio_write(iocb, iov, nr_segs, pos);
 }
 
-static void empty_write_end(struct page *page, unsigned from,
-			   unsigned to)
+static int empty_write_end(struct page *page, unsigned from,
+			   unsigned to, int mode)
 {
-	struct gfs2_inode *ip = GFS2_I(page->mapping->host);
+	struct inode *inode = page->mapping->host;
+	struct gfs2_inode *ip = GFS2_I(inode);
+	struct buffer_head *bh;
+	unsigned offset, blksize = 1 << inode->i_blkbits;
+	pgoff_t end_index = i_size_read(inode) >> PAGE_CACHE_SHIFT;
 
 	zero_user(page, from, to-from);
 	mark_page_accessed(page);
 
-	if (!gfs2_is_writeback(ip))
-		gfs2_page_add_databufs(ip, page, from, to);
+	if (page->index < end_index || !(mode & FALLOC_FL_KEEP_SIZE)) {
+		if (!gfs2_is_writeback(ip))
+			gfs2_page_add_databufs(ip, page, from, to);
+
+		block_commit_write(page, from, to);
+		return 0;
+	}
+
+	offset = 0;
+	bh = page_buffers(page);
+	while (offset < to) {
+		if (offset >= from) {
+			set_buffer_uptodate(bh);
+			mark_buffer_dirty(bh);
+			clear_buffer_new(bh);
+			write_dirty_buffer(bh, WRITE);
+		}
+		offset += blksize;
+		bh = bh->b_this_page;
+	}
 
-	block_commit_write(page, from, to);
+	offset = 0;
+	bh = page_buffers(page);
+	while (offset < to) {
+		if (offset >= from) {
+			wait_on_buffer(bh);
+			if (!buffer_uptodate(bh))
+				return -EIO;
+		}
+		offset += blksize;
+		bh = bh->b_this_page;
+	}
+	return 0;
 }
 
 static int needs_empty_write(sector_t block, struct inode *inode)
@@ -643,7 +676,8 @@ static int needs_empty_write(sector_t block, struct inode *inode)
 	return !buffer_mapped(&bh_map);
 }
 
-static int write_empty_blocks(struct page *page, unsigned from, unsigned to)
+static int write_empty_blocks(struct page *page, unsigned from, unsigned to,
+			      int mode)
 {
 	struct inode *inode = page->mapping->host;
 	unsigned start, end, next, blksize;
@@ -668,7 +702,9 @@ static int write_empty_blocks(struct page *page, unsigned from, unsigned to)
 							  gfs2_block_map);
 				if (unlikely(ret))
 					return ret;
-				empty_write_end(page, start, end);
+				ret = empty_write_end(page, start, end, mode);
+				if (unlikely(ret))
+					return ret;
 				end = 0;
 			}
 			start = next;
@@ -682,7 +718,9 @@ static int write_empty_blocks(struct page *page, unsigned from, unsigned to)
 		ret = __block_write_begin(page, start, end - start, gfs2_block_map);
 		if (unlikely(ret))
 			return ret;
-		empty_write_end(page, start, end);
+		ret = empty_write_end(page, start, end, mode);
+		if (unlikely(ret))
+			return ret;
 	}
 
 	return 0;
@@ -731,7 +769,7 @@ static int fallocate_chunk(struct inode *inode, loff_t offset, loff_t len,
 
 		if (curr == end)
 			to = end_offset;
-		error = write_empty_blocks(page, from, to);
+		error = write_empty_blocks(page, from, to, mode);
 		if (!error && offset + to > inode->i_size &&
 		    !(mode & FALLOC_FL_KEEP_SIZE)) {
 			i_size_write(inode, offset + to);
diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c
index f07643e21bfa..1019183232fe 100644
--- a/fs/gfs2/glock.c
+++ b/fs/gfs2/glock.c
@@ -160,6 +160,19 @@ static int demote_ok(const struct gfs2_glock *gl)
 }
 
 
+void gfs2_glock_add_to_lru(struct gfs2_glock *gl)
+{
+	spin_lock(&lru_lock);
+
+	if (!list_empty(&gl->gl_lru))
+		list_del_init(&gl->gl_lru);
+	else
+		atomic_inc(&lru_count);
+
+	list_add_tail(&gl->gl_lru, &lru_list);
+	spin_unlock(&lru_lock);
+}
+
 /**
  * __gfs2_glock_schedule_for_reclaim - Add a glock to the reclaim list
  * @gl: the glock
@@ -170,24 +183,8 @@ static int demote_ok(const struct gfs2_glock *gl)
 
 static void __gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
 {
-	if (demote_ok(gl)) {
-		spin_lock(&lru_lock);
-
-		if (!list_empty(&gl->gl_lru))
-			list_del_init(&gl->gl_lru);
-		else
-			atomic_inc(&lru_count);
-
-		list_add_tail(&gl->gl_lru, &lru_list);
-		spin_unlock(&lru_lock);
-	}
-}
-
-void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl)
-{
-	spin_lock(&gl->gl_spin);
-	__gfs2_glock_schedule_for_reclaim(gl);
-	spin_unlock(&gl->gl_spin);
+	if (demote_ok(gl))
+		gfs2_glock_add_to_lru(gl);
 }
 
 /**
diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h
index aea160690e94..6b2f757b9281 100644
--- a/fs/gfs2/glock.h
+++ b/fs/gfs2/glock.h
@@ -225,11 +225,10 @@ static inline int gfs2_glock_nq_init(struct gfs2_glock *gl,
 
 extern void gfs2_glock_cb(struct gfs2_glock *gl, unsigned int state);
 extern void gfs2_glock_complete(struct gfs2_glock *gl, int ret);
-extern void gfs2_reclaim_glock(struct gfs2_sbd *sdp);
 extern void gfs2_gl_hash_clear(struct gfs2_sbd *sdp);
 extern void gfs2_glock_finish_truncate(struct gfs2_inode *ip);
 extern void gfs2_glock_thaw(struct gfs2_sbd *sdp);
-extern void gfs2_glock_schedule_for_reclaim(struct gfs2_glock *gl);
+extern void gfs2_glock_add_to_lru(struct gfs2_glock *gl);
 extern void gfs2_glock_free(struct gfs2_glock *gl);
 
 extern int __init gfs2_glock_init(void);
diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c
index 3754e3cbf02b..25eeb2bcee47 100644
--- a/fs/gfs2/glops.c
+++ b/fs/gfs2/glops.c
@@ -385,6 +385,10 @@ static int trans_go_demote_ok(const struct gfs2_glock *gl)
 static void iopen_go_callback(struct gfs2_glock *gl)
 {
 	struct gfs2_inode *ip = (struct gfs2_inode *)gl->gl_object;
+	struct gfs2_sbd *sdp = gl->gl_sbd;
+
+	if (sdp->sd_vfs->s_flags & MS_RDONLY)
+		return;
 
 	if (gl->gl_demote_state == LM_ST_UNLOCKED &&
 	    gl->gl_state == LM_ST_SHARED && ip) {
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 97d54a28776a..9134dcb89479 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -40,37 +40,61 @@ struct gfs2_inum_range_host {
 	u64 ir_length;
 };
 
+struct gfs2_skip_data {
+	u64 no_addr;
+	int skipped;
+	int non_block;
+};
+
 static int iget_test(struct inode *inode, void *opaque)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
-	u64 *no_addr = opaque;
+	struct gfs2_skip_data *data = opaque;
 
-	if (ip->i_no_addr == *no_addr)
+	if (ip->i_no_addr == data->no_addr) {
+		if (data->non_block &&
+		    inode->i_state & (I_FREEING|I_CLEAR|I_WILL_FREE)) {
+			data->skipped = 1;
+			return 0;
+		}
 		return 1;
-
+	}
 	return 0;
 }
 
 static int iget_set(struct inode *inode, void *opaque)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
-	u64 *no_addr = opaque;
+	struct gfs2_skip_data *data = opaque;
 
-	inode->i_ino = (unsigned long)*no_addr;
-	ip->i_no_addr = *no_addr;
+	if (data->skipped)
+		return -ENOENT;
+	inode->i_ino = (unsigned long)(data->no_addr);
+	ip->i_no_addr = data->no_addr;
 	return 0;
 }
 
 struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr)
 {
 	unsigned long hash = (unsigned long)no_addr;
-	return ilookup5(sb, hash, iget_test, &no_addr);
+	struct gfs2_skip_data data;
+
+	data.no_addr = no_addr;
+	data.skipped = 0;
+	data.non_block = 0;
+	return ilookup5(sb, hash, iget_test, &data);
 }
 
-static struct inode *gfs2_iget(struct super_block *sb, u64 no_addr)
+static struct inode *gfs2_iget(struct super_block *sb, u64 no_addr,
+			       int non_block)
 {
+	struct gfs2_skip_data data;
 	unsigned long hash = (unsigned long)no_addr;
-	return iget5_locked(sb, hash, iget_test, iget_set, &no_addr);
+
+	data.no_addr = no_addr;
+	data.skipped = 0;
+	data.non_block = non_block;
+	return iget5_locked(sb, hash, iget_test, iget_set, &data);
 }
 
 /**
@@ -111,19 +135,20 @@ static void gfs2_set_iop(struct inode *inode)
  * @sb: The super block
  * @no_addr: The inode number
  * @type: The type of the inode
+ * non_block: Can we block on inodes that are being freed?
  *
  * Returns: A VFS inode, or an error
  */
 
 struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type,
-				u64 no_addr, u64 no_formal_ino)
+				u64 no_addr, u64 no_formal_ino, int non_block)
 {
 	struct inode *inode;
 	struct gfs2_inode *ip;
 	struct gfs2_glock *io_gl = NULL;
 	int error;
 
-	inode = gfs2_iget(sb, no_addr);
+	inode = gfs2_iget(sb, no_addr, non_block);
 	ip = GFS2_I(inode);
 
 	if (!inode)
@@ -185,11 +210,12 @@ struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
 {
 	struct super_block *sb = sdp->sd_vfs;
 	struct gfs2_holder i_gh;
-	struct inode *inode;
+	struct inode *inode = NULL;
 	int error;
 
+	/* Must not read in block until block type is verified */
 	error = gfs2_glock_nq_num(sdp, no_addr, &gfs2_inode_glops,
-				  LM_ST_SHARED, LM_FLAG_ANY, &i_gh);
+				  LM_ST_EXCLUSIVE, GL_SKIP, &i_gh);
 	if (error)
 		return ERR_PTR(error);
 
@@ -197,7 +223,7 @@ struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
 	if (error)
 		goto fail;
 
-	inode = gfs2_inode_lookup(sb, DT_UNKNOWN, no_addr, 0);
+	inode = gfs2_inode_lookup(sb, DT_UNKNOWN, no_addr, 0, 1);
 	if (IS_ERR(inode))
 		goto fail;
 
@@ -843,7 +869,7 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
 		goto fail_gunlock2;
 
 	inode = gfs2_inode_lookup(dir->i_sb, IF2DT(mode), inum.no_addr,
-				  inum.no_formal_ino);
+				  inum.no_formal_ino, 0);
 	if (IS_ERR(inode))
 		goto fail_gunlock2;
 
diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h
index 3e00a66e7cbd..099ca305e518 100644
--- a/fs/gfs2/inode.h
+++ b/fs/gfs2/inode.h
@@ -97,7 +97,8 @@ err:
 }
 
 extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, 
-				       u64 no_addr, u64 no_formal_ino);
+				       u64 no_addr, u64 no_formal_ino,
+				       int non_block);
 extern struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr,
 					 u64 *no_formal_ino,
 					 unsigned int blktype);
diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c
index 5b102c1887fd..3ebafa1efad0 100644
--- a/fs/gfs2/log.c
+++ b/fs/gfs2/log.c
@@ -91,6 +91,7 @@ static void gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai)
 __releases(&sdp->sd_ail_lock)
 __acquires(&sdp->sd_ail_lock)
 {
+	struct gfs2_glock *gl = NULL;
 	struct gfs2_bufdata *bd, *s;
 	struct buffer_head *bh;
 	int retry;
@@ -113,19 +114,13 @@ __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);
 			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);
-			}
+			filemap_fdatawrite(gfs2_glock2aspace(gl));
 			spin_lock(&sdp->sd_ail_lock);
 
 			retry = 1;
diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c
index 51d27f00ebb4..611a51d476b2 100644
--- a/fs/gfs2/lops.c
+++ b/fs/gfs2/lops.c
@@ -40,7 +40,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh)
 {
 	struct gfs2_bufdata *bd;
 
-	gfs2_assert_withdraw(sdp, test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags));
+	BUG_ON(!current->journal_info);
 
 	clear_buffer_dirty(bh);
 	if (test_set_buffer_pinned(bh))
@@ -65,6 +65,7 @@ static void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh)
  * @sdp: the filesystem the buffer belongs to
  * @bh: The buffer to unpin
  * @ai:
+ * @flags: The inode dirty flags
  *
  */
 
@@ -73,10 +74,8 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh,
 {
 	struct gfs2_bufdata *bd = bh->b_private;
 
-	gfs2_assert_withdraw(sdp, buffer_uptodate(bh));
-
-	if (!buffer_pinned(bh))
-		gfs2_assert_withdraw(sdp, 0);
+	BUG_ON(!buffer_uptodate(bh));
+	BUG_ON(!buffer_pinned(bh));
 
 	lock_buffer(bh);
 	mark_buffer_dirty(bh);
@@ -95,8 +94,7 @@ static void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh,
 	list_add(&bd->bd_ail_st_list, &ai->ai_ail1_list);
 	spin_unlock(&sdp->sd_ail_lock);
 
-	if (test_and_clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags))
-		gfs2_glock_schedule_for_reclaim(bd->bd_gl);
+	clear_bit(GLF_LFLUSH, &bd->bd_gl->gl_flags);
 	trace_gfs2_pin(bd, 0);
 	unlock_buffer(bh);
 	atomic_dec(&sdp->sd_log_pinned);
diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c
index 42ef24355afb..d3c69eb91c74 100644
--- a/fs/gfs2/ops_fstype.c
+++ b/fs/gfs2/ops_fstype.c
@@ -430,7 +430,7 @@ static int gfs2_lookup_root(struct super_block *sb, struct dentry **dptr,
 	struct dentry *dentry;
 	struct inode *inode;
 
-	inode = gfs2_inode_lookup(sb, DT_DIR, no_addr, 0);
+	inode = gfs2_inode_lookup(sb, DT_DIR, no_addr, 0, 0);
 	if (IS_ERR(inode)) {
 		fs_err(sdp, "can't read in %s inode: %ld\n", name, PTR_ERR(inode));
 		return PTR_ERR(inode);
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index cf930cd9664a..7273ad3c85ba 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -78,10 +78,11 @@ static u32 rgblk_search(struct gfs2_rgrpd *rgd, u32 goal,
 
 static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1,
 			       unsigned char *buf2, unsigned int offset,
-			       unsigned int buflen, u32 block,
+			       struct gfs2_bitmap *bi, u32 block,
 			       unsigned char new_state)
 {
 	unsigned char *byte1, *byte2, *end, cur_state;
+	unsigned int buflen = bi->bi_len;
 	const unsigned int bit = (block % GFS2_NBBY) * GFS2_BIT_SIZE;
 
 	byte1 = buf1 + offset + (block / GFS2_NBBY);
@@ -92,6 +93,16 @@ static inline void gfs2_setbit(struct gfs2_rgrpd *rgd, unsigned char *buf1,
 	cur_state = (*byte1 >> bit) & GFS2_BIT_MASK;
 
 	if (unlikely(!valid_change[new_state * 4 + cur_state])) {
+		printk(KERN_WARNING "GFS2: buf_blk = 0x%llx old_state=%d, "
+		       "new_state=%d\n",
+		       (unsigned long long)block, cur_state, new_state);
+		printk(KERN_WARNING "GFS2: rgrp=0x%llx bi_start=0x%lx\n",
+		       (unsigned long long)rgd->rd_addr,
+		       (unsigned long)bi->bi_start);
+		printk(KERN_WARNING "GFS2: bi_offset=0x%lx bi_len=0x%lx\n",
+		       (unsigned long)bi->bi_offset,
+		       (unsigned long)bi->bi_len);
+		dump_stack();
 		gfs2_consist_rgrpd(rgd);
 		return;
 	}
@@ -381,6 +392,7 @@ static void clear_rgrpdi(struct gfs2_sbd *sdp)
 
 		if (gl) {
 			gl->gl_object = NULL;
+			gfs2_glock_add_to_lru(gl);
 			gfs2_glock_put(gl);
 		}
 
@@ -945,7 +957,7 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip
 		/* rgblk_search can return a block < goal, so we need to
 		   keep it marching forward. */
 		no_addr = block + rgd->rd_data0;
-		goal++;
+		goal = max(block + 1, goal + 1);
 		if (*last_unlinked != NO_BLOCK && no_addr <= *last_unlinked)
 			continue;
 		if (no_addr == skip)
@@ -971,7 +983,7 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip
 			found++;
 
 		/* Limit reclaim to sensible number of tasks */
-		if (found > 2*NR_CPUS)
+		if (found > NR_CPUS)
 			return;
 	}
 
@@ -1365,7 +1377,7 @@ skip:
 
 	gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1);
 	gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset,
-		    bi->bi_len, blk, new_state);
+		    bi, blk, new_state);
 	goal = blk;
 	while (*n < elen) {
 		goal++;
@@ -1375,7 +1387,7 @@ skip:
 		    GFS2_BLKST_FREE)
 			break;
 		gfs2_setbit(rgd, bi->bi_bh->b_data, bi->bi_clone, bi->bi_offset,
-			    bi->bi_len, goal, new_state);
+			    bi, goal, new_state);
 		(*n)++;
 	}
 out:
@@ -1432,7 +1444,7 @@ static struct gfs2_rgrpd *rgblk_free(struct gfs2_sbd *sdp, u64 bstart,
 		}
 		gfs2_trans_add_bh(rgd->rd_gl, bi->bi_bh, 1);
 		gfs2_setbit(rgd, bi->bi_bh->b_data, NULL, bi->bi_offset,
-			    bi->bi_len, buf_blk, new_state);
+			    bi, buf_blk, new_state);
 	}
 
 	return rgd;
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index a4e23d68a398..b62c8427672c 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -704,7 +704,7 @@ void gfs2_unfreeze_fs(struct gfs2_sbd *sdp)
 /**
  * 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 +713,16 @@ 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 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 +746,11 @@ do_unlock:
 do_flush:
 	if (wbc->sync_mode == WB_SYNC_ALL)
 		gfs2_log_flush(GFS2_SB(inode), ip->i_gl);
+	filemap_fdatawrite(metamapping);
+	if (!ret && (wbc->sync_mode == WB_SYNC_ALL))
+		ret = filemap_fdatawait(metamapping);
+	if (ret)
+		mark_inode_dirty_sync(inode);
 	return ret;
 }
 
@@ -874,8 +880,9 @@ restart:
 
 static int gfs2_sync_fs(struct super_block *sb, int wait)
 {
-	if (wait && sb->s_fs_info)
-		gfs2_log_flush(sb->s_fs_info, NULL);
+	struct gfs2_sbd *sdp = sb->s_fs_info;
+	if (wait && sdp)
+		gfs2_log_flush(sdp, NULL);
 	return 0;
 }
 
@@ -1318,15 +1325,17 @@ static int gfs2_show_options(struct seq_file *s, struct vfsmount *mnt)
 
 static void gfs2_evict_inode(struct inode *inode)
 {
-	struct gfs2_sbd *sdp = inode->i_sb->s_fs_info;
+	struct super_block *sb = inode->i_sb;
+	struct gfs2_sbd *sdp = sb->s_fs_info;
 	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_holder gh;
 	int error;
 
-	if (inode->i_nlink)
+	if (inode->i_nlink || (sb->s_flags & MS_RDONLY))
 		goto out;
 
-	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh);
+	/* Must not read inode block until block type has been verified */
+	error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_SKIP, &gh);
 	if (unlikely(error)) {
 		gfs2_glock_dq_uninit(&ip->i_iopen_gh);
 		goto out;
@@ -1336,6 +1345,12 @@ static void gfs2_evict_inode(struct inode *inode)
 	if (error)
 		goto out_truncate;
 
+	if (test_bit(GIF_INVALID, &ip->i_flags)) {
+		error = gfs2_inode_refresh(ip);
+		if (error)
+			goto out_truncate;
+	}
+
 	ip->i_iopen_gh.gh_flags |= GL_NOCACHE;
 	gfs2_glock_dq_wait(&ip->i_iopen_gh);
 	gfs2_holder_reinit(LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB | GL_NOCACHE, &ip->i_iopen_gh);
@@ -1386,6 +1401,7 @@ out:
 	end_writeback(inode);
 
 	ip->i_gl->gl_object = NULL;
+	gfs2_glock_add_to_lru(ip->i_gl);
 	gfs2_glock_put(ip->i_gl);
 	ip->i_gl = NULL;
 	if (ip->i_iopen_gh.gh_gl) {