Merge branch 'master' into upstream

Conflicts:
	drivers/hid/hid-ids.h

436 files changed, 15591 insertions, 11857 deletions

diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 18f74ec4dce9..9d03d1ebca6f 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -1001,44 +1001,6 @@ done:
 }
 
 /**
- * v9fs_vfs_readlink - read a symlink's location
- * @dentry: dentry for symlink
- * @buffer: buffer to load symlink location into
- * @buflen: length of buffer
- *
- */
-
-static int v9fs_vfs_readlink(struct dentry *dentry, char __user * buffer,
-			     int buflen)
-{
-	int retval;
-	int ret;
-	char *link = __getname();
-
-	if (unlikely(!link))
-		return -ENOMEM;
-
-	if (buflen > PATH_MAX)
-		buflen = PATH_MAX;
-
-	P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_name.name,
-									dentry);
-
-	retval = v9fs_readlink(dentry, link, buflen);
-
-	if (retval > 0) {
-		if ((ret = copy_to_user(buffer, link, retval)) != 0) {
-			P9_DPRINTK(P9_DEBUG_ERROR,
-					"problem copying to user: %d\n", ret);
-			retval = ret;
-		}
-	}
-
-	__putname(link);
-	return retval;
-}
-
-/**
  * v9fs_vfs_follow_link - follow a symlink path
  * @dentry: dentry for symlink
  * @nd: nameidata
@@ -1230,7 +1192,6 @@ static const struct inode_operations v9fs_dir_inode_operations_ext = {
 	.rmdir = v9fs_vfs_rmdir,
 	.mknod = v9fs_vfs_mknod,
 	.rename = v9fs_vfs_rename,
-	.readlink = v9fs_vfs_readlink,
 	.getattr = v9fs_vfs_getattr,
 	.setattr = v9fs_vfs_setattr,
 };
@@ -1253,7 +1214,7 @@ static const struct inode_operations v9fs_file_inode_operations = {
 };
 
 static const struct inode_operations v9fs_symlink_inode_operations = {
-	.readlink = v9fs_vfs_readlink,
+	.readlink = generic_readlink,
 	.follow_link = v9fs_vfs_follow_link,
 	.put_link = v9fs_vfs_put_link,
 	.getattr = v9fs_vfs_getattr,
diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c
index 14a86448572c..69357c0d9899 100644
--- a/fs/9p/vfs_super.c
+++ b/fs/9p/vfs_super.c
@@ -188,7 +188,8 @@ static void v9fs_kill_super(struct super_block *s)
 
 	P9_DPRINTK(P9_DEBUG_VFS, " %p\n", s);
 
-	v9fs_dentry_release(s->s_root);	/* clunk root */
+	if (s->s_root)
+		v9fs_dentry_release(s->s_root);	/* clunk root */
 
 	kill_anon_super(s);
 
diff --git a/fs/affs/affs.h b/fs/affs/affs.h
index e511dc621a2e..0e40caaba456 100644
--- a/fs/affs/affs.h
+++ b/fs/affs/affs.h
@@ -106,8 +106,8 @@ struct affs_sb_info {
 	u32 s_last_bmap;
 	struct buffer_head *s_bmap_bh;
 	char *s_prefix;			/* Prefix for volumes and assigns. */
-	int s_prefix_len;		/* Length of prefix. */
 	char s_volume[32];		/* Volume prefix for absolute symlinks. */
+	spinlock_t symlink_lock;	/* protects the previous two */
 };
 
 #define SF_INTL		0x0001		/* International filesystem. */
diff --git a/fs/affs/namei.c b/fs/affs/namei.c
index 960d336ec694..d70bbbac6b7b 100644
--- a/fs/affs/namei.c
+++ b/fs/affs/namei.c
@@ -341,10 +341,13 @@ affs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 	p  = (char *)AFFS_HEAD(bh)->table;
 	lc = '/';
 	if (*symname == '/') {
+		struct affs_sb_info *sbi = AFFS_SB(sb);
 		while (*symname == '/')
 			symname++;
-		while (AFFS_SB(sb)->s_volume[i])	/* Cannot overflow */
-			*p++ = AFFS_SB(sb)->s_volume[i++];
+		spin_lock(&sbi->symlink_lock);
+		while (sbi->s_volume[i])	/* Cannot overflow */
+			*p++ = sbi->s_volume[i++];
+		spin_unlock(&sbi->symlink_lock);
 	}
 	while (i < maxlen && (c = *symname++)) {
 		if (c == '.' && lc == '/' && *symname == '.' && symname[1] == '/') {
diff --git a/fs/affs/super.c b/fs/affs/super.c
index 104fdcb3a7fc..d41e9673cd97 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -203,7 +203,7 @@ parse_options(char *options, uid_t *uid, gid_t *gid, int *mode, int *reserved, s
 		switch (token) {
 		case Opt_bs:
 			if (match_int(&args[0], &n))
-				return -EINVAL;
+				return 0;
 			if (n != 512 && n != 1024 && n != 2048
 			    && n != 4096) {
 				printk ("AFFS: Invalid blocksize (512, 1024, 2048, 4096 allowed)\n");
@@ -213,7 +213,7 @@ parse_options(char *options, uid_t *uid, gid_t *gid, int *mode, int *reserved, s
 			break;
 		case Opt_mode:
 			if (match_octal(&args[0], &option))
-				return 1;
+				return 0;
 			*mode = option & 0777;
 			*mount_opts |= SF_SETMODE;
 			break;
@@ -221,8 +221,6 @@ parse_options(char *options, uid_t *uid, gid_t *gid, int *mode, int *reserved, s
 			*mount_opts |= SF_MUFS;
 			break;
 		case Opt_prefix:
-			/* Free any previous prefix */
-			kfree(*prefix);
 			*prefix = match_strdup(&args[0]);
 			if (!*prefix)
 				return 0;
@@ -233,21 +231,21 @@ parse_options(char *options, uid_t *uid, gid_t *gid, int *mode, int *reserved, s
 			break;
 		case Opt_reserved:
 			if (match_int(&args[0], reserved))
-				return 1;
+				return 0;
 			break;
 		case Opt_root:
 			if (match_int(&args[0], root))
-				return 1;
+				return 0;
 			break;
 		case Opt_setgid:
 			if (match_int(&args[0], &option))
-				return 1;
+				return 0;
 			*gid = option;
 			*mount_opts |= SF_SETGID;
 			break;
 		case Opt_setuid:
 			if (match_int(&args[0], &option))
-				return -EINVAL;
+				return 0;
 			*uid = option;
 			*mount_opts |= SF_SETUID;
 			break;
@@ -311,11 +309,14 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent)
 		return -ENOMEM;
 	sb->s_fs_info = sbi;
 	mutex_init(&sbi->s_bmlock);
+	spin_lock_init(&sbi->symlink_lock);
 
 	if (!parse_options(data,&uid,&gid,&i,&reserved,&root_block,
 				&blocksize,&sbi->s_prefix,
 				sbi->s_volume, &mount_flags)) {
 		printk(KERN_ERR "AFFS: Error parsing options\n");
+		kfree(sbi->s_prefix);
+		kfree(sbi);
 		return -EINVAL;
 	}
 	/* N.B. after this point s_prefix must be released */
@@ -516,14 +517,18 @@ affs_remount(struct super_block *sb, int *flags, char *data)
 	unsigned long		 mount_flags;
 	int			 res = 0;
 	char			*new_opts = kstrdup(data, GFP_KERNEL);
+	char			 volume[32];
+	char			*prefix = NULL;
 
 	pr_debug("AFFS: remount(flags=0x%x,opts=\"%s\")\n",*flags,data);
 
 	*flags |= MS_NODIRATIME;
 
+	memcpy(volume, sbi->s_volume, 32);
 	if (!parse_options(data, &uid, &gid, &mode, &reserved, &root_block,
-			   &blocksize, &sbi->s_prefix, sbi->s_volume,
+			   &blocksize, &prefix, volume,
 			   &mount_flags)) {
+		kfree(prefix);
 		kfree(new_opts);
 		return -EINVAL;
 	}
@@ -534,6 +539,14 @@ affs_remount(struct super_block *sb, int *flags, char *data)
 	sbi->s_mode  = mode;
 	sbi->s_uid   = uid;
 	sbi->s_gid   = gid;
+	/* protect against readers */
+	spin_lock(&sbi->symlink_lock);
+	if (prefix) {
+		kfree(sbi->s_prefix);
+		sbi->s_prefix = prefix;
+	}
+	memcpy(sbi->s_volume, volume, 32);
+	spin_unlock(&sbi->symlink_lock);
 
 	if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY)) {
 		unlock_kernel();
diff --git a/fs/affs/symlink.c b/fs/affs/symlink.c
index 41782539c907..ee00f08c4f53 100644
--- a/fs/affs/symlink.c
+++ b/fs/affs/symlink.c
@@ -20,7 +20,6 @@ static int affs_symlink_readpage(struct file *file, struct page *page)
 	int			 i, j;
 	char			 c;
 	char			 lc;
-	char			*pf;
 
 	pr_debug("AFFS: follow_link(ino=%lu)\n",inode->i_ino);
 
@@ -32,11 +31,15 @@ static int affs_symlink_readpage(struct file *file, struct page *page)
 	j  = 0;
 	lf = (struct slink_front *)bh->b_data;
 	lc = 0;
-	pf = AFFS_SB(inode->i_sb)->s_prefix ? AFFS_SB(inode->i_sb)->s_prefix : "/";
 
 	if (strchr(lf->symname,':')) {	/* Handle assign or volume name */
+		struct affs_sb_info *sbi = AFFS_SB(inode->i_sb);
+		char *pf;
+		spin_lock(&sbi->symlink_lock);
+		pf = sbi->s_prefix ? sbi->s_prefix : "/";
 		while (i < 1023 && (c = pf[i]))
 			link[i++] = c;
+		spin_unlock(&sbi->symlink_lock);
 		while (i < 1023 && lf->symname[j] != ':')
 			link[i++] = lf->symname[j++];
 		if (i < 1023)
diff --git a/fs/afs/write.c b/fs/afs/write.c
index c63a3c8beb73..5e15a21dbf9f 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -671,7 +671,6 @@ ssize_t afs_file_write(struct kiocb *iocb, const struct iovec *iov,
 	struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode);
 	ssize_t result;
 	size_t count = iov_length(iov, nr_segs);
-	int ret;
 
 	_enter("{%x.%u},{%zu},%lu,",
 	       vnode->fid.vid, vnode->fid.vnode, count, nr_segs);
@@ -691,13 +690,6 @@ ssize_t afs_file_write(struct kiocb *iocb, const struct iovec *iov,
 		return result;
 	}
 
-	/* return error values for O_SYNC and IS_SYNC() */
-	if (IS_SYNC(&vnode->vfs_inode) || iocb->ki_filp->f_flags & O_SYNC) {
-		ret = afs_fsync(iocb->ki_filp, dentry, 1);
-		if (ret < 0)
-			result = ret;
-	}
-
 	_leave(" = %zd", result);
 	return result;
 }
diff --git a/fs/aio.c b/fs/aio.c
index c30dfc006108..1cf12b3dd83a 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -711,10 +711,8 @@ static ssize_t aio_run_iocb(struct kiocb *iocb)
 	 */
 	ret = retry(iocb);
 
-	if (ret != -EIOCBRETRY && ret != -EIOCBQUEUED) {
-		BUG_ON(!list_empty(&iocb->ki_wait.task_list));
+	if (ret != -EIOCBRETRY && ret != -EIOCBQUEUED)
 		aio_complete(iocb, ret, 0);
-	}
 out:
 	spin_lock_irq(&ctx->ctx_lock);
 
@@ -866,13 +864,6 @@ static void try_queue_kicked_iocb(struct kiocb *iocb)
 	unsigned long flags;
 	int run = 0;
 
-	/* We're supposed to be the only path putting the iocb back on the run
-	 * list.  If we find that the iocb is *back* on a wait queue already
-	 * than retry has happened before we could queue the iocb.  This also
-	 * means that the retry could have completed and freed our iocb, no
-	 * good. */
-	BUG_ON((!list_empty(&iocb->ki_wait.task_list)));
-
 	spin_lock_irqsave(&ctx->ctx_lock, flags);
 	/* set this inside the lock so that we can't race with aio_run_iocb()
 	 * testing it and putting the iocb on the run list under the lock */
@@ -886,7 +877,7 @@ static void try_queue_kicked_iocb(struct kiocb *iocb)
 /*
  * kick_iocb:
  *      Called typically from a wait queue callback context
- *      (aio_wake_function) to trigger a retry of the iocb.
+ *      to trigger a retry of the iocb.
  *      The retry is usually executed by aio workqueue
  *      threads (See aio_kick_handler).
  */
@@ -1520,31 +1511,6 @@ static ssize_t aio_setup_iocb(struct kiocb *kiocb)
 	return 0;
 }
 
-/*
- * aio_wake_function:
- * 	wait queue callback function for aio notification,
- * 	Simply triggers a retry of the operation via kick_iocb.
- *
- * 	This callback is specified in the wait queue entry in
- *	a kiocb.
- *
- * Note:
- * This routine is executed with the wait queue lock held.
- * Since kick_iocb acquires iocb->ctx->ctx_lock, it nests
- * the ioctx lock inside the wait queue lock. This is safe
- * because this callback isn't used for wait queues which
- * are nested inside ioctx lock (i.e. ctx->wait)
- */
-static int aio_wake_function(wait_queue_t *wait, unsigned mode,
-			     int sync, void *key)
-{
-	struct kiocb *iocb = container_of(wait, struct kiocb, ki_wait);
-
-	list_del_init(&wait->task_list);
-	kick_iocb(iocb);
-	return 1;
-}
-
 static void aio_batch_add(struct address_space *mapping,
 			  struct hlist_head *batch_hash)
 {
@@ -1642,8 +1608,6 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb,
 	req->ki_buf = (char __user *)(unsigned long)iocb->aio_buf;
 	req->ki_left = req->ki_nbytes = iocb->aio_nbytes;
 	req->ki_opcode = iocb->aio_lio_opcode;
-	init_waitqueue_func_entry(&req->ki_wait, aio_wake_function);
-	INIT_LIST_HEAD(&req->ki_wait.task_list);
 
 	ret = aio_setup_iocb(req);
 
diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c
index 2ca7a7cafdbf..9f0bf13291e5 100644
--- a/fs/anon_inodes.c
+++ b/fs/anon_inodes.c
@@ -35,14 +35,13 @@ static int anon_inodefs_get_sb(struct file_system_type *fs_type, int flags,
 			     mnt);
 }
 
-static int anon_inodefs_delete_dentry(struct dentry *dentry)
+/*
+ * anon_inodefs_dname() is called from d_path().
+ */
+static char *anon_inodefs_dname(struct dentry *dentry, char *buffer, int buflen)
 {
-	/*
-	 * We faked vfs to believe the dentry was hashed when we created it.
-	 * Now we restore the flag so that dput() will work correctly.
-	 */
-	dentry->d_flags |= DCACHE_UNHASHED;
-	return 1;
+	return dynamic_dname(dentry, buffer, buflen, "anon_inode:%s",
+				dentry->d_name.name);
 }
 
 static struct file_system_type anon_inode_fs_type = {
@@ -51,7 +50,7 @@ static struct file_system_type anon_inode_fs_type = {
 	.kill_sb	= kill_anon_super,
 };
 static const struct dentry_operations anon_inodefs_dentry_operations = {
-	.d_delete	= anon_inodefs_delete_dentry,
+	.d_dname	= anon_inodefs_dname,
 };
 
 /*
@@ -88,7 +87,7 @@ struct file *anon_inode_getfile(const char *name,
 				void *priv, int flags)
 {
 	struct qstr this;
-	struct dentry *dentry;
+	struct path path;
 	struct file *file;
 	int error;
 
@@ -106,10 +105,11 @@ struct file *anon_inode_getfile(const char *name,
 	this.name = name;
 	this.len = strlen(name);
 	this.hash = 0;
-	dentry = d_alloc(anon_inode_mnt->mnt_sb->s_root, &this);
-	if (!dentry)
+	path.dentry = d_alloc(anon_inode_mnt->mnt_sb->s_root, &this);
+	if (!path.dentry)
 		goto err_module;
 
+	path.mnt = mntget(anon_inode_mnt);
 	/*
 	 * We know the anon_inode inode count is always greater than zero,
 	 * so we can avoid doing an igrab() and we can use an open-coded
@@ -117,27 +117,24 @@ struct file *anon_inode_getfile(const char *name,
 	 */
 	atomic_inc(&anon_inode_inode->i_count);
 
-	dentry->d_op = &anon_inodefs_dentry_operations;
-	/* Do not publish this dentry inside the global dentry hash table */
-	dentry->d_flags &= ~DCACHE_UNHASHED;
-	d_instantiate(dentry, anon_inode_inode);
+	path.dentry->d_op = &anon_inodefs_dentry_operations;
+	d_instantiate(path.dentry, anon_inode_inode);
 
 	error = -ENFILE;
-	file = alloc_file(anon_inode_mnt, dentry,
-			  FMODE_READ | FMODE_WRITE, fops);
+	file = alloc_file(&path, OPEN_FMODE(flags), fops);
 	if (!file)
 		goto err_dput;
 	file->f_mapping = anon_inode_inode->i_mapping;
 
 	file->f_pos = 0;
-	file->f_flags = O_RDWR | (flags & O_NONBLOCK);
+	file->f_flags = flags & (O_ACCMODE | O_NONBLOCK);
 	file->f_version = 0;
 	file->private_data = priv;
 
 	return file;
 
 err_dput:
-	dput(dentry);
+	path_put(&path);
 err_module:
 	module_put(fops->owner);
 	return ERR_PTR(error);
diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 8f7cdde41733..0118d67221b2 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -60,6 +60,11 @@ do {							\
 		current->pid, __func__, ##args);	\
 } while (0)
 
+struct rehash_entry {
+	struct task_struct *task;
+	struct list_head list;
+};
+
 /* Unified info structure.  This is pointed to by both the dentry and
    inode structures.  Each file in the filesystem has an instance of this
    structure.  It holds a reference to the dentry, so dentries are never
@@ -75,6 +80,9 @@ struct autofs_info {
 	struct completion expire_complete;
 
 	struct list_head active;
+	int active_count;
+	struct list_head rehash_list;
+
 	struct list_head expiring;
 
 	struct autofs_sb_info *sbi;
@@ -95,6 +103,8 @@ struct autofs_info {
 
 #define AUTOFS_INF_EXPIRING	(1<<0) /* dentry is in the process of expiring */
 #define AUTOFS_INF_MOUNTPOINT	(1<<1) /* mountpoint status for direct expire */
+#define AUTOFS_INF_PENDING	(1<<2) /* dentry pending mount */
+#define AUTOFS_INF_REHASH	(1<<3) /* dentry in transit to ->lookup() */
 
 struct autofs_wait_queue {
 	wait_queue_head_t queue;
@@ -161,7 +171,7 @@ static inline int autofs4_ispending(struct dentry *dentry)
 {
 	struct autofs_info *inf = autofs4_dentry_ino(dentry);
 
-	if (dentry->d_flags & DCACHE_AUTOFS_PENDING)
+	if (inf->flags & AUTOFS_INF_PENDING)
 		return 1;
 
 	if (inf->flags & AUTOFS_INF_EXPIRING)
@@ -264,5 +274,31 @@ out:
 	return ret;
 }
 
+static inline void autofs4_add_expiring(struct dentry *dentry)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	if (ino) {
+		spin_lock(&sbi->lookup_lock);
+		if (list_empty(&ino->expiring))
+			list_add(&ino->expiring, &sbi->expiring_list);
+		spin_unlock(&sbi->lookup_lock);
+	}
+	return;
+}
+
+static inline void autofs4_del_expiring(struct dentry *dentry)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	if (ino) {
+		spin_lock(&sbi->lookup_lock);
+		if (!list_empty(&ino->expiring))
+			list_del_init(&ino->expiring);
+		spin_unlock(&sbi->lookup_lock);
+	}
+	return;
+}
+
 void autofs4_dentry_release(struct dentry *);
 extern void autofs4_kill_sb(struct super_block *);
diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index 3da18d453488..74bc9aa6df31 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -27,7 +27,7 @@ static inline int autofs4_can_expire(struct dentry *dentry,
 		return 0;
 
 	/* No point expiring a pending mount */
-	if (dentry->d_flags & DCACHE_AUTOFS_PENDING)
+	if (ino->flags & AUTOFS_INF_PENDING)
 		return 0;
 
 	if (!do_now) {
@@ -279,6 +279,7 @@ struct dentry *autofs4_expire_direct(struct super_block *sb,
 			root->d_mounted--;
 		}
 		ino->flags |= AUTOFS_INF_EXPIRING;
+		autofs4_add_expiring(root);
 		init_completion(&ino->expire_complete);
 		spin_unlock(&sbi->fs_lock);
 		return root;
@@ -406,6 +407,7 @@ found:
 		expired, (int)expired->d_name.len, expired->d_name.name);
 	ino = autofs4_dentry_ino(expired);
 	ino->flags |= AUTOFS_INF_EXPIRING;
+	autofs4_add_expiring(expired);
 	init_completion(&ino->expire_complete);
 	spin_unlock(&sbi->fs_lock);
 	spin_lock(&dcache_lock);
@@ -433,7 +435,7 @@ int autofs4_expire_wait(struct dentry *dentry)
 
 		DPRINTK("expire done status=%d", status);
 
-		if (d_unhashed(dentry))
+		if (d_unhashed(dentry) && IS_DEADDIR(dentry->d_inode))
 			return -EAGAIN;
 
 		return status;
@@ -473,6 +475,7 @@ int autofs4_expire_run(struct super_block *sb,
 	spin_lock(&sbi->fs_lock);
 	ino = autofs4_dentry_ino(dentry);
 	ino->flags &= ~AUTOFS_INF_EXPIRING;
+	autofs4_del_expiring(dentry);
 	complete_all(&ino->expire_complete);
 	spin_unlock(&sbi->fs_lock);
 
@@ -503,6 +506,7 @@ int autofs4_do_expire_multi(struct super_block *sb, struct vfsmount *mnt,
 			ino->flags &= ~AUTOFS_INF_MOUNTPOINT;
 		}
 		ino->flags &= ~AUTOFS_INF_EXPIRING;
+		autofs4_del_expiring(dentry);
 		complete_all(&ino->expire_complete);
 		spin_unlock(&sbi->fs_lock);
 		dput(dentry);
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 69c8142da838..d0a3de247458 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -49,6 +49,8 @@ struct autofs_info *autofs4_init_ino(struct autofs_info *ino,
 		ino->dentry = NULL;
 		ino->size = 0;
 		INIT_LIST_HEAD(&ino->active);
+		INIT_LIST_HEAD(&ino->rehash_list);
+		ino->active_count = 0;
 		INIT_LIST_HEAD(&ino->expiring);
 		atomic_set(&ino->count, 0);
 	}
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index b96a3c57359d..30cc9ddf4b70 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -72,6 +72,139 @@ const struct inode_operations autofs4_dir_inode_operations = {
 	.rmdir		= autofs4_dir_rmdir,
 };
 
+static void autofs4_add_active(struct dentry *dentry)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	if (ino) {
+		spin_lock(&sbi->lookup_lock);
+		if (!ino->active_count) {
+			if (list_empty(&ino->active))
+				list_add(&ino->active, &sbi->active_list);
+		}
+		ino->active_count++;
+		spin_unlock(&sbi->lookup_lock);
+	}
+	return;
+}
+
+static void autofs4_del_active(struct dentry *dentry)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	if (ino) {
+		spin_lock(&sbi->lookup_lock);
+		ino->active_count--;
+		if (!ino->active_count) {
+			if (!list_empty(&ino->active))
+				list_del_init(&ino->active);
+		}
+		spin_unlock(&sbi->lookup_lock);
+	}
+	return;
+}
+
+static void autofs4_add_rehash_entry(struct autofs_info *ino,
+				     struct rehash_entry *entry)
+{
+	entry->task = current;
+	INIT_LIST_HEAD(&entry->list);
+	list_add(&entry->list, &ino->rehash_list);
+	return;
+}
+
+static void autofs4_remove_rehash_entry(struct autofs_info *ino)
+{
+	struct list_head *head = &ino->rehash_list;
+	struct rehash_entry *entry;
+	list_for_each_entry(entry, head, list) {
+		if (entry->task == current) {
+			list_del(&entry->list);
+			kfree(entry);
+			break;
+		}
+	}
+	return;
+}
+
+static void autofs4_remove_rehash_entrys(struct autofs_info *ino)
+{
+	struct autofs_sb_info *sbi = ino->sbi;
+	struct rehash_entry *entry, *next;
+	struct list_head *head;
+
+	spin_lock(&sbi->fs_lock);
+	spin_lock(&sbi->lookup_lock);
+	if (!(ino->flags & AUTOFS_INF_REHASH)) {
+		spin_unlock(&sbi->lookup_lock);
+		spin_unlock(&sbi->fs_lock);
+		return;
+	}
+	ino->flags &= ~AUTOFS_INF_REHASH;
+	head = &ino->rehash_list;
+	list_for_each_entry_safe(entry, next, head, list) {
+		list_del(&entry->list);
+		kfree(entry);
+	}
+	spin_unlock(&sbi->lookup_lock);
+	spin_unlock(&sbi->fs_lock);
+	dput(ino->dentry);
+
+	return;
+}
+
+static void autofs4_revalidate_drop(struct dentry *dentry,
+				    struct rehash_entry *entry)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	/*
+	 * Add to the active list so we can pick this up in
+	 * ->lookup(). Also add an entry to a rehash list so
+	 * we know when there are no dentrys in flight so we
+	 * know when we can rehash the dentry.
+	 */
+	spin_lock(&sbi->lookup_lock);
+	if (list_empty(&ino->active))
+		list_add(&ino->active, &sbi->active_list);
+	autofs4_add_rehash_entry(ino, entry);
+	spin_unlock(&sbi->lookup_lock);
+	if (!(ino->flags & AUTOFS_INF_REHASH)) {
+		ino->flags |= AUTOFS_INF_REHASH;
+		dget(dentry);
+		spin_lock(&dentry->d_lock);
+		__d_drop(dentry);
+		spin_unlock(&dentry->d_lock);
+	}
+	return;
+}
+
+static void autofs4_revalidate_rehash(struct dentry *dentry)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	if (ino->flags & AUTOFS_INF_REHASH) {
+		spin_lock(&sbi->lookup_lock);
+		autofs4_remove_rehash_entry(ino);
+		if (list_empty(&ino->rehash_list)) {
+			spin_unlock(&sbi->lookup_lock);
+			ino->flags &= ~AUTOFS_INF_REHASH;
+			d_rehash(dentry);
+			dput(ino->dentry);
+		} else
+			spin_unlock(&sbi->lookup_lock);
+	}
+	return;
+}
+
+static unsigned int autofs4_need_mount(unsigned int flags)
+{
+	unsigned int res = 0;
+	if (flags & (TRIGGER_FLAGS | TRIGGER_INTENTS))
+		res = 1;
+	return res;
+}
+
 static int autofs4_dir_open(struct inode *inode, struct file *file)
 {
 	struct dentry *dentry = file->f_path.dentry;
@@ -93,7 +226,7 @@ static int autofs4_dir_open(struct inode *inode, struct file *file)
 	 * it.
 	 */
 	spin_lock(&dcache_lock);
-	if (!d_mountpoint(dentry) && __simple_empty(dentry)) {
+	if (!d_mountpoint(dentry) && list_empty(&dentry->d_subdirs)) {
 		spin_unlock(&dcache_lock);
 		return -ENOENT;
 	}
@@ -103,7 +236,7 @@ out:
 	return dcache_dir_open(inode, file);
 }
 
-static int try_to_fill_dentry(struct dentry *dentry, int flags)
+static int try_to_fill_dentry(struct dentry *dentry)
 {
 	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
 	struct autofs_info *ino = autofs4_dentry_ino(dentry);
@@ -116,55 +249,17 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 	 * Wait for a pending mount, triggering one if there
 	 * isn't one already
 	 */
-	if (dentry->d_inode == NULL) {
-		DPRINTK("waiting for mount name=%.*s",
-			 dentry->d_name.len, dentry->d_name.name);
-
-		status = autofs4_wait(sbi, dentry, NFY_MOUNT);
-
-		DPRINTK("mount done status=%d", status);
-
-		/* Turn this into a real negative dentry? */
-		if (status == -ENOENT) {
-			spin_lock(&dentry->d_lock);
-			dentry->d_flags &= ~DCACHE_AUTOFS_PENDING;
-			spin_unlock(&dentry->d_lock);
-			return status;
-		} else if (status) {
-			/* Return a negative dentry, but leave it "pending" */
-			return status;
-		}
-	/* Trigger mount for path component or follow link */
-	} else if (dentry->d_flags & DCACHE_AUTOFS_PENDING ||
-			flags & (TRIGGER_FLAGS | TRIGGER_INTENTS) ||
-			current->link_count) {
-		DPRINTK("waiting for mount name=%.*s",
-			dentry->d_name.len, dentry->d_name.name);
-
-		spin_lock(&dentry->d_lock);
-		dentry->d_flags |= DCACHE_AUTOFS_PENDING;
-		spin_unlock(&dentry->d_lock);
-		status = autofs4_wait(sbi, dentry, NFY_MOUNT);
+	DPRINTK("waiting for mount name=%.*s",
+		 dentry->d_name.len, dentry->d_name.name);
 
-		DPRINTK("mount done status=%d", status);
+	status = autofs4_wait(sbi, dentry, NFY_MOUNT);
 
-		if (status) {
-			spin_lock(&dentry->d_lock);
-			dentry->d_flags &= ~DCACHE_AUTOFS_PENDING;
-			spin_unlock(&dentry->d_lock);
-			return status;
-		}
-	}
-
-	/* Initialize expiry counter after successful mount */
-	if (ino)
-		ino->last_used = jiffies;
+	DPRINTK("mount done status=%d", status);
 
-	spin_lock(&dentry->d_lock);
-	dentry->d_flags &= ~DCACHE_AUTOFS_PENDING;
-	spin_unlock(&dentry->d_lock);
+	/* Update expiry counter */
+	ino->last_used = jiffies;
 
-	return 0;
+	return status;
 }
 
 /* For autofs direct mounts the follow link triggers the mount */
@@ -202,27 +297,39 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 	autofs4_expire_wait(dentry);
 
 	/* We trigger a mount for almost all flags */
-	lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS);
-	if (!(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING))
+	lookup_type = autofs4_need_mount(nd->flags);
+	spin_lock(&sbi->fs_lock);
+	spin_lock(&dcache_lock);
+	if (!(lookup_type || ino->flags & AUTOFS_INF_PENDING)) {
+		spin_unlock(&dcache_lock);
+		spin_unlock(&sbi->fs_lock);
 		goto follow;
+	}
 
 	/*
 	 * If the dentry contains directories then it is an autofs
 	 * multi-mount with no root mount offset. So don't try to
 	 * mount it again.
 	 */
-	spin_lock(&dcache_lock);
-	if (dentry->d_flags & DCACHE_AUTOFS_PENDING ||
-	    (!d_mountpoint(dentry) && __simple_empty(dentry))) {
+	if (ino->flags & AUTOFS_INF_PENDING ||
+	    (!d_mountpoint(dentry) && list_empty(&dentry->d_subdirs))) {
+		ino->flags |= AUTOFS_INF_PENDING;
 		spin_unlock(&dcache_lock);
+		spin_unlock(&sbi->fs_lock);
+
+		status = try_to_fill_dentry(dentry);
+
+		spin_lock(&sbi->fs_lock);
+		ino->flags &= ~AUTOFS_INF_PENDING;
+		spin_unlock(&sbi->fs_lock);
 
-		status = try_to_fill_dentry(dentry, 0);
 		if (status)
 			goto out_error;
 
 		goto follow;
 	}
 	spin_unlock(&dcache_lock);
+	spin_unlock(&sbi->fs_lock);
 follow:
 	/*
 	 * If there is no root mount it must be an autofs
@@ -254,18 +361,47 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
 	struct inode *dir = dentry->d_parent->d_inode;
 	struct autofs_sb_info *sbi = autofs4_sbi(dir->i_sb);
-	int oz_mode = autofs4_oz_mode(sbi);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	struct rehash_entry *entry;
 	int flags = nd ? nd->flags : 0;
-	int status = 1;
+	unsigned int mutex_aquired;
+
+	DPRINTK("name = %.*s oz_mode = %d",
+		dentry->d_name.len, dentry->d_name.name, oz_mode);
+
+	/* Daemon never causes a mount to trigger */
+	if (autofs4_oz_mode(sbi))
+		return 1;
+
+	entry = kmalloc(sizeof(struct rehash_entry), GFP_KERNEL);
+	if (!entry)
+		return -ENOMEM;
+
+	mutex_aquired = mutex_trylock(&dir->i_mutex);
 
-	/* Pending dentry */
 	spin_lock(&sbi->fs_lock);
+	spin_lock(&dcache_lock);
+	/* Pending dentry */
 	if (autofs4_ispending(dentry)) {
-		/* The daemon never causes a mount to trigger */
-		spin_unlock(&sbi->fs_lock);
+		int status;
 
-		if (oz_mode)
-			return 1;
+		/*
+		 * We can only unhash and send this to ->lookup() if
+		 * the directory mutex is held over d_revalidate() and
+		 * ->lookup(). This prevents the VFS from incorrectly
+		 * seeing the dentry as non-existent.
+		 */
+		ino->flags |= AUTOFS_INF_PENDING;
+		if (!mutex_aquired) {
+			autofs4_revalidate_drop(dentry, entry);
+			spin_unlock(&dcache_lock);
+			spin_unlock(&sbi->fs_lock);
+			return 0;
+		}
+		spin_unlock(&dcache_lock);
+		spin_unlock(&sbi->fs_lock);
+		mutex_unlock(&dir->i_mutex);
+		kfree(entry);
 
 		/*
 		 * If the directory has gone away due to an expire
@@ -279,46 +415,82 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 		 * A zero status is success otherwise we have a
 		 * negative error code.
 		 */
-		status = try_to_fill_dentry(dentry, flags);
+		status = try_to_fill_dentry(dentry);
+
+		spin_lock(&sbi->fs_lock);
+		ino->flags &= ~AUTOFS_INF_PENDING;
+		spin_unlock(&sbi->fs_lock);
+
 		if (status == 0)
 			return 1;
 
 		return status;
 	}
-	spin_unlock(&sbi->fs_lock);
-
-	/* Negative dentry.. invalidate if "old" */
-	if (dentry->d_inode == NULL)
-		return 0;
 
 	/* Check for a non-mountpoint directory with no contents */
-	spin_lock(&dcache_lock);
 	if (S_ISDIR(dentry->d_inode->i_mode) &&
-	    !d_mountpoint(dentry) && 
-	    __simple_empty(dentry)) {
+	    !d_mountpoint(dentry) && list_empty(&dentry->d_subdirs)) {
 		DPRINTK("dentry=%p %.*s, emptydir",
 			 dentry, dentry->d_name.len, dentry->d_name.name);
-		spin_unlock(&dcache_lock);
 
-		/* The daemon never causes a mount to trigger */
-		if (oz_mode)
-			return 1;
+		if (autofs4_need_mount(flags) || current->link_count) {
+			int status;
 
-		/*
-		 * A zero status is success otherwise we have a
-		 * negative error code.
-		 */
-		status = try_to_fill_dentry(dentry, flags);
-		if (status == 0)
-			return 1;
+			/*
+			 * We can only unhash and send this to ->lookup() if
+			 * the directory mutex is held over d_revalidate() and
+			 * ->lookup(). This prevents the VFS from incorrectly
+			 * seeing the dentry as non-existent.
+			 */
+			ino->flags |= AUTOFS_INF_PENDING;
+			if (!mutex_aquired) {
+				autofs4_revalidate_drop(dentry, entry);
+				spin_unlock(&dcache_lock);
+				spin_unlock(&sbi->fs_lock);
+				return 0;
+			}
+			spin_unlock(&dcache_lock);
+			spin_unlock(&sbi->fs_lock);
+			mutex_unlock(&dir->i_mutex);
+			kfree(entry);
 
-		return status;
+			/*
+			 * A zero status is success otherwise we have a
+			 * negative error code.
+			 */
+			status = try_to_fill_dentry(dentry);
+
+			spin_lock(&sbi->fs_lock);
+			ino->flags &= ~AUTOFS_INF_PENDING;
+			spin_unlock(&sbi->fs_lock);
+
+			if (status == 0)
+				return 1;
+
+			return status;
+		}
 	}
 	spin_unlock(&dcache_lock);
+	spin_unlock(&sbi->fs_lock);
+
+	if (mutex_aquired)
+		mutex_unlock(&dir->i_mutex);
+
+	kfree(entry);
 
 	return 1;
 }
 
+static void autofs4_free_rehash_entrys(struct autofs_info *inf)
+{
+	struct list_head *head = &inf->rehash_list;
+	struct rehash_entry *entry, *next;
+	list_for_each_entry_safe(entry, next, head, list) {
+		list_del(&entry->list);
+		kfree(entry);
+	}
+}
+
 void autofs4_dentry_release(struct dentry *de)
 {
 	struct autofs_info *inf;
@@ -337,6 +509,8 @@ void autofs4_dentry_release(struct dentry *de)
 				list_del(&inf->active);
 			if (!list_empty(&inf->expiring))
 				list_del(&inf->expiring);
+			if (!list_empty(&inf->rehash_list))
+				autofs4_free_rehash_entrys(inf);
 			spin_unlock(&sbi->lookup_lock);
 		}
 
@@ -359,35 +533,52 @@ static const struct dentry_operations autofs4_dentry_operations = {
 	.d_release	= autofs4_dentry_release,
 };
 
-static struct dentry *autofs4_lookup_active(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name)
+static struct dentry *autofs4_lookup_active(struct dentry *dentry)
 {
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct dentry *parent = dentry->d_parent;
+	struct qstr *name = &dentry->d_name;
 	unsigned int len = name->len;
 	unsigned int hash = name->hash;
 	const unsigned char *str = name->name;
 	struct list_head *p, *head;
 
+restart:
 	spin_lock(&dcache_lock);
 	spin_lock(&sbi->lookup_lock);
 	head = &sbi->active_list;
 	list_for_each(p, head) {
 		struct autofs_info *ino;
-		struct dentry *dentry;
+		struct dentry *active;
 		struct qstr *qstr;
 
 		ino = list_entry(p, struct autofs_info, active);
-		dentry = ino->dentry;
+		active = ino->dentry;
 
-		spin_lock(&dentry->d_lock);
+		spin_lock(&active->d_lock);
 
 		/* Already gone? */
-		if (atomic_read(&dentry->d_count) == 0)
+		if (atomic_read(&active->d_count) == 0)
 			goto next;
 
-		qstr = &dentry->d_name;
+		if (active->d_inode && IS_DEADDIR(active->d_inode)) {
+			if (!list_empty(&ino->rehash_list)) {
+				dget(active);
+				spin_unlock(&active->d_lock);
+				spin_unlock(&sbi->lookup_lock);
+				spin_unlock(&dcache_lock);
+				autofs4_remove_rehash_entrys(ino);
+				dput(active);
+				goto restart;
+			}
+			goto next;
+		}
+
+		qstr = &active->d_name;
 
-		if (dentry->d_name.hash != hash)
+		if (active->d_name.hash != hash)
 			goto next;
-		if (dentry->d_parent != parent)
+		if (active->d_parent != parent)
 			goto next;
 
 		if (qstr->len != len)
@@ -395,15 +586,13 @@ static struct dentry *autofs4_lookup_active(struct autofs_sb_info *sbi, struct d
 		if (memcmp(qstr->name, str, len))
 			goto next;
 
-		if (d_unhashed(dentry)) {
-			dget(dentry);
-			spin_unlock(&dentry->d_lock);
-			spin_unlock(&sbi->lookup_lock);
-			spin_unlock(&dcache_lock);
-			return dentry;
-		}
+		dget(active);
+		spin_unlock(&active->d_lock);
+		spin_unlock(&sbi->lookup_lock);
+		spin_unlock(&dcache_lock);
+		return active;
 next:
-		spin_unlock(&dentry->d_lock);
+		spin_unlock(&active->d_lock);
 	}
 	spin_unlock(&sbi->lookup_lock);
 	spin_unlock(&dcache_lock);
@@ -411,8 +600,11 @@ next:
 	return NULL;
 }
 
-static struct dentry *autofs4_lookup_expiring(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name)
+static struct dentry *autofs4_lookup_expiring(struct dentry *dentry)
 {
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct dentry *parent = dentry->d_parent;
+	struct qstr *name = &dentry->d_name;
 	unsigned int len = name->len;
 	unsigned int hash = name->hash;
 	const unsigned char *str = name->name;
@@ -423,23 +615,23 @@ static struct dentry *autofs4_lookup_expiring(struct autofs_sb_info *sbi, struct
 	head = &sbi->expiring_list;
 	list_for_each(p, head) {
 		struct autofs_info *ino;
-		struct dentry *dentry;
+		struct dentry *expiring;
 		struct qstr *qstr;
 
 		ino = list_entry(p, struct autofs_info, expiring);
-		dentry = ino->dentry;
+		expiring = ino->dentry;
 
-		spin_lock(&dentry->d_lock);
+		spin_lock(&expiring->d_lock);
 
 		/* Bad luck, we've already been dentry_iput */
-		if (!dentry->d_inode)
+		if (!expiring->d_inode)
 			goto next;
 
-		qstr = &dentry->d_name;
+		qstr = &expiring->d_name;
 
-		if (dentry->d_name.hash != hash)
+		if (expiring->d_name.hash != hash)
 			goto next;
-		if (dentry->d_parent != parent)
+		if (expiring->d_parent != parent)
 			goto next;
 
 		if (qstr->len != len)
@@ -447,15 +639,13 @@ static struct dentry *autofs4_lookup_expiring(struct autofs_sb_info *sbi, struct
 		if (memcmp(qstr->name, str, len))
 			goto next;
 
-		if (d_unhashed(dentry)) {
-			dget(dentry);
-			spin_unlock(&dentry->d_lock);
-			spin_unlock(&sbi->lookup_lock);
-			spin_unlock(&dcache_lock);
-			return dentry;
-		}
+		dget(expiring);
+		spin_unlock(&expiring->d_lock);
+		spin_unlock(&sbi->lookup_lock);
+		spin_unlock(&dcache_lock);
+		return expiring;
 next:
-		spin_unlock(&dentry->d_lock);
+		spin_unlock(&expiring->d_lock);
 	}
 	spin_unlock(&sbi->lookup_lock);
 	spin_unlock(&dcache_lock);
@@ -463,13 +653,56 @@ next:
 	return NULL;
 }
 
+static struct autofs_info *init_new_dentry(struct autofs_sb_info *sbi,
+					   struct dentry *dentry, int oz_mode)
+{
+	struct autofs_info *ino;
+
+	/*
+	 * Mark the dentry incomplete but don't hash it. We do this
+	 * to serialize our inode creation operations (symlink and
+	 * mkdir) which prevents deadlock during the callback to
+	 * the daemon. Subsequent user space lookups for the same
+	 * dentry are placed on the wait queue while the daemon
+	 * itself is allowed passage unresticted so the create
+	 * operation itself can then hash the dentry. Finally,
+	 * we check for the hashed dentry and return the newly
+	 * hashed dentry.
+	 */
+	dentry->d_op = &autofs4_root_dentry_operations;
+
+	/*
+	 * And we need to ensure that the same dentry is used for
+	 * all following lookup calls until it is hashed so that
+	 * the dentry flags are persistent throughout the request.
+	 */
+	ino = autofs4_init_ino(NULL, sbi, 0555);
+	if (!ino)
+		return ERR_PTR(-ENOMEM);
+
+	dentry->d_fsdata = ino;
+	ino->dentry = dentry;
+
+	/*
+	 * Only set the mount pending flag for new dentrys not created
+	 * by the daemon.
+	 */
+	if (!oz_mode)
+		ino->flags |= AUTOFS_INF_PENDING;
+
+	d_instantiate(dentry, NULL);
+
+	return ino;
+}
+
 /* Lookups in the root directory */
 static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 {
 	struct autofs_sb_info *sbi;
 	struct autofs_info *ino;
-	struct dentry *expiring, *unhashed;
+	struct dentry *expiring, *active;
 	int oz_mode;
+	int status = 0;
 
 	DPRINTK("name = %.*s",
 		dentry->d_name.len, dentry->d_name.name);
@@ -484,123 +717,100 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 	DPRINTK("pid = %u, pgrp = %u, catatonic = %d, oz_mode = %d",
 		 current->pid, task_pgrp_nr(current), sbi->catatonic, oz_mode);
 
-	unhashed = autofs4_lookup_active(sbi, dentry->d_parent, &dentry->d_name);
-	if (unhashed)
-		dentry = unhashed;
-	else {
-		/*
-		 * Mark the dentry incomplete but don't hash it. We do this
-		 * to serialize our inode creation operations (symlink and
-		 * mkdir) which prevents deadlock during the callback to
-		 * the daemon. Subsequent user space lookups for the same
-		 * dentry are placed on the wait queue while the daemon
-		 * itself is allowed passage unresticted so the create
-		 * operation itself can then hash the dentry. Finally,
-		 * we check for the hashed dentry and return the newly
-		 * hashed dentry.
-		 */
-		dentry->d_op = &autofs4_root_dentry_operations;
-
-		/*
-		 * And we need to ensure that the same dentry is used for
-		 * all following lookup calls until it is hashed so that
-		 * the dentry flags are persistent throughout the request.
-		 */
-		ino = autofs4_init_ino(NULL, sbi, 0555);
-		if (!ino)
-			return ERR_PTR(-ENOMEM);
-
-		dentry->d_fsdata = ino;
-		ino->dentry = dentry;
-
-		spin_lock(&sbi->lookup_lock);
-		list_add(&ino->active, &sbi->active_list);
-		spin_unlock(&sbi->lookup_lock);
-
-		d_instantiate(dentry, NULL);
+	spin_lock(&sbi->fs_lock);
+	active = autofs4_lookup_active(dentry);
+	if (active) {
+		dentry = active;
+		ino = autofs4_dentry_ino(dentry);
+		/* If this came from revalidate, rehash it */
+		autofs4_revalidate_rehash(dentry);
+		spin_unlock(&sbi->fs_lock);
+	} else {
+		spin_unlock(&sbi->fs_lock);
+		ino = init_new_dentry(sbi, dentry, oz_mode);
+		if (IS_ERR(ino))
+			return (struct dentry *) ino;
 	}
 
+	autofs4_add_active(dentry);
+
 	if (!oz_mode) {
+		expiring = autofs4_lookup_expiring(dentry);
 		mutex_unlock(&dir->i_mutex);
-		expiring = autofs4_lookup_expiring(sbi,
-						   dentry->d_parent,
-						   &dentry->d_name);
 		if (expiring) {
 			/*
 			 * If we are racing with expire the request might not
 			 * be quite complete but the directory has been removed
 			 * so it must have been successful, so just wait for it.
 			 */
-			ino = autofs4_dentry_ino(expiring);
 			autofs4_expire_wait(expiring);
-			spin_lock(&sbi->lookup_lock);
-			if (!list_empty(&ino->expiring))
-				list_del_init(&ino->expiring);
-			spin_unlock(&sbi->lookup_lock);
 			dput(expiring);
 		}
-
-		spin_lock(&dentry->d_lock);
-		dentry->d_flags |= DCACHE_AUTOFS_PENDING;
-		spin_unlock(&dentry->d_lock);
-		if (dentry->d_op && dentry->d_op->d_revalidate)
-			(dentry->d_op->d_revalidate)(dentry, nd);
+		status = try_to_fill_dentry(dentry);
 		mutex_lock(&dir->i_mutex);
+		spin_lock(&sbi->fs_lock);
+		ino->flags &= ~AUTOFS_INF_PENDING;
+		spin_unlock(&sbi->fs_lock);
 	}
 
+	autofs4_del_active(dentry);
+
 	/*
-	 * If we are still pending, check if we had to handle
+	 * If we had a mount fail, check if we had to handle
 	 * a signal. If so we can force a restart..
 	 */
-	if (dentry->d_flags & DCACHE_AUTOFS_PENDING) {
+	if (status) {
 		/* See if we were interrupted */
 		if (signal_pending(current)) {
 			sigset_t *sigset = &current->pending.signal;
 			if (sigismember (sigset, SIGKILL) ||
 			    sigismember (sigset, SIGQUIT) ||
 			    sigismember (sigset, SIGINT)) {
-			    if (unhashed)
-				dput(unhashed);
+			    if (active)
+				dput(active);
 			    return ERR_PTR(-ERESTARTNOINTR);
 			}
 		}
-		if (!oz_mode) {
-			spin_lock(&dentry->d_lock);
-			dentry->d_flags &= ~DCACHE_AUTOFS_PENDING;
-			spin_unlock(&dentry->d_lock);
+	}
+
+	/*
+	 * User space can (and has done in the past) remove and re-create
+	 * this directory during the callback. This can leave us with an
+	 * unhashed dentry, but a successful mount!  So we need to
+	 * perform another cached lookup in case the dentry now exists.
+	 */
+	if (!oz_mode && !have_submounts(dentry)) {
+		struct dentry *new;
+		new = d_lookup(dentry->d_parent, &dentry->d_name);
+		if (new) {
+			if (active)
+				dput(active);
+			return new;
+		} else {
+			if (!status)
+				status = -ENOENT;
 		}
 	}
 
 	/*
-	 * If this dentry is unhashed, then we shouldn't honour this
-	 * lookup.  Returning ENOENT here doesn't do the right thing
-	 * for all system calls, but it should be OK for the operations
-	 * we permit from an autofs.
+	 * If we had a mount failure, return status to user space.
+	 * If the mount succeeded and we used a dentry from the active queue
+	 * return it.
 	 */
-	if (!oz_mode && d_unhashed(dentry)) {
+	if (status) {
+		dentry = ERR_PTR(status);
+		if (active)
+			dput(active);
+		return dentry;
+	} else {
 		/*
-		 * A user space application can (and has done in the past)
-		 * remove and re-create this directory during the callback.
-		 * This can leave us with an unhashed dentry, but a
-		 * successful mount!  So we need to perform another
-		 * cached lookup in case the dentry now exists.
+		 * Valid successful mount, return active dentry or NULL
+		 * for a new dentry.
 		 */
-		struct dentry *parent = dentry->d_parent;
-		struct dentry *new = d_lookup(parent, &dentry->d_name);
-		if (new != NULL)
-			dentry = new;
-		else
-			dentry = ERR_PTR(-ENOENT);
-
-		if (unhashed)
-			dput(unhashed);
-
-		return dentry;
+		if (active)
+			return active;
 	}
 
-	if (unhashed)
-		return unhashed;
-
 	return NULL;
 }
 
@@ -624,11 +834,6 @@ static int autofs4_dir_symlink(struct inode *dir,
 	if (!ino)
 		return -ENOMEM;
 
-	spin_lock(&sbi->lookup_lock);
-	if (!list_empty(&ino->active))
-		list_del_init(&ino->active);
-	spin_unlock(&sbi->lookup_lock);
-
 	ino->size = strlen(symname);
 	cp = kmalloc(ino->size + 1, GFP_KERNEL);
 	if (!cp) {
@@ -705,10 +910,6 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry)
 	dir->i_mtime = CURRENT_TIME;
 
 	spin_lock(&dcache_lock);
-	spin_lock(&sbi->lookup_lock);
-	if (list_empty(&ino->expiring))
-		list_add(&ino->expiring, &sbi->expiring_list);
-	spin_unlock(&sbi->lookup_lock);
 	spin_lock(&dentry->d_lock);
 	__d_drop(dentry);
 	spin_unlock(&dentry->d_lock);
@@ -734,10 +935,6 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry)
 		spin_unlock(&dcache_lock);
 		return -ENOTEMPTY;
 	}
-	spin_lock(&sbi->lookup_lock);
-	if (list_empty(&ino->expiring))
-		list_add(&ino->expiring, &sbi->expiring_list);
-	spin_unlock(&sbi->lookup_lock);
 	spin_lock(&dentry->d_lock);
 	__d_drop(dentry);
 	spin_unlock(&dentry->d_lock);
@@ -775,11 +972,6 @@ static int autofs4_dir_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (!ino)
 		return -ENOMEM;
 
-	spin_lock(&sbi->lookup_lock);
-	if (!list_empty(&ino->active))
-		list_del_init(&ino->active);
-	spin_unlock(&sbi->lookup_lock);
-
 	inode = autofs4_get_inode(dir->i_sb, ino);
 	if (!inode) {
 		if (!dentry->d_fsdata)
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index 6f60336c6628..8f3d9fd89604 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -353,35 +353,35 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 	struct inode *inode;
 	unsigned i, imap_len;
 	struct bfs_sb_info *info;
-	long ret = -EINVAL;
+	int ret = -EINVAL;
 	unsigned long i_sblock, i_eblock, i_eoff, s_size;
 
 	info = kzalloc(sizeof(*info), GFP_KERNEL);
 	if (!info)
 		return -ENOMEM;
+	mutex_init(&info->bfs_lock);
 	s->s_fs_info = info;
 
 	sb_set_blocksize(s, BFS_BSIZE);
 
-	bh = sb_bread(s, 0);
-	if(!bh)
+	info->si_sbh = sb_bread(s, 0);
+	if (!info->si_sbh)
 		goto out;
-	bfs_sb = (struct bfs_super_block *)bh->b_data;
+	bfs_sb = (struct bfs_super_block *)info->si_sbh->b_data;
 	if (le32_to_cpu(bfs_sb->s_magic) != BFS_MAGIC) {
 		if (!silent)
 			printf("No BFS filesystem on %s (magic=%08x)\n", 
 				s->s_id,  le32_to_cpu(bfs_sb->s_magic));
-		goto out;
+		goto out1;
 	}
 	if (BFS_UNCLEAN(bfs_sb, s) && !silent)
 		printf("%s is unclean, continuing\n", s->s_id);
 
 	s->s_magic = BFS_MAGIC;
-	info->si_sbh = bh;
 
 	if (le32_to_cpu(bfs_sb->s_start) > le32_to_cpu(bfs_sb->s_end)) {
 		printf("Superblock is corrupted\n");
-		goto out;
+		goto out1;
 	}
 
 	info->si_lasti = (le32_to_cpu(bfs_sb->s_start) - BFS_BSIZE) /
@@ -390,7 +390,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 	imap_len = (info->si_lasti / 8) + 1;
 	info->si_imap = kzalloc(imap_len, GFP_KERNEL);
 	if (!info->si_imap)
-		goto out;
+		goto out1;
 	for (i = 0; i < BFS_ROOT_INO; i++)
 		set_bit(i, info->si_imap);
 
@@ -398,15 +398,13 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 	inode = bfs_iget(s, BFS_ROOT_INO);
 	if (IS_ERR(inode)) {
 		ret = PTR_ERR(inode);
-		kfree(info->si_imap);
-		goto out;
+		goto out2;
 	}
 	s->s_root = d_alloc_root(inode);
 	if (!s->s_root) {
 		iput(inode);
 		ret = -ENOMEM;
-		kfree(info->si_imap);
-		goto out;
+		goto out2;
 	}
 
 	info->si_blocks = (le32_to_cpu(bfs_sb->s_end) + 1) >> BFS_BSIZE_BITS;
@@ -419,10 +417,8 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 	bh = sb_bread(s, info->si_blocks - 1);
 	if (!bh) {
 		printf("Last block not available: %lu\n", info->si_blocks - 1);
-		iput(inode);
 		ret = -EIO;
-		kfree(info->si_imap);
-		goto out;
+		goto out3;
 	}
 	brelse(bh);
 
@@ -459,11 +455,8 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 			printf("Inode 0x%08x corrupted\n", i);
 
 			brelse(bh);
-			s->s_root = NULL;
-			kfree(info->si_imap);
-			kfree(info);
-			s->s_fs_info = NULL;
-			return -EIO;
+			ret = -EIO;
+			goto out3;
 		}
 
 		if (!di->i_ino) {
@@ -483,11 +476,17 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 		s->s_dirt = 1;
 	} 
 	dump_imap("read_super", s);
-	mutex_init(&info->bfs_lock);
 	return 0;
 
+out3:
+	dput(s->s_root);
+	s->s_root = NULL;
+out2:
+	kfree(info->si_imap);
+out1:
+	brelse(info->si_sbh);
 out:
-	brelse(bh);
+	mutex_destroy(&info->bfs_lock);
 	kfree(info);
 	s->s_fs_info = NULL;
 	return ret;
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index b639dcf7c778..fdd397099172 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -32,7 +32,7 @@
 
 static int load_aout_binary(struct linux_binprm *, struct pt_regs * regs);
 static int load_aout_library(struct file*);
-static int aout_core_dump(long signr, struct pt_regs *regs, struct file *file, unsigned long limit);
+static int aout_core_dump(struct coredump_params *cprm);
 
 static struct linux_binfmt aout_format = {
 	.module		= THIS_MODULE,
@@ -89,8 +89,9 @@ if (file->f_op->llseek) { \
  * dumping of the process results in another error..
  */
 
-static int aout_core_dump(long signr, struct pt_regs *regs, struct file *file, unsigned long limit)
+static int aout_core_dump(struct coredump_params *cprm)
 {
+	struct file *file = cprm->file;
 	mm_segment_t fs;
 	int has_dumped = 0;
 	unsigned long dump_start, dump_size;
@@ -108,16 +109,16 @@ static int aout_core_dump(long signr, struct pt_regs *regs, struct file *file, u
 	current->flags |= PF_DUMPCORE;
        	strncpy(dump.u_comm, current->comm, sizeof(dump.u_comm));
 	dump.u_ar0 = offsetof(struct user, regs);
-	dump.signal = signr;
-	aout_dump_thread(regs, &dump);
+	dump.signal = cprm->signr;
+	aout_dump_thread(cprm->regs, &dump);
 
 /* If the size of the dump file exceeds the rlimit, then see what would happen
    if we wrote the stack, but not the data area.  */
-	if ((dump.u_dsize + dump.u_ssize+1) * PAGE_SIZE > limit)
+	if ((dump.u_dsize + dump.u_ssize+1) * PAGE_SIZE > cprm->limit)
 		dump.u_dsize = 0;
 
 /* Make sure we have enough room to write the stack and data areas. */
-	if ((dump.u_ssize + 1) * PAGE_SIZE > limit)
+	if ((dump.u_ssize + 1) * PAGE_SIZE > cprm->limit)
 		dump.u_ssize = 0;
 
 /* make sure we actually have a data and stack area to dump */
@@ -263,6 +264,7 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 #else
 	set_personality(PER_LINUX);
 #endif
+	setup_new_exec(bprm);
 
 	current->mm->end_code = ex.a_text +
 		(current->mm->start_code = N_TXTADDR(ex));
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index d15ea1790bfb..fd5b2ea5d299 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -44,8 +44,8 @@ static unsigned long elf_map(struct file *, unsigned long, struct elf_phdr *,
  * If we don't support core dumping, then supply a NULL so we
  * don't even try.
  */
-#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE)
-static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, unsigned long limit);
+#ifdef CONFIG_ELF_CORE
+static int elf_core_dump(struct coredump_params *cprm);
 #else
 #define elf_core_dump	NULL
 #endif
@@ -662,27 +662,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 			if (elf_interpreter[elf_ppnt->p_filesz - 1] != '\0')
 				goto out_free_interp;
 
-			/*
-			 * The early SET_PERSONALITY here is so that the lookup
-			 * for the interpreter happens in the namespace of the 
-			 * to-be-execed image.  SET_PERSONALITY can select an
-			 * alternate root.
-			 *
-			 * However, SET_PERSONALITY is NOT allowed to switch
-			 * this task into the new images's memory mapping
-			 * policy - that is, TASK_SIZE must still evaluate to
-			 * that which is appropriate to the execing application.
-			 * This is because exit_mmap() needs to have TASK_SIZE
-			 * evaluate to the size of the old image.
-			 *
-			 * So if (say) a 64-bit application is execing a 32-bit
-			 * application it is the architecture's responsibility
-			 * to defer changing the value of TASK_SIZE until the
-			 * switch really is going to happen - do this in
-			 * flush_thread().	- akpm
-			 */
-			SET_PERSONALITY(loc->elf_ex);
-
 			interpreter = open_exec(elf_interpreter);
 			retval = PTR_ERR(interpreter);
 			if (IS_ERR(interpreter))
@@ -730,9 +709,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 		/* Verify the interpreter has a valid arch */
 		if (!elf_check_arch(&loc->interp_elf_ex))
 			goto out_free_dentry;
-	} else {
-		/* Executables without an interpreter also need a personality  */
-		SET_PERSONALITY(loc->elf_ex);
 	}
 
 	/* Flush all traces of the currently running executable */
@@ -752,7 +728,8 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 
 	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
 		current->flags |= PF_RANDOMIZE;
-	arch_pick_mmap_layout(current->mm);
+
+	setup_new_exec(bprm);
 
 	/* Do this so that we can load the interpreter, if need be.  We will
 	   change some of these later */
@@ -1101,12 +1078,7 @@ out:
 	return error;
 }
 
-/*
- * Note that some platforms still use traditional core dumps and not
- * the ELF core dump.  Each platform can select it as appropriate.
- */
-#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE)
-
+#ifdef CONFIG_ELF_CORE
 /*
  * ELF core dumper
  *
@@ -1277,8 +1249,9 @@ static int writenote(struct memelfnote *men, struct file *file,
 }
 #undef DUMP_WRITE
 
-#define DUMP_WRITE(addr, nr)	\
-	if ((size += (nr)) > limit || !dump_write(file, (addr), (nr))) \
+#define DUMP_WRITE(addr, nr)				\
+	if ((size += (nr)) > cprm->limit ||		\
+	    !dump_write(cprm->file, (addr), (nr)))	\
 		goto end_coredump;
 
 static void fill_elf_header(struct elfhdr *elf, int segs,
@@ -1906,7 +1879,7 @@ static struct vm_area_struct *next_vma(struct vm_area_struct *this_vma,
  * and then they are actually written out.  If we run out of core limit
  * we just truncate.
  */
-static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, unsigned long limit)
+static int elf_core_dump(struct coredump_params *cprm)
 {
 	int has_dumped = 0;
 	mm_segment_t fs;
@@ -1952,7 +1925,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un
 	 * notes.  This also sets up the file header.
 	 */
 	if (!fill_note_info(elf, segs + 1, /* including notes section */
-			    &info, signr, regs))
+			    &info, cprm->signr, cprm->regs))
 		goto cleanup;
 
 	has_dumped = 1;
@@ -2014,14 +1987,14 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un
 #endif
 
  	/* write out the notes section */
-	if (!write_note_info(&info, file, &foffset))
+	if (!write_note_info(&info, cprm->file, &foffset))
 		goto end_coredump;
 
-	if (elf_coredump_extra_notes_write(file, &foffset))
+	if (elf_coredump_extra_notes_write(cprm->file, &foffset))
 		goto end_coredump;
 
 	/* Align to page */
-	if (!dump_seek(file, dataoff - foffset))
+	if (!dump_seek(cprm->file, dataoff - foffset))
 		goto end_coredump;
 
 	for (vma = first_vma(current, gate_vma); vma != NULL;
@@ -2038,12 +2011,13 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file, un
 			page = get_dump_page(addr);
 			if (page) {
 				void *kaddr = kmap(page);
-				stop = ((size += PAGE_SIZE) > limit) ||
-					!dump_write(file, kaddr, PAGE_SIZE);
+				stop = ((size += PAGE_SIZE) > cprm->limit) ||
+					!dump_write(cprm->file, kaddr,
+						    PAGE_SIZE);
 				kunmap(page);
 				page_cache_release(page);
 			} else
-				stop = !dump_seek(file, PAGE_SIZE);
+				stop = !dump_seek(cprm->file, PAGE_SIZE);
 			if (stop)
 				goto end_coredump;
 		}
@@ -2063,7 +2037,7 @@ out:
 	return has_dumped;
 }
 
-#endif		/* USE_ELF_CORE_DUMP */
+#endif		/* CONFIG_ELF_CORE */
 
 static int __init init_elf_binfmt(void)
 {
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index 38502c67987c..18d77297ccc8 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -75,14 +75,14 @@ static int elf_fdpic_map_file_constdisp_on_uclinux(struct elf_fdpic_params *,
 static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *,
 					     struct file *, struct mm_struct *);
 
-#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE)
-static int elf_fdpic_core_dump(long, struct pt_regs *, struct file *, unsigned long limit);
+#ifdef CONFIG_ELF_CORE
+static int elf_fdpic_core_dump(struct coredump_params *cprm);
 #endif
 
 static struct linux_binfmt elf_fdpic_format = {
 	.module		= THIS_MODULE,
 	.load_binary	= load_elf_fdpic_binary,
-#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE)
+#ifdef CONFIG_ELF_CORE
 	.core_dump	= elf_fdpic_core_dump,
 #endif
 	.min_coredump	= ELF_EXEC_PAGESIZE,
@@ -171,6 +171,9 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm,
 #ifdef ELF_FDPIC_PLAT_INIT
 	unsigned long dynaddr;
 #endif
+#ifndef CONFIG_MMU
+	unsigned long stack_prot;
+#endif
 	struct file *interpreter = NULL; /* to shut gcc up */
 	char *interpreter_name = NULL;
 	int executable_stack;
@@ -316,6 +319,11 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm,
 	 * defunct, deceased, etc. after this point we have to exit via
 	 * error_kill */
 	set_personality(PER_LINUX_FDPIC);
+	if (elf_read_implies_exec(&exec_params.hdr, executable_stack))
+		current->personality |= READ_IMPLIES_EXEC;
+
+	setup_new_exec(bprm);
+
 	set_binfmt(&elf_fdpic_format);
 
 	current->mm->start_code = 0;
@@ -377,10 +385,15 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm,
 	if (stack_size < PAGE_SIZE * 2)
 		stack_size = PAGE_SIZE * 2;
 
+	stack_prot = PROT_READ | PROT_WRITE;
+	if (executable_stack == EXSTACK_ENABLE_X ||
+	    (executable_stack == EXSTACK_DEFAULT && VM_STACK_FLAGS & VM_EXEC))
+		stack_prot |= PROT_EXEC;
+
 	down_write(&current->mm->mmap_sem);
-	current->mm->start_brk = do_mmap(NULL, 0, stack_size,
-					 PROT_READ | PROT_WRITE | PROT_EXEC,
-					 MAP_PRIVATE | MAP_ANONYMOUS | MAP_GROWSDOWN,
+	current->mm->start_brk = do_mmap(NULL, 0, stack_size, stack_prot,
+					 MAP_PRIVATE | MAP_ANONYMOUS |
+					 MAP_UNINITIALIZED | MAP_GROWSDOWN,
 					 0);
 
 	if (IS_ERR_VALUE(current->mm->start_brk)) {
@@ -1200,7 +1213,7 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params,
  *
  * Modelled on fs/binfmt_elf.c core dumper
  */
-#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE)
+#ifdef CONFIG_ELF_CORE
 
 /*
  * These are the only things you should do on a core-file: use only these
@@ -1325,8 +1338,9 @@ static int writenote(struct memelfnote *men, struct file *file)
 #undef DUMP_WRITE
 #undef DUMP_SEEK
 
-#define DUMP_WRITE(addr, nr)	\
-	if ((size += (nr)) > limit || !dump_write(file, (addr), (nr))) \
+#define DUMP_WRITE(addr, nr)				\
+	if ((size += (nr)) > cprm->limit ||		\
+	    !dump_write(cprm->file, (addr), (nr)))	\
 		goto end_coredump;
 
 static inline void fill_elf_fdpic_header(struct elfhdr *elf, int segs)
@@ -1581,8 +1595,7 @@ static int elf_fdpic_dump_segments(struct file *file, size_t *size,
  * and then they are actually written out.  If we run out of core limit
  * we just truncate.
  */
-static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
-			       struct file *file, unsigned long limit)
+static int elf_fdpic_core_dump(struct coredump_params *cprm)
 {
 #define	NUM_NOTES	6
 	int has_dumped = 0;
@@ -1641,7 +1654,7 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 		goto cleanup;
 #endif
 
-	if (signr) {
+	if (cprm->signr) {
 		struct core_thread *ct;
 		struct elf_thread_status *tmp;
 
@@ -1660,14 +1673,14 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 			int sz;
 
 			tmp = list_entry(t, struct elf_thread_status, list);
-			sz = elf_dump_thread_status(signr, tmp);
+			sz = elf_dump_thread_status(cprm->signr, tmp);
 			thread_status_size += sz;
 		}
 	}
 
 	/* now collect the dump for the current */
-	fill_prstatus(prstatus, current, signr);
-	elf_core_copy_regs(&prstatus->pr_reg, regs);
+	fill_prstatus(prstatus, current, cprm->signr);
+	elf_core_copy_regs(&prstatus->pr_reg, cprm->regs);
 
 	segs = current->mm->map_count;
 #ifdef ELF_CORE_EXTRA_PHDRS
@@ -1702,7 +1715,7 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 
   	/* Try to dump the FPU. */
 	if ((prstatus->pr_fpvalid =
-	     elf_core_copy_task_fpregs(current, regs, fpu)))
+	     elf_core_copy_task_fpregs(current, cprm->regs, fpu)))
 		fill_note(notes + numnote++,
 			  "CORE", NT_PRFPREG, sizeof(*fpu), fpu);
 #ifdef ELF_CORE_COPY_XFPREGS
@@ -1773,7 +1786,7 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 
  	/* write out the notes section */
 	for (i = 0; i < numnote; i++)
-		if (!writenote(notes + i, file))
+		if (!writenote(notes + i, cprm->file))
 			goto end_coredump;
 
 	/* write out the thread status notes section */
@@ -1782,25 +1795,26 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 				list_entry(t, struct elf_thread_status, list);
 
 		for (i = 0; i < tmp->num_notes; i++)
-			if (!writenote(&tmp->notes[i], file))
+			if (!writenote(&tmp->notes[i], cprm->file))
 				goto end_coredump;
 	}
 
-	if (!dump_seek(file, dataoff))
+	if (!dump_seek(cprm->file, dataoff))
 		goto end_coredump;
 
-	if (elf_fdpic_dump_segments(file, &size, &limit, mm_flags) < 0)
+	if (elf_fdpic_dump_segments(cprm->file, &size, &cprm->limit,
+				    mm_flags) < 0)
 		goto end_coredump;
 
 #ifdef ELF_CORE_WRITE_EXTRA_DATA
 	ELF_CORE_WRITE_EXTRA_DATA;
 #endif
 
-	if (file->f_pos != offset) {
+	if (cprm->file->f_pos != offset) {
 		/* Sanity check */
 		printk(KERN_WARNING
 		       "elf_core_dump: file->f_pos (%lld) != offset (%lld)\n",
-		       file->f_pos, offset);
+		       cprm->file->f_pos, offset);
 	}
 
 end_coredump:
@@ -1825,4 +1839,4 @@ cleanup:
 #undef NUM_NOTES
 }
 
-#endif		/* USE_ELF_CORE_DUMP */
+#endif		/* CONFIG_ELF_CORE */
diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c
index a2796651e756..42c6b4a54445 100644
--- a/fs/binfmt_flat.c
+++ b/fs/binfmt_flat.c
@@ -87,7 +87,7 @@ static int load_flat_shared_library(int id, struct lib_info *p);
 #endif
 
 static int load_flat_binary(struct linux_binprm *, struct pt_regs * regs);
-static int flat_core_dump(long signr, struct pt_regs *regs, struct file *file, unsigned long limit);
+static int flat_core_dump(struct coredump_params *cprm);
 
 static struct linux_binfmt flat_format = {
 	.module		= THIS_MODULE,
@@ -102,10 +102,10 @@ static struct linux_binfmt flat_format = {
  * Currently only a stub-function.
  */
 
-static int flat_core_dump(long signr, struct pt_regs *regs, struct file *file, unsigned long limit)
+static int flat_core_dump(struct coredump_params *cprm)
 {
 	printk("Process %s:%d received signr %d and should have core dumped\n",
-			current->comm, current->pid, (int) signr);
+			current->comm, current->pid, (int) cprm->signr);
 	return(1);
 }
 
@@ -519,6 +519,7 @@ static int load_flat_file(struct linux_binprm * bprm,
 
 		/* OK, This is the point of no return */
 		set_personality(PER_LINUX_32BIT);
+		setup_new_exec(bprm);
 	}
 
 	/*
diff --git a/fs/binfmt_som.c b/fs/binfmt_som.c
index eff74b9c9e77..cc8560f6c9b0 100644
--- a/fs/binfmt_som.c
+++ b/fs/binfmt_som.c
@@ -43,7 +43,7 @@ static int load_som_library(struct file *);
  * don't even try.
  */
 #if 0
-static int som_core_dump(long signr, struct pt_regs *regs, unsigned long limit);
+static int som_core_dump(struct coredump_params *cprm);
 #else
 #define som_core_dump	NULL
 #endif
@@ -227,6 +227,7 @@ load_som_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 	/* OK, This is the point of no return */
 	current->flags &= ~PF_FORKNOEXEC;
 	current->personality = PER_HPUX;
+	setup_new_exec(bprm);
 
 	/* Set the task size for HP-UX processes such that
 	 * the gateway page is outside the address space.
diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c
index 49a34e7f7306..a16f29e888cd 100644
--- a/fs/bio-integrity.c
+++ b/fs/bio-integrity.c
@@ -61,7 +61,7 @@ static inline unsigned int vecs_to_idx(unsigned int nr)
 
 static inline int use_bip_pool(unsigned int idx)
 {
-	if (idx == BIOVEC_NR_POOLS)
+	if (idx == BIOVEC_MAX_IDX)
 		return 1;
 
 	return 0;
@@ -95,6 +95,7 @@ struct bio_integrity_payload *bio_integrity_alloc_bioset(struct bio *bio,
 
 	/* Use mempool if lower order alloc failed or max vecs were requested */
 	if (bip == NULL) {
+		idx = BIOVEC_MAX_IDX;  /* so we free the payload properly later */
 		bip = mempool_alloc(bs->bio_integrity_pool, gfp_mask);
 
 		if (unlikely(bip == NULL)) {
diff --git a/fs/bio.c b/fs/bio.c
index 76e6713abf94..88094afc29ea 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -78,7 +78,7 @@ static struct kmem_cache *bio_find_or_create_slab(unsigned int extra_size)
 
 	i = 0;
 	while (i < bio_slab_nr) {
-		struct bio_slab *bslab = &bio_slabs[i];
+		bslab = &bio_slabs[i];
 
 		if (!bslab->slab && entry == -1)
 			entry = i;
@@ -542,13 +542,18 @@ static int __bio_add_page(struct request_queue *q, struct bio *bio, struct page
 
 		if (page == prev->bv_page &&
 		    offset == prev->bv_offset + prev->bv_len) {
+			unsigned int prev_bv_len = prev->bv_len;
 			prev->bv_len += len;
 
 			if (q->merge_bvec_fn) {
 				struct bvec_merge_data bvm = {
+					/* prev_bvec is already charged in
+					   bi_size, discharge it in order to
+					   simulate merging updated prev_bvec
+					   as new bvec. */
 					.bi_bdev = bio->bi_bdev,
 					.bi_sector = bio->bi_sector,
-					.bi_size = bio->bi_size,
+					.bi_size = bio->bi_size - prev_bv_len,
 					.bi_rw = bio->bi_rw,
 				};
 
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 361604244271..6df6d6ed74fd 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -73,13 +73,13 @@ static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
 	return acl;
 }
 
-static int btrfs_xattr_get_acl(struct inode *inode, int type,
-			       void *value, size_t size)
+static int btrfs_xattr_acl_get(struct dentry *dentry, const char *name,
+		void *value, size_t size, int type)
 {
 	struct posix_acl *acl;
 	int ret = 0;
 
-	acl = btrfs_get_acl(inode, type);
+	acl = btrfs_get_acl(dentry->d_inode, type);
 
 	if (IS_ERR(acl))
 		return PTR_ERR(acl);
@@ -94,7 +94,8 @@ static int btrfs_xattr_get_acl(struct inode *inode, int type,
 /*
  * Needs to be called with fs_mutex held
  */
-static int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
+static int btrfs_set_acl(struct btrfs_trans_handle *trans,
+			 struct inode *inode, struct posix_acl *acl, int type)
 {
 	int ret, size = 0;
 	const char *name;
@@ -111,12 +112,14 @@ static int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		mode = inode->i_mode;
-		ret = posix_acl_equiv_mode(acl, &mode);
-		if (ret < 0)
-			return ret;
-		ret = 0;
-		inode->i_mode = mode;
 		name = POSIX_ACL_XATTR_ACCESS;
+		if (acl) {
+			ret = posix_acl_equiv_mode(acl, &mode);
+			if (ret < 0)
+				return ret;
+			inode->i_mode = mode;
+		}
+		ret = 0;
 		break;
 	case ACL_TYPE_DEFAULT:
 		if (!S_ISDIR(inode->i_mode))
@@ -140,8 +143,7 @@ static int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 			goto out;
 	}
 
-	ret = __btrfs_setxattr(inode, name, value, size, 0);
-
+	ret = __btrfs_setxattr(trans, inode, name, value, size, 0);
 out:
 	kfree(value);
 
@@ -151,10 +153,10 @@ out:
 	return ret;
 }
 
-static int btrfs_xattr_set_acl(struct inode *inode, int type,
-			       const void *value, size_t size)
+static int btrfs_xattr_acl_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
-	int ret = 0;
+	int ret;
 	struct posix_acl *acl = NULL;
 
 	if (value) {
@@ -167,38 +169,13 @@ static int btrfs_xattr_set_acl(struct inode *inode, int type,
 		}
 	}
 
-	ret = btrfs_set_acl(inode, acl, type);
+	ret = btrfs_set_acl(NULL, dentry->d_inode, acl, type);
 
 	posix_acl_release(acl);
 
 	return ret;
 }
 
-
-static int btrfs_xattr_acl_access_get(struct inode *inode, const char *name,
-				      void *value, size_t size)
-{
-	return btrfs_xattr_get_acl(inode, ACL_TYPE_ACCESS, value, size);
-}
-
-static int btrfs_xattr_acl_access_set(struct inode *inode, const char *name,
-				      const void *value, size_t size, int flags)
-{
-	return btrfs_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size);
-}
-
-static int btrfs_xattr_acl_default_get(struct inode *inode, const char *name,
-				       void *value, size_t size)
-{
-	return btrfs_xattr_get_acl(inode, ACL_TYPE_DEFAULT, value, size);
-}
-
-static int btrfs_xattr_acl_default_set(struct inode *inode, const char *name,
-			       const void *value, size_t size, int flags)
-{
-	return btrfs_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size);
-}
-
 int btrfs_check_acl(struct inode *inode, int mask)
 {
 	struct posix_acl *acl;
@@ -221,7 +198,8 @@ int btrfs_check_acl(struct inode *inode, int mask)
  * stuff has been fixed to work with that.  If the locking stuff changes, we
  * need to re-evaluate the acl locking stuff.
  */
-int btrfs_init_acl(struct inode *inode, struct inode *dir)
+int btrfs_init_acl(struct btrfs_trans_handle *trans,
+		   struct inode *inode, struct inode *dir)
 {
 	struct posix_acl *acl = NULL;
 	int ret = 0;
@@ -246,7 +224,8 @@ int btrfs_init_acl(struct inode *inode, struct inode *dir)
 		mode_t mode;
 
 		if (S_ISDIR(inode->i_mode)) {
-			ret = btrfs_set_acl(inode, acl, ACL_TYPE_DEFAULT);
+			ret = btrfs_set_acl(trans, inode, acl,
+					    ACL_TYPE_DEFAULT);
 			if (ret)
 				goto failed;
 		}
@@ -261,10 +240,11 @@ int btrfs_init_acl(struct inode *inode, struct inode *dir)
 			inode->i_mode = mode;
 			if (ret > 0) {
 				/* we need an acl */
-				ret = btrfs_set_acl(inode, clone,
+				ret = btrfs_set_acl(trans, inode, clone,
 						    ACL_TYPE_ACCESS);
 			}
 		}
+		posix_acl_release(clone);
 	}
 failed:
 	posix_acl_release(acl);
@@ -294,7 +274,7 @@ int btrfs_acl_chmod(struct inode *inode)
 
 	ret = posix_acl_chmod_masq(clone, inode->i_mode);
 	if (!ret)
-		ret = btrfs_set_acl(inode, clone, ACL_TYPE_ACCESS);
+		ret = btrfs_set_acl(NULL, inode, clone, ACL_TYPE_ACCESS);
 
 	posix_acl_release(clone);
 
@@ -303,14 +283,16 @@ int btrfs_acl_chmod(struct inode *inode)
 
 struct xattr_handler btrfs_xattr_acl_default_handler = {
 	.prefix = POSIX_ACL_XATTR_DEFAULT,
-	.get	= btrfs_xattr_acl_default_get,
-	.set	= btrfs_xattr_acl_default_set,
+	.flags	= ACL_TYPE_DEFAULT,
+	.get	= btrfs_xattr_acl_get,
+	.set	= btrfs_xattr_acl_set,
 };
 
 struct xattr_handler btrfs_xattr_acl_access_handler = {
 	.prefix = POSIX_ACL_XATTR_ACCESS,
-	.get	= btrfs_xattr_acl_access_get,
-	.set	= btrfs_xattr_acl_access_set,
+	.flags	= ACL_TYPE_ACCESS,
+	.get	= btrfs_xattr_acl_get,
+	.set	= btrfs_xattr_acl_set,
 };
 
 #else /* CONFIG_BTRFS_FS_POSIX_ACL */
@@ -320,7 +302,8 @@ int btrfs_acl_chmod(struct inode *inode)
 	return 0;
 }
 
-int btrfs_init_acl(struct inode *inode, struct inode *dir)
+int btrfs_init_acl(struct btrfs_trans_handle *trans,
+		   struct inode *inode, struct inode *dir)
 {
 	return 0;
 }
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index f6783a42f010..3f1f50d9d916 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -44,9 +44,6 @@ struct btrfs_inode {
 	 */
 	struct extent_io_tree io_failure_tree;
 
-	/* held while inesrting or deleting extents from files */
-	struct mutex extent_mutex;
-
 	/* held while logging the inode in tree-log.c */
 	struct mutex log_mutex;
 
@@ -166,7 +163,7 @@ static inline struct btrfs_inode *BTRFS_I(struct inode *inode)
 
 static inline void btrfs_i_size_write(struct inode *inode, u64 size)
 {
-	inode->i_size = size;
+	i_size_write(inode, size);
 	BTRFS_I(inode)->disk_i_size = size;
 }
 
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index ec96f3a6d536..c4bc570a396e 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -37,6 +37,11 @@ static int balance_node_right(struct btrfs_trans_handle *trans,
 			      struct extent_buffer *src_buf);
 static int del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root,
 		   struct btrfs_path *path, int level, int slot);
+static int setup_items_for_insert(struct btrfs_trans_handle *trans,
+			struct btrfs_root *root, struct btrfs_path *path,
+			struct btrfs_key *cpu_key, u32 *data_size,
+			u32 total_data, u32 total_size, int nr);
+
 
 struct btrfs_path *btrfs_alloc_path(void)
 {
@@ -451,9 +456,8 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		extent_buffer_get(cow);
 		spin_unlock(&root->node_lock);
 
-		btrfs_free_extent(trans, root, buf->start, buf->len,
-				  parent_start, root->root_key.objectid,
-				  level, 0);
+		btrfs_free_tree_block(trans, root, buf->start, buf->len,
+				parent_start, root->root_key.objectid, level);
 		free_extent_buffer(buf);
 		add_root_to_dirty_list(root);
 	} else {
@@ -468,9 +472,8 @@ static noinline int __btrfs_cow_block(struct btrfs_trans_handle *trans,
 		btrfs_set_node_ptr_generation(parent, parent_slot,
 					      trans->transid);
 		btrfs_mark_buffer_dirty(parent);
-		btrfs_free_extent(trans, root, buf->start, buf->len,
-				  parent_start, root->root_key.objectid,
-				  level, 0);
+		btrfs_free_tree_block(trans, root, buf->start, buf->len,
+				parent_start, root->root_key.objectid, level);
 	}
 	if (unlock_orig)
 		btrfs_tree_unlock(buf);
@@ -1030,8 +1033,8 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		btrfs_tree_unlock(mid);
 		/* once for the path */
 		free_extent_buffer(mid);
-		ret = btrfs_free_extent(trans, root, mid->start, mid->len,
-					0, root->root_key.objectid, level, 1);
+		ret = btrfs_free_tree_block(trans, root, mid->start, mid->len,
+					    0, root->root_key.objectid, level);
 		/* once for the root ptr */
 		free_extent_buffer(mid);
 		return ret;
@@ -1095,10 +1098,10 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 				       1);
 			if (wret)
 				ret = wret;
-			wret = btrfs_free_extent(trans, root, bytenr,
-						 blocksize, 0,
-						 root->root_key.objectid,
-						 level, 0);
+			wret = btrfs_free_tree_block(trans, root,
+						     bytenr, blocksize, 0,
+						     root->root_key.objectid,
+						     level);
 			if (wret)
 				ret = wret;
 		} else {
@@ -1143,9 +1146,8 @@ static noinline int balance_level(struct btrfs_trans_handle *trans,
 		wret = del_ptr(trans, root, path, level + 1, pslot);
 		if (wret)
 			ret = wret;
-		wret = btrfs_free_extent(trans, root, bytenr, blocksize,
-					 0, root->root_key.objectid,
-					 level, 0);
+		wret = btrfs_free_tree_block(trans, root, bytenr, blocksize,
+					 0, root->root_key.objectid, level);
 		if (wret)
 			ret = wret;
 	} else {
@@ -2997,75 +2999,85 @@ again:
 	return ret;
 }
 
-/*
- * This function splits a single item into two items,
- * giving 'new_key' to the new item and splitting the
- * old one at split_offset (from the start of the item).
- *
- * The path may be released by this operation.  After
- * the split, the path is pointing to the old item.  The
- * new item is going to be in the same node as the old one.
- *
- * Note, the item being split must be smaller enough to live alone on
- * a tree block with room for one extra struct btrfs_item
- *
- * This allows us to split the item in place, keeping a lock on the
- * leaf the entire time.
- */
-int btrfs_split_item(struct btrfs_trans_handle *trans,
-		     struct btrfs_root *root,
-		     struct btrfs_path *path,
-		     struct btrfs_key *new_key,
-		     unsigned long split_offset)
+static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
+					 struct btrfs_root *root,
+					 struct btrfs_path *path, int ins_len)
 {
-	u32 item_size;
+	struct btrfs_key key;
 	struct extent_buffer *leaf;
-	struct btrfs_key orig_key;
-	struct btrfs_item *item;
-	struct btrfs_item *new_item;
-	int ret = 0;
-	int slot;
-	u32 nritems;
-	u32 orig_offset;
-	struct btrfs_disk_key disk_key;
-	char *buf;
+	struct btrfs_file_extent_item *fi;
+	u64 extent_len = 0;
+	u32 item_size;
+	int ret;
 
 	leaf = path->nodes[0];
-	btrfs_item_key_to_cpu(leaf, &orig_key, path->slots[0]);
-	if (btrfs_leaf_free_space(root, leaf) >= sizeof(struct btrfs_item))
-		goto split;
+	btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
+
+	BUG_ON(key.type != BTRFS_EXTENT_DATA_KEY &&
+	       key.type != BTRFS_EXTENT_CSUM_KEY);
+
+	if (btrfs_leaf_free_space(root, leaf) >= ins_len)
+		return 0;
 
 	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	if (key.type == BTRFS_EXTENT_DATA_KEY) {
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+				    struct btrfs_file_extent_item);
+		extent_len = btrfs_file_extent_num_bytes(leaf, fi);
+	}
 	btrfs_release_path(root, path);
 
-	path->search_for_split = 1;
 	path->keep_locks = 1;
-
-	ret = btrfs_search_slot(trans, root, &orig_key, path, 0, 1);
+	path->search_for_split = 1;
+	ret = btrfs_search_slot(trans, root, &key, path, 0, 1);
 	path->search_for_split = 0;
+	if (ret < 0)
+		goto err;
 
+	ret = -EAGAIN;
+	leaf = path->nodes[0];
 	/* if our item isn't there or got smaller, return now */
-	if (ret != 0 || item_size != btrfs_item_size_nr(path->nodes[0],
-							path->slots[0])) {
-		path->keep_locks = 0;
-		return -EAGAIN;
+	if (ret > 0 || item_size != btrfs_item_size_nr(leaf, path->slots[0]))
+		goto err;
+
+	if (key.type == BTRFS_EXTENT_DATA_KEY) {
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+				    struct btrfs_file_extent_item);
+		if (extent_len != btrfs_file_extent_num_bytes(leaf, fi))
+			goto err;
 	}
 
 	btrfs_set_path_blocking(path);
-	ret = split_leaf(trans, root, &orig_key, path,
-			 sizeof(struct btrfs_item), 1);
-	path->keep_locks = 0;
+	ret = split_leaf(trans, root, &key, path, ins_len, 1);
 	BUG_ON(ret);
 
+	path->keep_locks = 0;
 	btrfs_unlock_up_safe(path, 1);
+	return 0;
+err:
+	path->keep_locks = 0;
+	return ret;
+}
+
+static noinline int split_item(struct btrfs_trans_handle *trans,
+			       struct btrfs_root *root,
+			       struct btrfs_path *path,
+			       struct btrfs_key *new_key,
+			       unsigned long split_offset)
+{
+	struct extent_buffer *leaf;
+	struct btrfs_item *item;
+	struct btrfs_item *new_item;
+	int slot;
+	char *buf;
+	u32 nritems;
+	u32 item_size;
+	u32 orig_offset;
+	struct btrfs_disk_key disk_key;
+
 	leaf = path->nodes[0];
 	BUG_ON(btrfs_leaf_free_space(root, leaf) < sizeof(struct btrfs_item));
 
-split:
-	/*
-	 * make sure any changes to the path from split_leaf leave it
-	 * in a blocking state
-	 */
 	btrfs_set_path_blocking(path);
 
 	item = btrfs_item_nr(leaf, path->slots[0]);
@@ -3073,19 +3085,19 @@ split:
 	item_size = btrfs_item_size(leaf, item);
 
 	buf = kmalloc(item_size, GFP_NOFS);
+	if (!buf)
+		return -ENOMEM;
+
 	read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf,
 			    path->slots[0]), item_size);
-	slot = path->slots[0] + 1;
-	leaf = path->nodes[0];
 
+	slot = path->slots[0] + 1;
 	nritems = btrfs_header_nritems(leaf);
-
 	if (slot != nritems) {
 		/* shift the items */
 		memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + 1),
-			      btrfs_item_nr_offset(slot),
-			      (nritems - slot) * sizeof(struct btrfs_item));
-
+				btrfs_item_nr_offset(slot),
+				(nritems - slot) * sizeof(struct btrfs_item));
 	}
 
 	btrfs_cpu_key_to_disk(&disk_key, new_key);
@@ -3113,16 +3125,81 @@ split:
 			    item_size - split_offset);
 	btrfs_mark_buffer_dirty(leaf);
 
-	ret = 0;
-	if (btrfs_leaf_free_space(root, leaf) < 0) {
-		btrfs_print_leaf(root, leaf);
-		BUG();
-	}
+	BUG_ON(btrfs_leaf_free_space(root, leaf) < 0);
 	kfree(buf);
+	return 0;
+}
+
+/*
+ * This function splits a single item into two items,
+ * giving 'new_key' to the new item and splitting the
+ * old one at split_offset (from the start of the item).
+ *
+ * The path may be released by this operation.  After
+ * the split, the path is pointing to the old item.  The
+ * new item is going to be in the same node as the old one.
+ *
+ * Note, the item being split must be smaller enough to live alone on
+ * a tree block with room for one extra struct btrfs_item
+ *
+ * This allows us to split the item in place, keeping a lock on the
+ * leaf the entire time.
+ */
+int btrfs_split_item(struct btrfs_trans_handle *trans,
+		     struct btrfs_root *root,
+		     struct btrfs_path *path,
+		     struct btrfs_key *new_key,
+		     unsigned long split_offset)
+{
+	int ret;
+	ret = setup_leaf_for_split(trans, root, path,
+				   sizeof(struct btrfs_item));
+	if (ret)
+		return ret;
+
+	ret = split_item(trans, root, path, new_key, split_offset);
 	return ret;
 }
 
 /*
+ * This function duplicate a item, giving 'new_key' to the new item.
+ * It guarantees both items live in the same tree leaf and the new item
+ * is contiguous with the original item.
+ *
+ * This allows us to split file extent in place, keeping a lock on the
+ * leaf the entire time.
+ */
+int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
+			 struct btrfs_root *root,
+			 struct btrfs_path *path,
+			 struct btrfs_key *new_key)
+{
+	struct extent_buffer *leaf;
+	int ret;
+	u32 item_size;
+
+	leaf = path->nodes[0];
+	item_size = btrfs_item_size_nr(leaf, path->slots[0]);
+	ret = setup_leaf_for_split(trans, root, path,
+				   item_size + sizeof(struct btrfs_item));
+	if (ret)
+		return ret;
+
+	path->slots[0]++;
+	ret = setup_items_for_insert(trans, root, path, new_key, &item_size,
+				     item_size, item_size +
+				     sizeof(struct btrfs_item), 1);
+	BUG_ON(ret);
+
+	leaf = path->nodes[0];
+	memcpy_extent_buffer(leaf,
+			     btrfs_item_ptr_offset(leaf, path->slots[0]),
+			     btrfs_item_ptr_offset(leaf, path->slots[0] - 1),
+			     item_size);
+	return 0;
+}
+
+/*
  * make the item pointed to by the path smaller.  new_size indicates
  * how small to make it, and from_end tells us if we just chop bytes
  * off the end of the item or if we shift the item to chop bytes off
@@ -3714,8 +3791,8 @@ static noinline int btrfs_del_leaf(struct btrfs_trans_handle *trans,
 	 */
 	btrfs_unlock_up_safe(path, 0);
 
-	ret = btrfs_free_extent(trans, root, leaf->start, leaf->len,
-				0, root->root_key.objectid, 0, 0);
+	ret = btrfs_free_tree_block(trans, root, leaf->start, leaf->len,
+				    0, root->root_key.objectid, 0);
 	return ret;
 }
 /*
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 444b3e9b92a4..2aa8ec6a0981 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -310,6 +310,9 @@ struct btrfs_header {
 #define BTRFS_MAX_INLINE_DATA_SIZE(r) (BTRFS_LEAF_DATA_SIZE(r) - \
 					sizeof(struct btrfs_item) - \
 					sizeof(struct btrfs_file_extent_item))
+#define BTRFS_MAX_XATTR_SIZE(r)	(BTRFS_LEAF_DATA_SIZE(r) - \
+				 sizeof(struct btrfs_item) -\
+				 sizeof(struct btrfs_dir_item))
 
 
 /*
@@ -859,8 +862,9 @@ struct btrfs_fs_info {
 	struct mutex ordered_operations_mutex;
 	struct rw_semaphore extent_commit_sem;
 
-	struct rw_semaphore subvol_sem;
+	struct rw_semaphore cleanup_work_sem;
 
+	struct rw_semaphore subvol_sem;
 	struct srcu_struct subvol_srcu;
 
 	struct list_head trans_list;
@@ -868,6 +872,9 @@ struct btrfs_fs_info {
 	struct list_head dead_roots;
 	struct list_head caching_block_groups;
 
+	spinlock_t delayed_iput_lock;
+	struct list_head delayed_iputs;
+
 	atomic_t nr_async_submits;
 	atomic_t async_submit_draining;
 	atomic_t nr_async_bios;
@@ -1034,12 +1041,12 @@ struct btrfs_root {
 	int ref_cows;
 	int track_dirty;
 	int in_radix;
+	int clean_orphans;
 
 	u64 defrag_trans_start;
 	struct btrfs_key defrag_progress;
 	struct btrfs_key defrag_max;
 	int defrag_running;
-	int defrag_level;
 	char *name;
 	int in_sysfs;
 
@@ -1154,6 +1161,7 @@ struct btrfs_root {
 #define BTRFS_MOUNT_SSD_SPREAD		(1 << 8)
 #define BTRFS_MOUNT_NOSSD		(1 << 9)
 #define BTRFS_MOUNT_DISCARD		(1 << 10)
+#define BTRFS_MOUNT_FORCE_COMPRESS      (1 << 11)
 
 #define btrfs_clear_opt(o, opt)		((o) &= ~BTRFS_MOUNT_##opt)
 #define btrfs_set_opt(o, opt)		((o) |= BTRFS_MOUNT_##opt)
@@ -1975,6 +1983,10 @@ struct extent_buffer *btrfs_alloc_free_block(struct btrfs_trans_handle *trans,
 					u64 parent, u64 root_objectid,
 					struct btrfs_disk_key *key, int level,
 					u64 hint, u64 empty_size);
+int btrfs_free_tree_block(struct btrfs_trans_handle *trans,
+			  struct btrfs_root *root,
+			  u64 bytenr, u32 blocksize,
+			  u64 parent, u64 root_objectid, int level);
 struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
 					    struct btrfs_root *root,
 					    u64 bytenr, u32 blocksize,
@@ -2089,6 +2101,10 @@ int btrfs_split_item(struct btrfs_trans_handle *trans,
 		     struct btrfs_path *path,
 		     struct btrfs_key *new_key,
 		     unsigned long split_offset);
+int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
+			 struct btrfs_root *root,
+			 struct btrfs_path *path,
+			 struct btrfs_key *new_key);
 int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root
 		      *root, struct btrfs_key *key, struct btrfs_path *p, int
 		      ins_len, int cow);
@@ -2196,9 +2212,10 @@ int btrfs_delete_one_dir_name(struct btrfs_trans_handle *trans,
 			      struct btrfs_path *path,
 			      struct btrfs_dir_item *di);
 int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
-			    struct btrfs_root *root, const char *name,
-			    u16 name_len, const void *data, u16 data_len,
-			    u64 dir);
+			    struct btrfs_root *root,
+			    struct btrfs_path *path, u64 objectid,
+			    const char *name, u16 name_len,
+			    const void *data, u16 data_len);
 struct btrfs_dir_item *btrfs_lookup_xattr(struct btrfs_trans_handle *trans,
 					  struct btrfs_root *root,
 					  struct btrfs_path *path, u64 dir,
@@ -2292,7 +2309,7 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 			       struct inode *inode, u64 new_size,
 			       u32 min_type);
 
-int btrfs_start_delalloc_inodes(struct btrfs_root *root);
+int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput);
 int btrfs_set_extent_delalloc(struct inode *inode, u64 start, u64 end);
 int btrfs_writepages(struct address_space *mapping,
 		     struct writeback_control *wbc);
@@ -2332,6 +2349,8 @@ int btrfs_orphan_del(struct btrfs_trans_handle *trans, struct inode *inode);
 void btrfs_orphan_cleanup(struct btrfs_root *root);
 int btrfs_cont_expand(struct inode *inode, loff_t size);
 int btrfs_invalidate_inodes(struct btrfs_root *root);
+void btrfs_add_delayed_iput(struct inode *inode);
+void btrfs_run_delayed_iputs(struct btrfs_root *root);
 extern const struct dentry_operations btrfs_dentry_operations;
 
 /* ioctl.c */
@@ -2345,12 +2364,9 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
 			    int skip_pinned);
 int btrfs_check_file(struct btrfs_root *root, struct inode *inode);
 extern const struct file_operations btrfs_file_operations;
-int btrfs_drop_extents(struct btrfs_trans_handle *trans,
-		       struct btrfs_root *root, struct inode *inode,
-		       u64 start, u64 end, u64 locked_end,
-		       u64 inline_limit, u64 *hint_block, int drop_cache);
+int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode,
+		       u64 start, u64 end, u64 *hint_byte, int drop_cache);
 int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
-			      struct btrfs_root *root,
 			      struct inode *inode, u64 start, u64 end);
 int btrfs_release_file(struct inode *inode, struct file *file);
 
@@ -2380,7 +2396,8 @@ int btrfs_check_acl(struct inode *inode, int mask);
 #else
 #define btrfs_check_acl NULL
 #endif
-int btrfs_init_acl(struct inode *inode, struct inode *dir);
+int btrfs_init_acl(struct btrfs_trans_handle *trans,
+		   struct inode *inode, struct inode *dir);
 int btrfs_acl_chmod(struct inode *inode);
 
 /* relocation.c */
diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c
index f3a6075519cc..e9103b3baa49 100644
--- a/fs/btrfs/dir-item.c
+++ b/fs/btrfs/dir-item.c
@@ -68,12 +68,12 @@ static struct btrfs_dir_item *insert_with_overflow(struct btrfs_trans_handle
  * into the tree
  */
 int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
-			    struct btrfs_root *root, const char *name,
-			    u16 name_len, const void *data, u16 data_len,
-			    u64 dir)
+			    struct btrfs_root *root,
+			    struct btrfs_path *path, u64 objectid,
+			    const char *name, u16 name_len,
+			    const void *data, u16 data_len)
 {
 	int ret = 0;
-	struct btrfs_path *path;
 	struct btrfs_dir_item *dir_item;
 	unsigned long name_ptr, data_ptr;
 	struct btrfs_key key, location;
@@ -81,15 +81,11 @@ int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
 	struct extent_buffer *leaf;
 	u32 data_size;
 
-	key.objectid = dir;
+	BUG_ON(name_len + data_len > BTRFS_MAX_XATTR_SIZE(root));
+
+	key.objectid = objectid;
 	btrfs_set_key_type(&key, BTRFS_XATTR_ITEM_KEY);
 	key.offset = btrfs_name_hash(name, name_len);
-	path = btrfs_alloc_path();
-	if (!path)
-		return -ENOMEM;
-	if (name_len + data_len + sizeof(struct btrfs_dir_item) >
-	    BTRFS_LEAF_DATA_SIZE(root) - sizeof(struct btrfs_item))
-		return -ENOSPC;
 
 	data_size = sizeof(*dir_item) + name_len + data_len;
 	dir_item = insert_with_overflow(trans, root, path, &key, data_size,
@@ -117,7 +113,6 @@ int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
 	write_extent_buffer(leaf, data, data_ptr, data_len);
 	btrfs_mark_buffer_dirty(path->nodes[0]);
 
-	btrfs_free_path(path);
 	return ret;
 }
 
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 02b6afbd7450..87b25543d7d1 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -892,6 +892,8 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	root->stripesize = stripesize;
 	root->ref_cows = 0;
 	root->track_dirty = 0;
+	root->in_radix = 0;
+	root->clean_orphans = 0;
 
 	root->fs_info = fs_info;
 	root->objectid = objectid;
@@ -928,7 +930,6 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
 	root->defrag_trans_start = fs_info->generation;
 	init_completion(&root->kobj_unregister);
 	root->defrag_running = 0;
-	root->defrag_level = 0;
 	root->root_key.objectid = objectid;
 	root->anon_super.s_root = NULL;
 	root->anon_super.s_dev = 0;
@@ -980,12 +981,12 @@ int btrfs_free_log_root_tree(struct btrfs_trans_handle *trans,
 
 	while (1) {
 		ret = find_first_extent_bit(&log_root_tree->dirty_log_pages,
-				    0, &start, &end, EXTENT_DIRTY);
+				0, &start, &end, EXTENT_DIRTY | EXTENT_NEW);
 		if (ret)
 			break;
 
-		clear_extent_dirty(&log_root_tree->dirty_log_pages,
-				   start, end, GFP_NOFS);
+		clear_extent_bits(&log_root_tree->dirty_log_pages, start, end,
+				  EXTENT_DIRTY | EXTENT_NEW, GFP_NOFS);
 	}
 	eb = fs_info->log_root_tree->node;
 
@@ -1210,8 +1211,10 @@ again:
 	ret = radix_tree_insert(&fs_info->fs_roots_radix,
 				(unsigned long)root->root_key.objectid,
 				root);
-	if (ret == 0)
+	if (ret == 0) {
 		root->in_radix = 1;
+		root->clean_orphans = 1;
+	}
 	spin_unlock(&fs_info->fs_roots_radix_lock);
 	radix_tree_preload_end();
 	if (ret) {
@@ -1225,10 +1228,6 @@ again:
 	ret = btrfs_find_dead_roots(fs_info->tree_root,
 				    root->root_key.objectid);
 	WARN_ON(ret);
-
-	if (!(fs_info->sb->s_flags & MS_RDONLY))
-		btrfs_orphan_cleanup(root);
-
 	return root;
 fail:
 	free_fs_root(root);
@@ -1477,6 +1476,7 @@ static int cleaner_kthread(void *arg)
 
 		if (!(root->fs_info->sb->s_flags & MS_RDONLY) &&
 		    mutex_trylock(&root->fs_info->cleaner_mutex)) {
+			btrfs_run_delayed_iputs(root);
 			btrfs_clean_old_snapshots(root);
 			mutex_unlock(&root->fs_info->cleaner_mutex);
 		}
@@ -1606,6 +1606,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC);
 	INIT_LIST_HEAD(&fs_info->trans_list);
 	INIT_LIST_HEAD(&fs_info->dead_roots);
+	INIT_LIST_HEAD(&fs_info->delayed_iputs);
 	INIT_LIST_HEAD(&fs_info->hashers);
 	INIT_LIST_HEAD(&fs_info->delalloc_inodes);
 	INIT_LIST_HEAD(&fs_info->ordered_operations);
@@ -1614,6 +1615,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	spin_lock_init(&fs_info->new_trans_lock);
 	spin_lock_init(&fs_info->ref_cache_lock);
 	spin_lock_init(&fs_info->fs_roots_radix_lock);
+	spin_lock_init(&fs_info->delayed_iput_lock);
 
 	init_completion(&fs_info->kobj_unregister);
 	fs_info->tree_root = tree_root;
@@ -1689,6 +1691,7 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	mutex_init(&fs_info->cleaner_mutex);
 	mutex_init(&fs_info->volume_mutex);
 	init_rwsem(&fs_info->extent_commit_sem);
+	init_rwsem(&fs_info->cleanup_work_sem);
 	init_rwsem(&fs_info->subvol_sem);
 
 	btrfs_init_free_cluster(&fs_info->meta_alloc_cluster);
@@ -1990,6 +1993,12 @@ struct btrfs_root *open_ctree(struct super_block *sb,
 	if (!fs_info->fs_root)
 		goto fail_trans_kthread;
 
+	if (!(sb->s_flags & MS_RDONLY)) {
+		down_read(&fs_info->cleanup_work_sem);
+		btrfs_orphan_cleanup(fs_info->fs_root);
+		up_read(&fs_info->cleanup_work_sem);
+	}
+
 	return tree_root;
 
 fail_trans_kthread:
@@ -2386,8 +2395,14 @@ int btrfs_commit_super(struct btrfs_root *root)
 	int ret;
 
 	mutex_lock(&root->fs_info->cleaner_mutex);
+	btrfs_run_delayed_iputs(root);
 	btrfs_clean_old_snapshots(root);
 	mutex_unlock(&root->fs_info->cleaner_mutex);
+
+	/* wait until ongoing cleanup work done */
+	down_write(&root->fs_info->cleanup_work_sem);
+	up_write(&root->fs_info->cleanup_work_sem);
+
 	trans = btrfs_start_transaction(root, 1);
 	ret = btrfs_commit_transaction(trans, root);
 	BUG_ON(ret);
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 94627c4cc193..432a2da4641e 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -83,6 +83,17 @@ static int block_group_bits(struct btrfs_block_group_cache *cache, u64 bits)
 	return (cache->flags & bits) == bits;
 }
 
+void btrfs_get_block_group(struct btrfs_block_group_cache *cache)
+{
+	atomic_inc(&cache->count);
+}
+
+void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
+{
+	if (atomic_dec_and_test(&cache->count))
+		kfree(cache);
+}
+
 /*
  * this adds the block group to the fs_info rb tree for the block group
  * cache
@@ -156,7 +167,7 @@ block_group_cache_tree_search(struct btrfs_fs_info *info, u64 bytenr,
 		}
 	}
 	if (ret)
-		atomic_inc(&ret->count);
+		btrfs_get_block_group(ret);
 	spin_unlock(&info->block_group_cache_lock);
 
 	return ret;
@@ -195,6 +206,14 @@ static int exclude_super_stripes(struct btrfs_root *root,
 	int stripe_len;
 	int i, nr, ret;
 
+	if (cache->key.objectid < BTRFS_SUPER_INFO_OFFSET) {
+		stripe_len = BTRFS_SUPER_INFO_OFFSET - cache->key.objectid;
+		cache->bytes_super += stripe_len;
+		ret = add_excluded_extent(root, cache->key.objectid,
+					  stripe_len);
+		BUG_ON(ret);
+	}
+
 	for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
 		bytenr = btrfs_sb_offset(i);
 		ret = btrfs_rmap_block(&root->fs_info->mapping_tree,
@@ -255,7 +274,7 @@ static u64 add_new_free_space(struct btrfs_block_group_cache *block_group,
 		if (ret)
 			break;
 
-		if (extent_start == start) {
+		if (extent_start <= start) {
 			start = extent_end + 1;
 		} else if (extent_start > start && extent_start < end) {
 			size = extent_start - start;
@@ -399,6 +418,8 @@ err:
 
 	put_caching_control(caching_ctl);
 	atomic_dec(&block_group->space_info->caching_threads);
+	btrfs_put_block_group(block_group);
+
 	return 0;
 }
 
@@ -439,6 +460,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache)
 	up_write(&fs_info->extent_commit_sem);
 
 	atomic_inc(&cache->space_info->caching_threads);
+	btrfs_get_block_group(cache);
 
 	tsk = kthread_run(caching_kthread, cache, "btrfs-cache-%llu\n",
 			  cache->key.objectid);
@@ -478,12 +500,6 @@ struct btrfs_block_group_cache *btrfs_lookup_block_group(
 	return cache;
 }
 
-void btrfs_put_block_group(struct btrfs_block_group_cache *cache)
-{
-	if (atomic_dec_and_test(&cache->count))
-		kfree(cache);
-}
-
 static struct btrfs_space_info *__find_space_info(struct btrfs_fs_info *info,
 						  u64 flags)
 {
@@ -2574,7 +2590,7 @@ next_block_group(struct btrfs_root *root,
 	if (node) {
 		cache = rb_entry(node, struct btrfs_block_group_cache,
 				 cache_node);
-		atomic_inc(&cache->count);
+		btrfs_get_block_group(cache);
 	} else
 		cache = NULL;
 	spin_unlock(&root->fs_info->block_group_cache_lock);
@@ -2880,9 +2896,9 @@ static noinline void flush_delalloc_async(struct btrfs_work *work)
 	root = async->root;
 	info = async->info;
 
-	btrfs_start_delalloc_inodes(root);
+	btrfs_start_delalloc_inodes(root, 0);
 	wake_up(&info->flush_wait);
-	btrfs_wait_ordered_extents(root, 0);
+	btrfs_wait_ordered_extents(root, 0, 0);
 
 	spin_lock(&info->lock);
 	info->flushing = 0;
@@ -2956,8 +2972,8 @@ static void flush_delalloc(struct btrfs_root *root,
 	return;
 
 flush:
-	btrfs_start_delalloc_inodes(root);
-	btrfs_wait_ordered_extents(root, 0);
+	btrfs_start_delalloc_inodes(root, 0);
+	btrfs_wait_ordered_extents(root, 0, 0);
 
 	spin_lock(&info->lock);
 	info->flushing = 0;
@@ -3454,14 +3470,6 @@ static int update_block_group(struct btrfs_trans_handle *trans,
 	else
 		old_val -= num_bytes;
 	btrfs_set_super_bytes_used(&info->super_copy, old_val);
-
-	/* block accounting for root item */
-	old_val = btrfs_root_used(&root->root_item);
-	if (alloc)
-		old_val += num_bytes;
-	else
-		old_val -= num_bytes;
-	btrfs_set_root_used(&root->root_item, old_val);
 	spin_unlock(&info->delalloc_lock);
 
 	while (total) {
@@ -4049,6 +4057,21 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans,
 	return ret;
 }
 
+int btrfs_free_tree_block(struct btrfs_trans_handle *trans,
+			  struct btrfs_root *root,
+			  u64 bytenr, u32 blocksize,
+			  u64 parent, u64 root_objectid, int level)
+{
+	u64 used;
+	spin_lock(&root->node_lock);
+	used = btrfs_root_used(&root->root_item) - blocksize;
+	btrfs_set_root_used(&root->root_item, used);
+	spin_unlock(&root->node_lock);
+
+	return btrfs_free_extent(trans, root, bytenr, blocksize,
+				 parent, root_objectid, level, 0);
+}
+
 static u64 stripe_align(struct btrfs_root *root, u64 val)
 {
 	u64 mask = ((u64)root->stripesize - 1);
@@ -4212,7 +4235,7 @@ search:
 		u64 offset;
 		int cached;
 
-		atomic_inc(&block_group->count);
+		btrfs_get_block_group(block_group);
 		search_start = block_group->key.objectid;
 
 have_block_group:
@@ -4300,7 +4323,7 @@ have_block_group:
 
 				btrfs_put_block_group(block_group);
 				block_group = last_ptr->block_group;
-				atomic_inc(&block_group->count);
+				btrfs_get_block_group(block_group);
 				spin_unlock(&last_ptr->lock);
 				spin_unlock(&last_ptr->refill_lock);
 
@@ -4578,7 +4601,6 @@ int btrfs_reserve_extent(struct btrfs_trans_handle *trans,
 {
 	int ret;
 	u64 search_start = 0;
-	struct btrfs_fs_info *info = root->fs_info;
 
 	data = btrfs_get_alloc_profile(root, data);
 again:
@@ -4586,17 +4608,9 @@ again:
 	 * the only place that sets empty_size is btrfs_realloc_node, which
 	 * is not called recursively on allocations
 	 */
-	if (empty_size || root->ref_cows) {
-		if (!(data & BTRFS_BLOCK_GROUP_METADATA)) {
-			ret = do_chunk_alloc(trans, root->fs_info->extent_root,
-				     2 * 1024 * 1024,
-				     BTRFS_BLOCK_GROUP_METADATA |
-				     (info->metadata_alloc_profile &
-				      info->avail_metadata_alloc_bits), 0);
-		}
+	if (empty_size || root->ref_cows)
 		ret = do_chunk_alloc(trans, root->fs_info->extent_root,
 				     num_bytes + 2 * 1024 * 1024, data, 0);
-	}
 
 	WARN_ON(num_bytes < root->sectorsize);
 	ret = find_free_extent(trans, root, num_bytes, empty_size,
@@ -4897,6 +4911,14 @@ static int alloc_tree_block(struct btrfs_trans_handle *trans,
 					extent_op);
 		BUG_ON(ret);
 	}
+
+	if (root_objectid == root->root_key.objectid) {
+		u64 used;
+		spin_lock(&root->node_lock);
+		used = btrfs_root_used(&root->root_item) + num_bytes;
+		btrfs_set_root_used(&root->root_item, used);
+		spin_unlock(&root->node_lock);
+	}
 	return ret;
 }
 
@@ -4919,8 +4941,16 @@ struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans,
 	btrfs_set_buffer_uptodate(buf);
 
 	if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
-		set_extent_dirty(&root->dirty_log_pages, buf->start,
-			 buf->start + buf->len - 1, GFP_NOFS);
+		/*
+		 * we allow two log transactions at a time, use different
+		 * EXENT bit to differentiate dirty pages.
+		 */
+		if (root->log_transid % 2 == 0)
+			set_extent_dirty(&root->dirty_log_pages, buf->start,
+					buf->start + buf->len - 1, GFP_NOFS);
+		else
+			set_extent_new(&root->dirty_log_pages, buf->start,
+					buf->start + buf->len - 1, GFP_NOFS);
 	} else {
 		set_extent_dirty(&trans->transaction->dirty_pages, buf->start,
 			 buf->start + buf->len - 1, GFP_NOFS);
@@ -7373,9 +7403,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
 			wait_block_group_cache_done(block_group);
 
 		btrfs_remove_free_space_cache(block_group);
-
-		WARN_ON(atomic_read(&block_group->count) != 1);
-		kfree(block_group);
+		btrfs_put_block_group(block_group);
 
 		spin_lock(&info->block_group_cache_lock);
 	}
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 46bea0f4dc7b..428fcac45f90 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -155,20 +155,6 @@ static struct rb_node *__tree_search(struct rb_root *root, u64 offset,
 	return NULL;
 }
 
-/*
- * look for an offset in the tree, and if it can't be found, return
- * the first offset we can find smaller than 'offset'.
- */
-static inline struct rb_node *tree_search(struct rb_root *root, u64 offset)
-{
-	struct rb_node *prev;
-	struct rb_node *ret;
-	ret = __tree_search(root, offset, &prev, NULL);
-	if (!ret)
-		return prev;
-	return ret;
-}
-
 /* check to see if two extent_map structs are adjacent and safe to merge */
 static int mergable_maps(struct extent_map *prev, struct extent_map *next)
 {
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 06550affbd27..c02033596f02 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -179,18 +179,14 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
 		}
 		flags = em->flags;
 		if (skip_pinned && test_bit(EXTENT_FLAG_PINNED, &em->flags)) {
-			if (em->start <= start &&
-			    (!testend || em->start + em->len >= start + len)) {
+			if (testend && em->start + em->len >= start + len) {
 				free_extent_map(em);
 				write_unlock(&em_tree->lock);
 				break;
 			}
-			if (start < em->start) {
-				len = em->start - start;
-			} else {
+			start = em->start + em->len;
+			if (testend)
 				len = start + len - (em->start + em->len);
-				start = em->start + em->len;
-			}
 			free_extent_map(em);
 			write_unlock(&em_tree->lock);
 			continue;
@@ -265,324 +261,253 @@ int btrfs_drop_extent_cache(struct inode *inode, u64 start, u64 end,
  * If an extent intersects the range but is not entirely inside the range
  * it is either truncated or split.  Anything entirely inside the range
  * is deleted from the tree.
- *
- * inline_limit is used to tell this code which offsets in the file to keep
- * if they contain inline extents.
  */
-noinline int btrfs_drop_extents(struct btrfs_trans_handle *trans,
-		       struct btrfs_root *root, struct inode *inode,
-		       u64 start, u64 end, u64 locked_end,
-		       u64 inline_limit, u64 *hint_byte, int drop_cache)
+int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode,
+		       u64 start, u64 end, u64 *hint_byte, int drop_cache)
 {
-	u64 extent_end = 0;
-	u64 search_start = start;
-	u64 ram_bytes = 0;
-	u64 disk_bytenr = 0;
-	u64 orig_locked_end = locked_end;
-	u8 compression;
-	u8 encryption;
-	u16 other_encoding = 0;
+	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_buffer *leaf;
-	struct btrfs_file_extent_item *extent;
+	struct btrfs_file_extent_item *fi;
 	struct btrfs_path *path;
 	struct btrfs_key key;
-	struct btrfs_file_extent_item old;
-	int keep;
-	int slot;
-	int bookend;
-	int found_type = 0;
-	int found_extent;
-	int found_inline;
+	struct btrfs_key new_key;
+	u64 search_start = start;
+	u64 disk_bytenr = 0;
+	u64 num_bytes = 0;
+	u64 extent_offset = 0;
+	u64 extent_end = 0;
+	int del_nr = 0;
+	int del_slot = 0;
+	int extent_type;
 	int recow;
 	int ret;
 
-	inline_limit = 0;
 	if (drop_cache)
 		btrfs_drop_extent_cache(inode, start, end - 1, 0);
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
+
 	while (1) {
 		recow = 0;
-		btrfs_release_path(root, path);
 		ret = btrfs_lookup_file_extent(trans, root, path, inode->i_ino,
 					       search_start, -1);
 		if (ret < 0)
-			goto out;
-		if (ret > 0) {
-			if (path->slots[0] == 0) {
-				ret = 0;
-				goto out;
-			}
-			path->slots[0]--;
+			break;
+		if (ret > 0 && path->slots[0] > 0 && search_start == start) {
+			leaf = path->nodes[0];
+			btrfs_item_key_to_cpu(leaf, &key, path->slots[0] - 1);
+			if (key.objectid == inode->i_ino &&
+			    key.type == BTRFS_EXTENT_DATA_KEY)
+				path->slots[0]--;
 		}
+		ret = 0;
 next_slot:
-		keep = 0;
-		bookend = 0;
-		found_extent = 0;
-		found_inline = 0;
-		compression = 0;
-		encryption = 0;
-		extent = NULL;
 		leaf = path->nodes[0];
-		slot = path->slots[0];
-		ret = 0;
-		btrfs_item_key_to_cpu(leaf, &key, slot);
-		if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY &&
-		    key.offset >= end) {
-			goto out;
-		}
-		if (btrfs_key_type(&key) > BTRFS_EXTENT_DATA_KEY ||
-		    key.objectid != inode->i_ino) {
-			goto out;
-		}
-		if (recow) {
-			search_start = max(key.offset, start);
-			continue;
-		}
-		if (btrfs_key_type(&key) == BTRFS_EXTENT_DATA_KEY) {
-			extent = btrfs_item_ptr(leaf, slot,
-						struct btrfs_file_extent_item);
-			found_type = btrfs_file_extent_type(leaf, extent);
-			compression = btrfs_file_extent_compression(leaf,
-								    extent);
-			encryption = btrfs_file_extent_encryption(leaf,
-								  extent);
-			other_encoding = btrfs_file_extent_other_encoding(leaf,
-								  extent);
-			if (found_type == BTRFS_FILE_EXTENT_REG ||
-			    found_type == BTRFS_FILE_EXTENT_PREALLOC) {
-				extent_end =
-				     btrfs_file_extent_disk_bytenr(leaf,
-								   extent);
-				if (extent_end)
-					*hint_byte = extent_end;
-
-				extent_end = key.offset +
-				     btrfs_file_extent_num_bytes(leaf, extent);
-				ram_bytes = btrfs_file_extent_ram_bytes(leaf,
-								extent);
-				found_extent = 1;
-			} else if (found_type == BTRFS_FILE_EXTENT_INLINE) {
-				found_inline = 1;
-				extent_end = key.offset +
-				     btrfs_file_extent_inline_len(leaf, extent);
+		if (path->slots[0] >= btrfs_header_nritems(leaf)) {
+			BUG_ON(del_nr > 0);
+			ret = btrfs_next_leaf(root, path);
+			if (ret < 0)
+				break;
+			if (ret > 0) {
+				ret = 0;
+				break;
 			}
+			leaf = path->nodes[0];
+			recow = 1;
+		}
+
+		btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
+		if (key.objectid > inode->i_ino ||
+		    key.type > BTRFS_EXTENT_DATA_KEY || key.offset >= end)
+			break;
+
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+				    struct btrfs_file_extent_item);
+		extent_type = btrfs_file_extent_type(leaf, fi);
+
+		if (extent_type == BTRFS_FILE_EXTENT_REG ||
+		    extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
+			disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
+			num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
+			extent_offset = btrfs_file_extent_offset(leaf, fi);
+			extent_end = key.offset +
+				btrfs_file_extent_num_bytes(leaf, fi);
+		} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
+			extent_end = key.offset +
+				btrfs_file_extent_inline_len(leaf, fi);
 		} else {
+			WARN_ON(1);
 			extent_end = search_start;
 		}
 
-		/* we found nothing we can drop */
-		if ((!found_extent && !found_inline) ||
-		    search_start >= extent_end) {
-			int nextret;
-			u32 nritems;
-			nritems = btrfs_header_nritems(leaf);
-			if (slot >= nritems - 1) {
-				nextret = btrfs_next_leaf(root, path);
-				if (nextret)
-					goto out;
-				recow = 1;
-			} else {
-				path->slots[0]++;
-			}
+		if (extent_end <= search_start) {
+			path->slots[0]++;
 			goto next_slot;
 		}
 
-		if (end <= extent_end && start >= key.offset && found_inline)
-			*hint_byte = EXTENT_MAP_INLINE;
-
-		if (found_extent) {
-			read_extent_buffer(leaf, &old, (unsigned long)extent,
-					   sizeof(old));
-		}
-
-		if (end < extent_end && end >= key.offset) {
-			bookend = 1;
-			if (found_inline && start <= key.offset)
-				keep = 1;
+		search_start = max(key.offset, start);
+		if (recow) {
+			btrfs_release_path(root, path);
+			continue;
 		}
 
-		if (bookend && found_extent) {
-			if (locked_end < extent_end) {
-				ret = try_lock_extent(&BTRFS_I(inode)->io_tree,
-						locked_end, extent_end - 1,
-						GFP_NOFS);
-				if (!ret) {
-					btrfs_release_path(root, path);
-					lock_extent(&BTRFS_I(inode)->io_tree,
-						locked_end, extent_end - 1,
-						GFP_NOFS);
-					locked_end = extent_end;
-					continue;
-				}
-				locked_end = extent_end;
+		/*
+		 *     | - range to drop - |
+		 *  | -------- extent -------- |
+		 */
+		if (start > key.offset && end < extent_end) {
+			BUG_ON(del_nr > 0);
+			BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
+
+			memcpy(&new_key, &key, sizeof(new_key));
+			new_key.offset = start;
+			ret = btrfs_duplicate_item(trans, root, path,
+						   &new_key);
+			if (ret == -EAGAIN) {
+				btrfs_release_path(root, path);
+				continue;
 			}
-			disk_bytenr = le64_to_cpu(old.disk_bytenr);
-			if (disk_bytenr != 0) {
+			if (ret < 0)
+				break;
+
+			leaf = path->nodes[0];
+			fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							start - key.offset);
+
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+
+			extent_offset += start - key.offset;
+			btrfs_set_file_extent_offset(leaf, fi, extent_offset);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							extent_end - start);
+			btrfs_mark_buffer_dirty(leaf);
+
+			if (disk_bytenr > 0) {
 				ret = btrfs_inc_extent_ref(trans, root,
-					   disk_bytenr,
-					   le64_to_cpu(old.disk_num_bytes), 0,
-					   root->root_key.objectid,
-					   key.objectid, key.offset -
-					   le64_to_cpu(old.offset));
+						disk_bytenr, num_bytes, 0,
+						root->root_key.objectid,
+						new_key.objectid,
+						start - extent_offset);
 				BUG_ON(ret);
+				*hint_byte = disk_bytenr;
 			}
+			key.offset = start;
 		}
+		/*
+		 *  | ---- range to drop ----- |
+		 *      | -------- extent -------- |
+		 */
+		if (start <= key.offset && end < extent_end) {
+			BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
 
-		if (found_inline) {
-			u64 mask = root->sectorsize - 1;
-			search_start = (extent_end + mask) & ~mask;
-		} else
-			search_start = extent_end;
-
-		/* truncate existing extent */
-		if (start > key.offset) {
-			u64 new_num;
-			u64 old_num;
-			keep = 1;
-			WARN_ON(start & (root->sectorsize - 1));
-			if (found_extent) {
-				new_num = start - key.offset;
-				old_num = btrfs_file_extent_num_bytes(leaf,
-								      extent);
-				*hint_byte =
-					btrfs_file_extent_disk_bytenr(leaf,
-								      extent);
-				if (btrfs_file_extent_disk_bytenr(leaf,
-								  extent)) {
-					inode_sub_bytes(inode, old_num -
-							new_num);
-				}
-				btrfs_set_file_extent_num_bytes(leaf,
-							extent, new_num);
-				btrfs_mark_buffer_dirty(leaf);
-			} else if (key.offset < inline_limit &&
-				   (end > extent_end) &&
-				   (inline_limit < extent_end)) {
-				u32 new_size;
-				new_size = btrfs_file_extent_calc_inline_size(
-						   inline_limit - key.offset);
-				inode_sub_bytes(inode, extent_end -
-						inline_limit);
-				btrfs_set_file_extent_ram_bytes(leaf, extent,
-							new_size);
-				if (!compression && !encryption) {
-					btrfs_truncate_item(trans, root, path,
-							    new_size, 1);
-				}
+			memcpy(&new_key, &key, sizeof(new_key));
+			new_key.offset = end;
+			btrfs_set_item_key_safe(trans, root, path, &new_key);
+
+			extent_offset += end - key.offset;
+			btrfs_set_file_extent_offset(leaf, fi, extent_offset);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							extent_end - end);
+			btrfs_mark_buffer_dirty(leaf);
+			if (disk_bytenr > 0) {
+				inode_sub_bytes(inode, end - key.offset);
+				*hint_byte = disk_bytenr;
 			}
+			break;
 		}
-		/* delete the entire extent */
-		if (!keep) {
-			if (found_inline)
-				inode_sub_bytes(inode, extent_end -
-						key.offset);
-			ret = btrfs_del_item(trans, root, path);
-			/* TODO update progress marker and return */
-			BUG_ON(ret);
-			extent = NULL;
-			btrfs_release_path(root, path);
-			/* the extent will be freed later */
-		}
-		if (bookend && found_inline && start <= key.offset) {
-			u32 new_size;
-			new_size = btrfs_file_extent_calc_inline_size(
-						   extent_end - end);
-			inode_sub_bytes(inode, end - key.offset);
-			btrfs_set_file_extent_ram_bytes(leaf, extent,
-							new_size);
-			if (!compression && !encryption)
-				ret = btrfs_truncate_item(trans, root, path,
-							  new_size, 0);
-			BUG_ON(ret);
-		}
-		/* create bookend, splitting the extent in two */
-		if (bookend && found_extent) {
-			struct btrfs_key ins;
-			ins.objectid = inode->i_ino;
-			ins.offset = end;
-			btrfs_set_key_type(&ins, BTRFS_EXTENT_DATA_KEY);
 
-			btrfs_release_path(root, path);
-			path->leave_spinning = 1;
-			ret = btrfs_insert_empty_item(trans, root, path, &ins,
-						      sizeof(*extent));
-			BUG_ON(ret);
+		search_start = extent_end;
+		/*
+		 *       | ---- range to drop ----- |
+		 *  | -------- extent -------- |
+		 */
+		if (start > key.offset && end >= extent_end) {
+			BUG_ON(del_nr > 0);
+			BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
 
-			leaf = path->nodes[0];
-			extent = btrfs_item_ptr(leaf, path->slots[0],
-						struct btrfs_file_extent_item);
-			write_extent_buffer(leaf, &old,
-					    (unsigned long)extent, sizeof(old));
-
-			btrfs_set_file_extent_compression(leaf, extent,
-							  compression);
-			btrfs_set_file_extent_encryption(leaf, extent,
-							 encryption);
-			btrfs_set_file_extent_other_encoding(leaf, extent,
-							     other_encoding);
-			btrfs_set_file_extent_offset(leaf, extent,
-				    le64_to_cpu(old.offset) + end - key.offset);
-			WARN_ON(le64_to_cpu(old.num_bytes) <
-				(extent_end - end));
-			btrfs_set_file_extent_num_bytes(leaf, extent,
-							extent_end - end);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							start - key.offset);
+			btrfs_mark_buffer_dirty(leaf);
+			if (disk_bytenr > 0) {
+				inode_sub_bytes(inode, extent_end - start);
+				*hint_byte = disk_bytenr;
+			}
+			if (end == extent_end)
+				break;
 
-			/*
-			 * set the ram bytes to the size of the full extent
-			 * before splitting.  This is a worst case flag,
-			 * but its the best we can do because we don't know
-			 * how splitting affects compression
-			 */
-			btrfs_set_file_extent_ram_bytes(leaf, extent,
-							ram_bytes);
-			btrfs_set_file_extent_type(leaf, extent, found_type);
-
-			btrfs_unlock_up_safe(path, 1);
-			btrfs_mark_buffer_dirty(path->nodes[0]);
-			btrfs_set_lock_blocking(path->nodes[0]);
-
-			path->leave_spinning = 0;
-			btrfs_release_path(root, path);
-			if (disk_bytenr != 0)
-				inode_add_bytes(inode, extent_end - end);
+			path->slots[0]++;
+			goto next_slot;
 		}
 
-		if (found_extent && !keep) {
-			u64 old_disk_bytenr = le64_to_cpu(old.disk_bytenr);
+		/*
+		 *  | ---- range to drop ----- |
+		 *    | ------ extent ------ |
+		 */
+		if (start <= key.offset && end >= extent_end) {
+			if (del_nr == 0) {
+				del_slot = path->slots[0];
+				del_nr = 1;
+			} else {
+				BUG_ON(del_slot + del_nr != path->slots[0]);
+				del_nr++;
+			}
 
-			if (old_disk_bytenr != 0) {
+			if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
 				inode_sub_bytes(inode,
-						le64_to_cpu(old.num_bytes));
+						extent_end - key.offset);
+				extent_end = ALIGN(extent_end,
+						   root->sectorsize);
+			} else if (disk_bytenr > 0) {
 				ret = btrfs_free_extent(trans, root,
-						old_disk_bytenr,
-						le64_to_cpu(old.disk_num_bytes),
-						0, root->root_key.objectid,
+						disk_bytenr, num_bytes, 0,
+						root->root_key.objectid,
 						key.objectid, key.offset -
-						le64_to_cpu(old.offset));
+						extent_offset);
 				BUG_ON(ret);
-				*hint_byte = old_disk_bytenr;
+				inode_sub_bytes(inode,
+						extent_end - key.offset);
+				*hint_byte = disk_bytenr;
 			}
-		}
 
-		if (search_start >= end) {
-			ret = 0;
-			goto out;
+			if (end == extent_end)
+				break;
+
+			if (path->slots[0] + 1 < btrfs_header_nritems(leaf)) {
+				path->slots[0]++;
+				goto next_slot;
+			}
+
+			ret = btrfs_del_items(trans, root, path, del_slot,
+					      del_nr);
+			BUG_ON(ret);
+
+			del_nr = 0;
+			del_slot = 0;
+
+			btrfs_release_path(root, path);
+			continue;
 		}
+
+		BUG_ON(1);
 	}
-out:
-	btrfs_free_path(path);
-	if (locked_end > orig_locked_end) {
-		unlock_extent(&BTRFS_I(inode)->io_tree, orig_locked_end,
-			      locked_end - 1, GFP_NOFS);
+
+	if (del_nr > 0) {
+		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
+		BUG_ON(ret);
 	}
+
+	btrfs_free_path(path);
 	return ret;
 }
 
 static int extent_mergeable(struct extent_buffer *leaf, int slot,
-			    u64 objectid, u64 bytenr, u64 *start, u64 *end)
+			    u64 objectid, u64 bytenr, u64 orig_offset,
+			    u64 *start, u64 *end)
 {
 	struct btrfs_file_extent_item *fi;
 	struct btrfs_key key;
@@ -598,6 +523,7 @@ static int extent_mergeable(struct extent_buffer *leaf, int slot,
 	fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item);
 	if (btrfs_file_extent_type(leaf, fi) != BTRFS_FILE_EXTENT_REG ||
 	    btrfs_file_extent_disk_bytenr(leaf, fi) != bytenr ||
+	    btrfs_file_extent_offset(leaf, fi) != key.offset - orig_offset ||
 	    btrfs_file_extent_compression(leaf, fi) ||
 	    btrfs_file_extent_encryption(leaf, fi) ||
 	    btrfs_file_extent_other_encoding(leaf, fi))
@@ -620,23 +546,24 @@ static int extent_mergeable(struct extent_buffer *leaf, int slot,
  * two or three.
  */
 int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
-			      struct btrfs_root *root,
 			      struct inode *inode, u64 start, u64 end)
 {
+	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct extent_buffer *leaf;
 	struct btrfs_path *path;
 	struct btrfs_file_extent_item *fi;
 	struct btrfs_key key;
+	struct btrfs_key new_key;
 	u64 bytenr;
 	u64 num_bytes;
 	u64 extent_end;
 	u64 orig_offset;
 	u64 other_start;
 	u64 other_end;
-	u64 split = start;
-	u64 locked_end = end;
-	int extent_type;
-	int split_end = 1;
+	u64 split;
+	int del_nr = 0;
+	int del_slot = 0;
+	int recow;
 	int ret;
 
 	btrfs_drop_extent_cache(inode, start, end - 1, 0);
@@ -644,12 +571,11 @@ int btrfs_mark_extent_written(struct btrfs_trans_handle *trans,
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 again:
+	recow = 0;
+	split = start;
 	key.objectid = inode->i_ino;
 	key.type = BTRFS_EXTENT_DATA_KEY;
-	if (split == start)
-		key.offset = split;
-	else
-		key.offset = split - 1;
+	key.offset = split;
 
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
 	if (ret > 0 && path->slots[0] > 0)
@@ -661,159 +587,156 @@ again:
 	       key.type != BTRFS_EXTENT_DATA_KEY);
 	fi = btrfs_item_ptr(leaf, path->slots[0],
 			    struct btrfs_file_extent_item);
-	extent_type = btrfs_file_extent_type(leaf, fi);
-	BUG_ON(extent_type != BTRFS_FILE_EXTENT_PREALLOC);
+	BUG_ON(btrfs_file_extent_type(leaf, fi) !=
+	       BTRFS_FILE_EXTENT_PREALLOC);
 	extent_end = key.offset + btrfs_file_extent_num_bytes(leaf, fi);
 	BUG_ON(key.offset > start || extent_end < end);
 
 	bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
 	num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
 	orig_offset = key.offset - btrfs_file_extent_offset(leaf, fi);
+	memcpy(&new_key, &key, sizeof(new_key));
 
-	if (key.offset == start)
-		split = end;
-
-	if (key.offset == start && extent_end == end) {
-		int del_nr = 0;
-		int del_slot = 0;
-		other_start = end;
-		other_end = 0;
-		if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
-				     bytenr, &other_start, &other_end)) {
-			extent_end = other_end;
-			del_slot = path->slots[0] + 1;
-			del_nr++;
-			ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
-						0, root->root_key.objectid,
-						inode->i_ino, orig_offset);
-			BUG_ON(ret);
-		}
+	if (start == key.offset && end < extent_end) {
 		other_start = 0;
 		other_end = start;
-		if (extent_mergeable(leaf, path->slots[0] - 1, inode->i_ino,
-				     bytenr, &other_start, &other_end)) {
-			key.offset = other_start;
-			del_slot = path->slots[0];
-			del_nr++;
-			ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
-						0, root->root_key.objectid,
-						inode->i_ino, orig_offset);
-			BUG_ON(ret);
-		}
-		split_end = 0;
-		if (del_nr == 0) {
-			btrfs_set_file_extent_type(leaf, fi,
-						   BTRFS_FILE_EXTENT_REG);
-			goto done;
-		}
-
-		fi = btrfs_item_ptr(leaf, del_slot - 1,
-				    struct btrfs_file_extent_item);
-		btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG);
-		btrfs_set_file_extent_num_bytes(leaf, fi,
-						extent_end - key.offset);
-		btrfs_mark_buffer_dirty(leaf);
-
-		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
-		BUG_ON(ret);
-		goto release;
-	} else if (split == start) {
-		if (locked_end < extent_end) {
-			ret = try_lock_extent(&BTRFS_I(inode)->io_tree,
-					locked_end, extent_end - 1, GFP_NOFS);
-			if (!ret) {
-				btrfs_release_path(root, path);
-				lock_extent(&BTRFS_I(inode)->io_tree,
-					locked_end, extent_end - 1, GFP_NOFS);
-				locked_end = extent_end;
-				goto again;
-			}
-			locked_end = extent_end;
+		if (extent_mergeable(leaf, path->slots[0] - 1,
+				     inode->i_ino, bytenr, orig_offset,
+				     &other_start, &other_end)) {
+			new_key.offset = end;
+			btrfs_set_item_key_safe(trans, root, path, &new_key);
+			fi = btrfs_item_ptr(leaf, path->slots[0],
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							extent_end - end);
+			btrfs_set_file_extent_offset(leaf, fi,
+						     end - orig_offset);
+			fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
+					    struct btrfs_file_extent_item);
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							end - other_start);
+			btrfs_mark_buffer_dirty(leaf);
+			goto out;
 		}
-		btrfs_set_file_extent_num_bytes(leaf, fi, split - key.offset);
-	} else  {
-		BUG_ON(key.offset != start);
-		key.offset = split;
-		btrfs_set_file_extent_offset(leaf, fi, key.offset -
-					     orig_offset);
-		btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - split);
-		btrfs_set_item_key_safe(trans, root, path, &key);
-		extent_end = split;
 	}
 
-	if (extent_end == end) {
-		split_end = 0;
-		extent_type = BTRFS_FILE_EXTENT_REG;
-	}
-	if (extent_end == end && split == start) {
+	if (start > key.offset && end == extent_end) {
 		other_start = end;
 		other_end = 0;
-		if (extent_mergeable(leaf, path->slots[0] + 1, inode->i_ino,
-				     bytenr, &other_start, &other_end)) {
-			path->slots[0]++;
+		if (extent_mergeable(leaf, path->slots[0] + 1,
+				     inode->i_ino, bytenr, orig_offset,
+				     &other_start, &other_end)) {
 			fi = btrfs_item_ptr(leaf, path->slots[0],
 					    struct btrfs_file_extent_item);
-			key.offset = split;
-			btrfs_set_item_key_safe(trans, root, path, &key);
-			btrfs_set_file_extent_offset(leaf, fi, key.offset -
-						     orig_offset);
 			btrfs_set_file_extent_num_bytes(leaf, fi,
-							other_end - split);
-			goto done;
-		}
-	}
-	if (extent_end == end && split == end) {
-		other_start = 0;
-		other_end = start;
-		if (extent_mergeable(leaf, path->slots[0] - 1 , inode->i_ino,
-				     bytenr, &other_start, &other_end)) {
-			path->slots[0]--;
+							start - key.offset);
+			path->slots[0]++;
+			new_key.offset = start;
+			btrfs_set_item_key_safe(trans, root, path, &new_key);
+
 			fi = btrfs_item_ptr(leaf, path->slots[0],
 					    struct btrfs_file_extent_item);
-			btrfs_set_file_extent_num_bytes(leaf, fi, extent_end -
-							other_start);
-			goto done;
+			btrfs_set_file_extent_num_bytes(leaf, fi,
+							other_end - start);
+			btrfs_set_file_extent_offset(leaf, fi,
+						     start - orig_offset);
+			btrfs_mark_buffer_dirty(leaf);
+			goto out;
 		}
 	}
 
-	btrfs_mark_buffer_dirty(leaf);
+	while (start > key.offset || end < extent_end) {
+		if (key.offset == start)
+			split = end;
 
-	ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, 0,
-				   root->root_key.objectid,
-				   inode->i_ino, orig_offset);
-	BUG_ON(ret);
-	btrfs_release_path(root, path);
+		new_key.offset = split;
+		ret = btrfs_duplicate_item(trans, root, path, &new_key);
+		if (ret == -EAGAIN) {
+			btrfs_release_path(root, path);
+			goto again;
+		}
+		BUG_ON(ret < 0);
 
-	key.offset = start;
-	ret = btrfs_insert_empty_item(trans, root, path, &key, sizeof(*fi));
-	BUG_ON(ret);
+		leaf = path->nodes[0];
+		fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
+				    struct btrfs_file_extent_item);
+		btrfs_set_file_extent_num_bytes(leaf, fi,
+						split - key.offset);
 
-	leaf = path->nodes[0];
-	fi = btrfs_item_ptr(leaf, path->slots[0],
-			    struct btrfs_file_extent_item);
-	btrfs_set_file_extent_generation(leaf, fi, trans->transid);
-	btrfs_set_file_extent_type(leaf, fi, extent_type);
-	btrfs_set_file_extent_disk_bytenr(leaf, fi, bytenr);
-	btrfs_set_file_extent_disk_num_bytes(leaf, fi, num_bytes);
-	btrfs_set_file_extent_offset(leaf, fi, key.offset - orig_offset);
-	btrfs_set_file_extent_num_bytes(leaf, fi, extent_end - key.offset);
-	btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes);
-	btrfs_set_file_extent_compression(leaf, fi, 0);
-	btrfs_set_file_extent_encryption(leaf, fi, 0);
-	btrfs_set_file_extent_other_encoding(leaf, fi, 0);
-done:
-	btrfs_mark_buffer_dirty(leaf);
-
-release:
-	btrfs_release_path(root, path);
-	if (split_end && split == start) {
-		split = end;
-		goto again;
+		fi = btrfs_item_ptr(leaf, path->slots[0],
+				    struct btrfs_file_extent_item);
+
+		btrfs_set_file_extent_offset(leaf, fi, split - orig_offset);
+		btrfs_set_file_extent_num_bytes(leaf, fi,
+						extent_end - split);
+		btrfs_mark_buffer_dirty(leaf);
+
+		ret = btrfs_inc_extent_ref(trans, root, bytenr, num_bytes, 0,
+					   root->root_key.objectid,
+					   inode->i_ino, orig_offset);
+		BUG_ON(ret);
+
+		if (split == start) {
+			key.offset = start;
+		} else {
+			BUG_ON(start != key.offset);
+			path->slots[0]--;
+			extent_end = end;
+		}
+		recow = 1;
 	}
-	if (locked_end > end) {
-		unlock_extent(&BTRFS_I(inode)->io_tree, end, locked_end - 1,
-			      GFP_NOFS);
+
+	other_start = end;
+	other_end = 0;
+	if (extent_mergeable(leaf, path->slots[0] + 1,
+			     inode->i_ino, bytenr, orig_offset,
+			     &other_start, &other_end)) {
+		if (recow) {
+			btrfs_release_path(root, path);
+			goto again;
+		}
+		extent_end = other_end;
+		del_slot = path->slots[0] + 1;
+		del_nr++;
+		ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
+					0, root->root_key.objectid,
+					inode->i_ino, orig_offset);
+		BUG_ON(ret);
+	}
+	other_start = 0;
+	other_end = start;
+	if (extent_mergeable(leaf, path->slots[0] - 1,
+			     inode->i_ino, bytenr, orig_offset,
+			     &other_start, &other_end)) {
+		if (recow) {
+			btrfs_release_path(root, path);
+			goto again;
+		}
+		key.offset = other_start;
+		del_slot = path->slots[0];
+		del_nr++;
+		ret = btrfs_free_extent(trans, root, bytenr, num_bytes,
+					0, root->root_key.objectid,
+					inode->i_ino, orig_offset);
+		BUG_ON(ret);
 	}
+	fi = btrfs_item_ptr(leaf, path->slots[0],
+			   struct btrfs_file_extent_item);
+	if (del_nr == 0) {
+		btrfs_set_file_extent_type(leaf, fi,
+					   BTRFS_FILE_EXTENT_REG);
+		btrfs_mark_buffer_dirty(leaf);
+	} else {
+		btrfs_set_file_extent_type(leaf, fi,
+					   BTRFS_FILE_EXTENT_REG);
+		btrfs_set_file_extent_num_bytes(leaf, fi,
+						extent_end - key.offset);
+		btrfs_mark_buffer_dirty(leaf);
+
+		ret = btrfs_del_items(trans, root, path, del_slot, del_nr);
+		BUG_ON(ret);
+	}
+out:
 	btrfs_free_path(path);
 	return 0;
 }
@@ -909,7 +832,7 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf,
 	unsigned long last_index;
 	int will_write;
 
-	will_write = ((file->f_flags & O_SYNC) || IS_SYNC(inode) ||
+	will_write = ((file->f_flags & O_DSYNC) || IS_SYNC(inode) ||
 		      (file->f_flags & O_DIRECT));
 
 	nrptrs = min((count + PAGE_CACHE_SIZE - 1) / PAGE_CACHE_SIZE,
@@ -1076,7 +999,7 @@ out_nolock:
 		if (err)
 			num_written = err;
 
-		if ((file->f_flags & O_SYNC) || IS_SYNC(inode)) {
+		if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
 			trans = btrfs_start_transaction(root, 1);
 			ret = btrfs_log_dentry_safe(trans, root,
 						    file->f_dentry);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b3ad168a0bfc..8cd109972fa6 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -88,13 +88,14 @@ static noinline int cow_file_range(struct inode *inode,
 				   u64 start, u64 end, int *page_started,
 				   unsigned long *nr_written, int unlock);
 
-static int btrfs_init_inode_security(struct inode *inode,  struct inode *dir)
+static int btrfs_init_inode_security(struct btrfs_trans_handle *trans,
+				     struct inode *inode,  struct inode *dir)
 {
 	int err;
 
-	err = btrfs_init_acl(inode, dir);
+	err = btrfs_init_acl(trans, inode, dir);
 	if (!err)
-		err = btrfs_xattr_security_init(inode, dir);
+		err = btrfs_xattr_security_init(trans, inode, dir);
 	return err;
 }
 
@@ -188,8 +189,18 @@ static noinline int insert_inline_extent(struct btrfs_trans_handle *trans,
 	btrfs_mark_buffer_dirty(leaf);
 	btrfs_free_path(path);
 
+	/*
+	 * we're an inline extent, so nobody can
+	 * extend the file past i_size without locking
+	 * a page we already have locked.
+	 *
+	 * We must do any isize and inode updates
+	 * before we unlock the pages.  Otherwise we
+	 * could end up racing with unlink.
+	 */
 	BTRFS_I(inode)->disk_i_size = inode->i_size;
 	btrfs_update_inode(trans, root, inode);
+
 	return 0;
 fail:
 	btrfs_free_path(path);
@@ -230,8 +241,7 @@ static noinline int cow_file_range_inline(struct btrfs_trans_handle *trans,
 		return 1;
 	}
 
-	ret = btrfs_drop_extents(trans, root, inode, start,
-				 aligned_end, aligned_end, start,
+	ret = btrfs_drop_extents(trans, inode, start, aligned_end,
 				 &hint_byte, 1);
 	BUG_ON(ret);
 
@@ -416,7 +426,6 @@ again:
 						    start, end,
 						    total_compressed, pages);
 		}
-		btrfs_end_transaction(trans, root);
 		if (ret == 0) {
 			/*
 			 * inline extent creation worked, we don't need
@@ -430,9 +439,11 @@ again:
 			     EXTENT_CLEAR_DELALLOC |
 			     EXTENT_CLEAR_ACCOUNTING |
 			     EXTENT_SET_WRITEBACK | EXTENT_END_WRITEBACK);
-			ret = 0;
+
+			btrfs_end_transaction(trans, root);
 			goto free_pages_out;
 		}
+		btrfs_end_transaction(trans, root);
 	}
 
 	if (will_compress) {
@@ -472,7 +483,8 @@ again:
 		nr_pages_ret = 0;
 
 		/* flag the file so we don't compress in the future */
-		BTRFS_I(inode)->flags |= BTRFS_INODE_NOCOMPRESS;
+		if (!btrfs_test_opt(root, FORCE_COMPRESS))
+			BTRFS_I(inode)->flags |= BTRFS_INODE_NOCOMPRESS;
 	}
 	if (will_compress) {
 		*num_added += 1;
@@ -543,7 +555,6 @@ static noinline int submit_compressed_extents(struct inode *inode,
 	if (list_empty(&async_cow->extents))
 		return 0;
 
-	trans = btrfs_join_transaction(root, 1);
 
 	while (!list_empty(&async_cow->extents)) {
 		async_extent = list_entry(async_cow->extents.next,
@@ -590,19 +601,15 @@ retry:
 		lock_extent(io_tree, async_extent->start,
 			    async_extent->start + async_extent->ram_size - 1,
 			    GFP_NOFS);
-		/*
-		 * here we're doing allocation and writeback of the
-		 * compressed pages
-		 */
-		btrfs_drop_extent_cache(inode, async_extent->start,
-					async_extent->start +
-					async_extent->ram_size - 1, 0);
 
+		trans = btrfs_join_transaction(root, 1);
 		ret = btrfs_reserve_extent(trans, root,
 					   async_extent->compressed_size,
 					   async_extent->compressed_size,
 					   0, alloc_hint,
 					   (u64)-1, &ins, 1);
+		btrfs_end_transaction(trans, root);
+
 		if (ret) {
 			int i;
 			for (i = 0; i < async_extent->nr_pages; i++) {
@@ -618,6 +625,14 @@ retry:
 			goto retry;
 		}
 
+		/*
+		 * here we're doing allocation and writeback of the
+		 * compressed pages
+		 */
+		btrfs_drop_extent_cache(inode, async_extent->start,
+					async_extent->start +
+					async_extent->ram_size - 1, 0);
+
 		em = alloc_extent_map(GFP_NOFS);
 		em->start = async_extent->start;
 		em->len = async_extent->ram_size;
@@ -649,8 +664,6 @@ retry:
 					       BTRFS_ORDERED_COMPRESSED);
 		BUG_ON(ret);
 
-		btrfs_end_transaction(trans, root);
-
 		/*
 		 * clear dirty, set writeback and unlock the pages.
 		 */
@@ -672,13 +685,11 @@ retry:
 				    async_extent->nr_pages);
 
 		BUG_ON(ret);
-		trans = btrfs_join_transaction(root, 1);
 		alloc_hint = ins.objectid + ins.offset;
 		kfree(async_extent);
 		cond_resched();
 	}
 
-	btrfs_end_transaction(trans, root);
 	return 0;
 }
 
@@ -742,6 +753,7 @@ static noinline int cow_file_range(struct inode *inode,
 				     EXTENT_CLEAR_DIRTY |
 				     EXTENT_SET_WRITEBACK |
 				     EXTENT_END_WRITEBACK);
+
 			*nr_written = *nr_written +
 			     (end - start + PAGE_CACHE_SIZE) / PAGE_CACHE_SIZE;
 			*page_started = 1;
@@ -1596,7 +1608,6 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans,
 				       struct inode *inode, u64 file_pos,
 				       u64 disk_bytenr, u64 disk_num_bytes,
 				       u64 num_bytes, u64 ram_bytes,
-				       u64 locked_end,
 				       u8 compression, u8 encryption,
 				       u16 other_encoding, int extent_type)
 {
@@ -1622,9 +1633,8 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans,
 	 * the caller is expected to unpin it and allow it to be merged
 	 * with the others.
 	 */
-	ret = btrfs_drop_extents(trans, root, inode, file_pos,
-				 file_pos + num_bytes, locked_end,
-				 file_pos, &hint, 0);
+	ret = btrfs_drop_extents(trans, inode, file_pos, file_pos + num_bytes,
+				 &hint, 0);
 	BUG_ON(ret);
 
 	ins.objectid = inode->i_ino;
@@ -1730,23 +1740,32 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 		}
 	}
 
-	trans = btrfs_join_transaction(root, 1);
-
 	if (!ordered_extent)
 		ordered_extent = btrfs_lookup_ordered_extent(inode, start);
 	BUG_ON(!ordered_extent);
-	if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags))
-		goto nocow;
+	if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) {
+		BUG_ON(!list_empty(&ordered_extent->list));
+		ret = btrfs_ordered_update_i_size(inode, 0, ordered_extent);
+		if (!ret) {
+			trans = btrfs_join_transaction(root, 1);
+			ret = btrfs_update_inode(trans, root, inode);
+			BUG_ON(ret);
+			btrfs_end_transaction(trans, root);
+		}
+		goto out;
+	}
 
 	lock_extent(io_tree, ordered_extent->file_offset,
 		    ordered_extent->file_offset + ordered_extent->len - 1,
 		    GFP_NOFS);
 
+	trans = btrfs_join_transaction(root, 1);
+
 	if (test_bit(BTRFS_ORDERED_COMPRESSED, &ordered_extent->flags))
 		compressed = 1;
 	if (test_bit(BTRFS_ORDERED_PREALLOC, &ordered_extent->flags)) {
 		BUG_ON(compressed);
-		ret = btrfs_mark_extent_written(trans, root, inode,
+		ret = btrfs_mark_extent_written(trans, inode,
 						ordered_extent->file_offset,
 						ordered_extent->file_offset +
 						ordered_extent->len);
@@ -1758,8 +1777,6 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 						ordered_extent->disk_len,
 						ordered_extent->len,
 						ordered_extent->len,
-						ordered_extent->file_offset +
-						ordered_extent->len,
 						compressed, 0, 0,
 						BTRFS_FILE_EXTENT_REG);
 		unpin_extent_cache(&BTRFS_I(inode)->extent_tree,
@@ -1770,22 +1787,20 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end)
 	unlock_extent(io_tree, ordered_extent->file_offset,
 		    ordered_extent->file_offset + ordered_extent->len - 1,
 		    GFP_NOFS);
-nocow:
 	add_pending_csums(trans, inode, ordered_extent->file_offset,
 			  &ordered_extent->list);
 
-	mutex_lock(&BTRFS_I(inode)->extent_mutex);
-	btrfs_ordered_update_i_size(inode, ordered_extent);
-	btrfs_update_inode(trans, root, inode);
-	btrfs_remove_ordered_extent(inode, ordered_extent);
-	mutex_unlock(&BTRFS_I(inode)->extent_mutex);
-
+	/* this also removes the ordered extent from the tree */
+	btrfs_ordered_update_i_size(inode, 0, ordered_extent);
+	ret = btrfs_update_inode(trans, root, inode);
+	BUG_ON(ret);
+	btrfs_end_transaction(trans, root);
+out:
 	/* once for us */
 	btrfs_put_ordered_extent(ordered_extent);
 	/* once for the tree */
 	btrfs_put_ordered_extent(ordered_extent);
 
-	btrfs_end_transaction(trans, root);
 	return 0;
 }
 
@@ -2008,6 +2023,54 @@ zeroit:
 	return -EIO;
 }
 
+struct delayed_iput {
+	struct list_head list;
+	struct inode *inode;
+};
+
+void btrfs_add_delayed_iput(struct inode *inode)
+{
+	struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info;
+	struct delayed_iput *delayed;
+
+	if (atomic_add_unless(&inode->i_count, -1, 1))
+		return;
+
+	delayed = kmalloc(sizeof(*delayed), GFP_NOFS | __GFP_NOFAIL);
+	delayed->inode = inode;
+
+	spin_lock(&fs_info->delayed_iput_lock);
+	list_add_tail(&delayed->list, &fs_info->delayed_iputs);
+	spin_unlock(&fs_info->delayed_iput_lock);
+}
+
+void btrfs_run_delayed_iputs(struct btrfs_root *root)
+{
+	LIST_HEAD(list);
+	struct btrfs_fs_info *fs_info = root->fs_info;
+	struct delayed_iput *delayed;
+	int empty;
+
+	spin_lock(&fs_info->delayed_iput_lock);
+	empty = list_empty(&fs_info->delayed_iputs);
+	spin_unlock(&fs_info->delayed_iput_lock);
+	if (empty)
+		return;
+
+	down_read(&root->fs_info->cleanup_work_sem);
+	spin_lock(&fs_info->delayed_iput_lock);
+	list_splice_init(&fs_info->delayed_iputs, &list);
+	spin_unlock(&fs_info->delayed_iput_lock);
+
+	while (!list_empty(&list)) {
+		delayed = list_entry(list.next, struct delayed_iput, list);
+		list_del(&delayed->list);
+		iput(delayed->inode);
+		kfree(delayed);
+	}
+	up_read(&root->fs_info->cleanup_work_sem);
+}
+
 /*
  * This creates an orphan entry for the given inode in case something goes
  * wrong in the middle of an unlink/truncate.
@@ -2080,16 +2143,17 @@ void btrfs_orphan_cleanup(struct btrfs_root *root)
 	struct inode *inode;
 	int ret = 0, nr_unlink = 0, nr_truncate = 0;
 
-	path = btrfs_alloc_path();
-	if (!path)
+	if (!xchg(&root->clean_orphans, 0))
 		return;
+
+	path = btrfs_alloc_path();
+	BUG_ON(!path);
 	path->reada = -1;
 
 	key.objectid = BTRFS_ORPHAN_OBJECTID;
 	btrfs_set_key_type(&key, BTRFS_ORPHAN_ITEM_KEY);
 	key.offset = (u64)-1;
 
-
 	while (1) {
 		ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
 		if (ret < 0) {
@@ -2834,37 +2898,40 @@ out:
  * min_type is the minimum key type to truncate down to.  If set to 0, this
  * will kill all the items on this inode, including the INODE_ITEM_KEY.
  */
-noinline int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
-					struct btrfs_root *root,
-					struct inode *inode,
-					u64 new_size, u32 min_type)
+int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
+			       struct btrfs_root *root,
+			       struct inode *inode,
+			       u64 new_size, u32 min_type)
 {
-	int ret;
 	struct btrfs_path *path;
-	struct btrfs_key key;
-	struct btrfs_key found_key;
-	u32 found_type = (u8)-1;
 	struct extent_buffer *leaf;
 	struct btrfs_file_extent_item *fi;
+	struct btrfs_key key;
+	struct btrfs_key found_key;
 	u64 extent_start = 0;
 	u64 extent_num_bytes = 0;
 	u64 extent_offset = 0;
 	u64 item_end = 0;
+	u64 mask = root->sectorsize - 1;
+	u32 found_type = (u8)-1;
 	int found_extent;
 	int del_item;
 	int pending_del_nr = 0;
 	int pending_del_slot = 0;
 	int extent_type = -1;
 	int encoding;
-	u64 mask = root->sectorsize - 1;
+	int ret;
+	int err = 0;
+
+	BUG_ON(new_size > 0 && min_type != BTRFS_EXTENT_DATA_KEY);
 
 	if (root->ref_cows)
 		btrfs_drop_extent_cache(inode, new_size & (~mask), (u64)-1, 0);
+
 	path = btrfs_alloc_path();
 	BUG_ON(!path);
 	path->reada = -1;
 
-	/* FIXME, add redo link to tree so we don't leak on crash */
 	key.objectid = inode->i_ino;
 	key.offset = (u64)-1;
 	key.type = (u8)-1;
@@ -2872,17 +2939,17 @@ noinline int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
 search_again:
 	path->leave_spinning = 1;
 	ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
-	if (ret < 0)
-		goto error;
+	if (ret < 0) {
+		err = ret;
+		goto out;
+	}
 
 	if (ret > 0) {
 		/* there are no items in the tree for us to truncate, we're
 		 * done
 		 */
-		if (path->slots[0] == 0) {
-			ret = 0;
-			goto error;
-		}
+		if (path->slots[0] == 0)
+			goto out;
 		path->slots[0]--;
 	}
 
@@ -2917,28 +2984,17 @@ search_again:
 			}
 			item_end--;
 		}
-		if (item_end < new_size) {
-			if (found_type == BTRFS_DIR_ITEM_KEY)
-				found_type = BTRFS_INODE_ITEM_KEY;
-			else if (found_type == BTRFS_EXTENT_ITEM_KEY)
-				found_type = BTRFS_EXTENT_DATA_KEY;
-			else if (found_type == BTRFS_EXTENT_DATA_KEY)
-				found_type = BTRFS_XATTR_ITEM_KEY;
-			else if (found_type == BTRFS_XATTR_ITEM_KEY)
-				found_type = BTRFS_INODE_REF_KEY;
-			else if (found_type)
-				found_type--;
-			else
+		if (found_type > min_type) {
+			del_item = 1;
+		} else {
+			if (item_end < new_size)
 				break;
-			btrfs_set_key_type(&key, found_type);
-			goto next;
+			if (found_key.offset >= new_size)
+				del_item = 1;
+			else
+				del_item = 0;
 		}
-		if (found_key.offset >= new_size)
-			del_item = 1;
-		else
-			del_item = 0;
 		found_extent = 0;
-
 		/* FIXME, shrink the extent if the ref count is only 1 */
 		if (found_type != BTRFS_EXTENT_DATA_KEY)
 			goto delete;
@@ -3025,42 +3081,36 @@ delete:
 						inode->i_ino, extent_offset);
 			BUG_ON(ret);
 		}
-next:
-		if (path->slots[0] == 0) {
-			if (pending_del_nr)
-				goto del_pending;
-			btrfs_release_path(root, path);
-			if (found_type == BTRFS_INODE_ITEM_KEY)
-				break;
-			goto search_again;
-		}
 
-		path->slots[0]--;
-		if (pending_del_nr &&
-		    path->slots[0] + 1 != pending_del_slot) {
-			struct btrfs_key debug;
-del_pending:
-			btrfs_item_key_to_cpu(path->nodes[0], &debug,
-					      pending_del_slot);
-			ret = btrfs_del_items(trans, root, path,
-					      pending_del_slot,
-					      pending_del_nr);
-			BUG_ON(ret);
-			pending_del_nr = 0;
+		if (found_type == BTRFS_INODE_ITEM_KEY)
+			break;
+
+		if (path->slots[0] == 0 ||
+		    path->slots[0] != pending_del_slot) {
+			if (root->ref_cows) {
+				err = -EAGAIN;
+				goto out;
+			}
+			if (pending_del_nr) {
+				ret = btrfs_del_items(trans, root, path,
+						pending_del_slot,
+						pending_del_nr);
+				BUG_ON(ret);
+				pending_del_nr = 0;
+			}
 			btrfs_release_path(root, path);
-			if (found_type == BTRFS_INODE_ITEM_KEY)
-				break;
 			goto search_again;
+		} else {
+			path->slots[0]--;
 		}
 	}
-	ret = 0;
-error:
+out:
 	if (pending_del_nr) {
 		ret = btrfs_del_items(trans, root, path, pending_del_slot,
 				      pending_del_nr);
 	}
 	btrfs_free_path(path);
-	return ret;
+	return err;
 }
 
 /*
@@ -3180,10 +3230,6 @@ int btrfs_cont_expand(struct inode *inode, loff_t size)
 	if (size <= hole_start)
 		return 0;
 
-	err = btrfs_truncate_page(inode->i_mapping, inode->i_size);
-	if (err)
-		return err;
-
 	while (1) {
 		struct btrfs_ordered_extent *ordered;
 		btrfs_wait_ordered_range(inode, hole_start,
@@ -3196,9 +3242,6 @@ int btrfs_cont_expand(struct inode *inode, loff_t size)
 		btrfs_put_ordered_extent(ordered);
 	}
 
-	trans = btrfs_start_transaction(root, 1);
-	btrfs_set_trans_block_group(trans, inode);
-
 	cur_offset = hole_start;
 	while (1) {
 		em = btrfs_get_extent(inode, NULL, 0, cur_offset,
@@ -3206,40 +3249,120 @@ int btrfs_cont_expand(struct inode *inode, loff_t size)
 		BUG_ON(IS_ERR(em) || !em);
 		last_byte = min(extent_map_end(em), block_end);
 		last_byte = (last_byte + mask) & ~mask;
-		if (test_bit(EXTENT_FLAG_VACANCY, &em->flags)) {
+		if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) {
 			u64 hint_byte = 0;
 			hole_size = last_byte - cur_offset;
-			err = btrfs_drop_extents(trans, root, inode,
-						 cur_offset,
-						 cur_offset + hole_size,
-						 block_end,
-						 cur_offset, &hint_byte, 1);
-			if (err)
-				break;
 
-			err = btrfs_reserve_metadata_space(root, 1);
+			err = btrfs_reserve_metadata_space(root, 2);
 			if (err)
 				break;
 
+			trans = btrfs_start_transaction(root, 1);
+			btrfs_set_trans_block_group(trans, inode);
+
+			err = btrfs_drop_extents(trans, inode, cur_offset,
+						 cur_offset + hole_size,
+						 &hint_byte, 1);
+			BUG_ON(err);
+
 			err = btrfs_insert_file_extent(trans, root,
 					inode->i_ino, cur_offset, 0,
 					0, hole_size, 0, hole_size,
 					0, 0, 0);
+			BUG_ON(err);
+
 			btrfs_drop_extent_cache(inode, hole_start,
 					last_byte - 1, 0);
-			btrfs_unreserve_metadata_space(root, 1);
+
+			btrfs_end_transaction(trans, root);
+			btrfs_unreserve_metadata_space(root, 2);
 		}
 		free_extent_map(em);
 		cur_offset = last_byte;
-		if (err || cur_offset >= block_end)
+		if (cur_offset >= block_end)
 			break;
 	}
 
-	btrfs_end_transaction(trans, root);
 	unlock_extent(io_tree, hole_start, block_end - 1, GFP_NOFS);
 	return err;
 }
 
+static int btrfs_setattr_size(struct inode *inode, struct iattr *attr)
+{
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	struct btrfs_trans_handle *trans;
+	unsigned long nr;
+	int ret;
+
+	if (attr->ia_size == inode->i_size)
+		return 0;
+
+	if (attr->ia_size > inode->i_size) {
+		unsigned long limit;
+		limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur;
+		if (attr->ia_size > inode->i_sb->s_maxbytes)
+			return -EFBIG;
+		if (limit != RLIM_INFINITY && attr->ia_size > limit) {
+			send_sig(SIGXFSZ, current, 0);
+			return -EFBIG;
+		}
+	}
+
+	ret = btrfs_reserve_metadata_space(root, 1);
+	if (ret)
+		return ret;
+
+	trans = btrfs_start_transaction(root, 1);
+	btrfs_set_trans_block_group(trans, inode);
+
+	ret = btrfs_orphan_add(trans, inode);
+	BUG_ON(ret);
+
+	nr = trans->blocks_used;
+	btrfs_end_transaction(trans, root);
+	btrfs_unreserve_metadata_space(root, 1);
+	btrfs_btree_balance_dirty(root, nr);
+
+	if (attr->ia_size > inode->i_size) {
+		ret = btrfs_cont_expand(inode, attr->ia_size);
+		if (ret) {
+			btrfs_truncate(inode);
+			return ret;
+		}
+
+		i_size_write(inode, attr->ia_size);
+		btrfs_ordered_update_i_size(inode, inode->i_size, NULL);
+
+		trans = btrfs_start_transaction(root, 1);
+		btrfs_set_trans_block_group(trans, inode);
+
+		ret = btrfs_update_inode(trans, root, inode);
+		BUG_ON(ret);
+		if (inode->i_nlink > 0) {
+			ret = btrfs_orphan_del(trans, inode);
+			BUG_ON(ret);
+		}
+		nr = trans->blocks_used;
+		btrfs_end_transaction(trans, root);
+		btrfs_btree_balance_dirty(root, nr);
+		return 0;
+	}
+
+	/*
+	 * We're truncating a file that used to have good data down to
+	 * zero. Make sure it gets into the ordered flush list so that
+	 * any new writes get down to disk quickly.
+	 */
+	if (attr->ia_size == 0)
+		BTRFS_I(inode)->ordered_data_close = 1;
+
+	/* we don't support swapfiles, so vmtruncate shouldn't fail */
+	ret = vmtruncate(inode, attr->ia_size);
+	BUG_ON(ret);
+
+	return 0;
+}
+
 static int btrfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = dentry->d_inode;
@@ -3250,23 +3373,14 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr)
 		return err;
 
 	if (S_ISREG(inode->i_mode) && (attr->ia_valid & ATTR_SIZE)) {
-		if (attr->ia_size > inode->i_size) {
-			err = btrfs_cont_expand(inode, attr->ia_size);
-			if (err)
-				return err;
-		} else if (inode->i_size > 0 &&
-			   attr->ia_size == 0) {
-
-			/* we're truncating a file that used to have good
-			 * data down to zero.  Make sure it gets into
-			 * the ordered flush list so that any new writes
-			 * get down to disk quickly.
-			 */
-			BTRFS_I(inode)->ordered_data_close = 1;
-		}
+		err = btrfs_setattr_size(inode, attr);
+		if (err)
+			return err;
 	}
+	attr->ia_valid &= ~ATTR_SIZE;
 
-	err = inode_setattr(inode, attr);
+	if (attr->ia_valid)
+		err = inode_setattr(inode, attr);
 
 	if (!err && ((attr->ia_valid & ATTR_MODE)))
 		err = btrfs_acl_chmod(inode);
@@ -3287,36 +3401,43 @@ void btrfs_delete_inode(struct inode *inode)
 	}
 	btrfs_wait_ordered_range(inode, 0, (u64)-1);
 
+	if (root->fs_info->log_root_recovering) {
+		BUG_ON(!list_empty(&BTRFS_I(inode)->i_orphan));
+		goto no_delete;
+	}
+
 	if (inode->i_nlink > 0) {
 		BUG_ON(btrfs_root_refs(&root->root_item) != 0);
 		goto no_delete;
 	}
 
 	btrfs_i_size_write(inode, 0);
-	trans = btrfs_join_transaction(root, 1);
 
-	btrfs_set_trans_block_group(trans, inode);
-	ret = btrfs_truncate_inode_items(trans, root, inode, inode->i_size, 0);
-	if (ret) {
-		btrfs_orphan_del(NULL, inode);
-		goto no_delete_lock;
-	}
+	while (1) {
+		trans = btrfs_start_transaction(root, 1);
+		btrfs_set_trans_block_group(trans, inode);
+		ret = btrfs_truncate_inode_items(trans, root, inode, 0, 0);
 
-	btrfs_orphan_del(trans, inode);
+		if (ret != -EAGAIN)
+			break;
 
-	nr = trans->blocks_used;
-	clear_inode(inode);
+		nr = trans->blocks_used;
+		btrfs_end_transaction(trans, root);
+		trans = NULL;
+		btrfs_btree_balance_dirty(root, nr);
+	}
 
-	btrfs_end_transaction(trans, root);
-	btrfs_btree_balance_dirty(root, nr);
-	return;
+	if (ret == 0) {
+		ret = btrfs_orphan_del(trans, inode);
+		BUG_ON(ret);
+	}
 
-no_delete_lock:
 	nr = trans->blocks_used;
 	btrfs_end_transaction(trans, root);
 	btrfs_btree_balance_dirty(root, nr);
 no_delete:
 	clear_inode(inode);
+	return;
 }
 
 /*
@@ -3569,7 +3690,6 @@ static noinline void init_btrfs_i(struct inode *inode)
 	INIT_LIST_HEAD(&BTRFS_I(inode)->ordered_operations);
 	RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node);
 	btrfs_ordered_inode_tree_init(&BTRFS_I(inode)->ordered_tree);
-	mutex_init(&BTRFS_I(inode)->extent_mutex);
 	mutex_init(&BTRFS_I(inode)->log_mutex);
 }
 
@@ -3695,6 +3815,13 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry)
 	}
 	srcu_read_unlock(&root->fs_info->subvol_srcu, index);
 
+	if (root != sub_root) {
+		down_read(&root->fs_info->cleanup_work_sem);
+		if (!(inode->i_sb->s_flags & MS_RDONLY))
+			btrfs_orphan_cleanup(sub_root);
+		up_read(&root->fs_info->cleanup_work_sem);
+	}
+
 	return inode;
 }
 
@@ -3869,7 +3996,11 @@ skip:
 
 	/* Reached end of directory/root. Bump pos past the last item. */
 	if (key_type == BTRFS_DIR_INDEX_KEY)
-		filp->f_pos = INT_LIMIT(off_t);
+		/*
+		 * 32-bit glibc will use getdents64, but then strtol -
+		 * so the last number we can serve is this.
+		 */
+		filp->f_pos = 0x7fffffff;
 	else
 		filp->f_pos++;
 nopos:
@@ -4219,7 +4350,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
 	if (IS_ERR(inode))
 		goto out_unlock;
 
-	err = btrfs_init_inode_security(inode, dir);
+	err = btrfs_init_inode_security(trans, inode, dir);
 	if (err) {
 		drop_inode = 1;
 		goto out_unlock;
@@ -4290,7 +4421,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
 	if (IS_ERR(inode))
 		goto out_unlock;
 
-	err = btrfs_init_inode_security(inode, dir);
+	err = btrfs_init_inode_security(trans, inode, dir);
 	if (err) {
 		drop_inode = 1;
 		goto out_unlock;
@@ -4336,6 +4467,10 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 	if (inode->i_nlink == 0)
 		return -ENOENT;
 
+	/* do not allow sys_link's with other subvols of the same device */
+	if (root->objectid != BTRFS_I(inode)->root->objectid)
+		return -EPERM;
+
 	/*
 	 * 1 item for inode ref
 	 * 2 items for dir items
@@ -4423,7 +4558,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 
 	drop_on_err = 1;
 
-	err = btrfs_init_inode_security(inode, dir);
+	err = btrfs_init_inode_security(trans, inode, dir);
 	if (err)
 		goto out_fail;
 
@@ -5074,17 +5209,20 @@ static void btrfs_truncate(struct inode *inode)
 	unsigned long nr;
 	u64 mask = root->sectorsize - 1;
 
-	if (!S_ISREG(inode->i_mode))
-		return;
-	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
+	if (!S_ISREG(inode->i_mode)) {
+		WARN_ON(1);
 		return;
+	}
 
 	ret = btrfs_truncate_page(inode->i_mapping, inode->i_size);
 	if (ret)
 		return;
+
 	btrfs_wait_ordered_range(inode, inode->i_size & (~mask), (u64)-1);
+	btrfs_ordered_update_i_size(inode, inode->i_size, NULL);
 
 	trans = btrfs_start_transaction(root, 1);
+	btrfs_set_trans_block_group(trans, inode);
 
 	/*
 	 * setattr is responsible for setting the ordered_data_close flag,
@@ -5106,21 +5244,32 @@ static void btrfs_truncate(struct inode *inode)
 	if (inode->i_size == 0 && BTRFS_I(inode)->ordered_data_close)
 		btrfs_add_ordered_operation(trans, root, inode);
 
-	btrfs_set_trans_block_group(trans, inode);
-	btrfs_i_size_write(inode, inode->i_size);
+	while (1) {
+		ret = btrfs_truncate_inode_items(trans, root, inode,
+						 inode->i_size,
+						 BTRFS_EXTENT_DATA_KEY);
+		if (ret != -EAGAIN)
+			break;
 
-	ret = btrfs_orphan_add(trans, inode);
-	if (ret)
-		goto out;
-	/* FIXME, add redo link to tree so we don't leak on crash */
-	ret = btrfs_truncate_inode_items(trans, root, inode, inode->i_size,
-				      BTRFS_EXTENT_DATA_KEY);
-	btrfs_update_inode(trans, root, inode);
+		ret = btrfs_update_inode(trans, root, inode);
+		BUG_ON(ret);
+
+		nr = trans->blocks_used;
+		btrfs_end_transaction(trans, root);
+		btrfs_btree_balance_dirty(root, nr);
+
+		trans = btrfs_start_transaction(root, 1);
+		btrfs_set_trans_block_group(trans, inode);
+	}
 
-	ret = btrfs_orphan_del(trans, inode);
+	if (ret == 0 && inode->i_nlink > 0) {
+		ret = btrfs_orphan_del(trans, inode);
+		BUG_ON(ret);
+	}
+
+	ret = btrfs_update_inode(trans, root, inode);
 	BUG_ON(ret);
 
-out:
 	nr = trans->blocks_used;
 	ret = btrfs_end_transaction_throttle(trans, root);
 	BUG_ON(ret);
@@ -5217,9 +5366,9 @@ void btrfs_destroy_inode(struct inode *inode)
 
 	spin_lock(&root->list_lock);
 	if (!list_empty(&BTRFS_I(inode)->i_orphan)) {
-		printk(KERN_ERR "BTRFS: inode %lu: inode still on the orphan"
-		       " list\n", inode->i_ino);
-		dump_stack();
+		printk(KERN_INFO "BTRFS: inode %lu still on the orphan list\n",
+		       inode->i_ino);
+		list_del_init(&BTRFS_I(inode)->i_orphan);
 	}
 	spin_unlock(&root->list_lock);
 
@@ -5476,7 +5625,7 @@ out_fail:
  * some fairly slow code that needs optimization. This walks the list
  * of all the inodes with pending delalloc and forces them to disk.
  */
-int btrfs_start_delalloc_inodes(struct btrfs_root *root)
+int btrfs_start_delalloc_inodes(struct btrfs_root *root, int delay_iput)
 {
 	struct list_head *head = &root->fs_info->delalloc_inodes;
 	struct btrfs_inode *binode;
@@ -5495,7 +5644,10 @@ int btrfs_start_delalloc_inodes(struct btrfs_root *root)
 		spin_unlock(&root->fs_info->delalloc_lock);
 		if (inode) {
 			filemap_flush(inode->i_mapping);
-			iput(inode);
+			if (delay_iput)
+				btrfs_add_delayed_iput(inode);
+			else
+				iput(inode);
 		}
 		cond_resched();
 		spin_lock(&root->fs_info->delalloc_lock);
@@ -5569,7 +5721,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 	if (IS_ERR(inode))
 		goto out_unlock;
 
-	err = btrfs_init_inode_security(inode, dir);
+	err = btrfs_init_inode_security(trans, inode, dir);
 	if (err) {
 		drop_inode = 1;
 		goto out_unlock;
@@ -5641,57 +5793,75 @@ out_fail:
 	return err;
 }
 
-static int prealloc_file_range(struct btrfs_trans_handle *trans,
-			       struct inode *inode, u64 start, u64 end,
-			       u64 locked_end, u64 alloc_hint, int mode)
+static int prealloc_file_range(struct inode *inode, u64 start, u64 end,
+			u64 alloc_hint, int mode, loff_t actual_len)
 {
+	struct btrfs_trans_handle *trans;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
 	struct btrfs_key ins;
 	u64 alloc_size;
 	u64 cur_offset = start;
 	u64 num_bytes = end - start;
 	int ret = 0;
+	u64 i_size;
 
 	while (num_bytes > 0) {
 		alloc_size = min(num_bytes, root->fs_info->max_extent);
 
-		ret = btrfs_reserve_metadata_space(root, 1);
-		if (ret)
-			goto out;
+		trans = btrfs_start_transaction(root, 1);
 
 		ret = btrfs_reserve_extent(trans, root, alloc_size,
 					   root->sectorsize, 0, alloc_hint,
 					   (u64)-1, &ins, 1);
 		if (ret) {
 			WARN_ON(1);
-			goto out;
+			goto stop_trans;
+		}
+
+		ret = btrfs_reserve_metadata_space(root, 3);
+		if (ret) {
+			btrfs_free_reserved_extent(root, ins.objectid,
+						   ins.offset);
+			goto stop_trans;
 		}
+
 		ret = insert_reserved_file_extent(trans, inode,
 						  cur_offset, ins.objectid,
 						  ins.offset, ins.offset,
-						  ins.offset, locked_end,
-						  0, 0, 0,
+						  ins.offset, 0, 0, 0,
 						  BTRFS_FILE_EXTENT_PREALLOC);
 		BUG_ON(ret);
 		btrfs_drop_extent_cache(inode, cur_offset,
 					cur_offset + ins.offset -1, 0);
+
 		num_bytes -= ins.offset;
 		cur_offset += ins.offset;
 		alloc_hint = ins.objectid + ins.offset;
-		btrfs_unreserve_metadata_space(root, 1);
-	}
-out:
-	if (cur_offset > start) {
+
 		inode->i_ctime = CURRENT_TIME;
 		BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC;
 		if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-		    cur_offset > i_size_read(inode))
-			btrfs_i_size_write(inode, cur_offset);
+		    cur_offset > inode->i_size) {
+			if (cur_offset > actual_len)
+				i_size  = actual_len;
+			else
+				i_size = cur_offset;
+			i_size_write(inode, i_size);
+			btrfs_ordered_update_i_size(inode, i_size, NULL);
+		}
+
 		ret = btrfs_update_inode(trans, root, inode);
 		BUG_ON(ret);
+
+		btrfs_end_transaction(trans, root);
+		btrfs_unreserve_metadata_space(root, 3);
 	}
+	return ret;
 
+stop_trans:
+	btrfs_end_transaction(trans, root);
 	return ret;
+
 }
 
 static long btrfs_fallocate(struct inode *inode, int mode,
@@ -5705,8 +5875,6 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 	u64 locked_end;
 	u64 mask = BTRFS_I(inode)->root->sectorsize - 1;
 	struct extent_map *em;
-	struct btrfs_trans_handle *trans;
-	struct btrfs_root *root;
 	int ret;
 
 	alloc_start = offset & ~mask;
@@ -5725,9 +5893,7 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 			goto out;
 	}
 
-	root = BTRFS_I(inode)->root;
-
-	ret = btrfs_check_data_free_space(root, inode,
+	ret = btrfs_check_data_free_space(BTRFS_I(inode)->root, inode,
 					  alloc_end - alloc_start);
 	if (ret)
 		goto out;
@@ -5736,12 +5902,6 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 	while (1) {
 		struct btrfs_ordered_extent *ordered;
 
-		trans = btrfs_start_transaction(BTRFS_I(inode)->root, 1);
-		if (!trans) {
-			ret = -EIO;
-			goto out_free;
-		}
-
 		/* the extent lock is ordered inside the running
 		 * transaction
 		 */
@@ -5755,8 +5915,6 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 			btrfs_put_ordered_extent(ordered);
 			unlock_extent(&BTRFS_I(inode)->io_tree,
 				      alloc_start, locked_end, GFP_NOFS);
-			btrfs_end_transaction(trans, BTRFS_I(inode)->root);
-
 			/*
 			 * we can't wait on the range with the transaction
 			 * running or with the extent lock held
@@ -5777,10 +5935,12 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 		BUG_ON(IS_ERR(em) || !em);
 		last_byte = min(extent_map_end(em), alloc_end);
 		last_byte = (last_byte + mask) & ~mask;
-		if (em->block_start == EXTENT_MAP_HOLE) {
-			ret = prealloc_file_range(trans, inode, cur_offset,
-					last_byte, locked_end + 1,
-					alloc_hint, mode);
+		if (em->block_start == EXTENT_MAP_HOLE ||
+		    (cur_offset >= inode->i_size &&
+		     !test_bit(EXTENT_FLAG_PREALLOC, &em->flags))) {
+			ret = prealloc_file_range(inode,
+						  cur_offset, last_byte,
+						alloc_hint, mode, offset+len);
 			if (ret < 0) {
 				free_extent_map(em);
 				break;
@@ -5799,9 +5959,8 @@ static long btrfs_fallocate(struct inode *inode, int mode,
 	unlock_extent(&BTRFS_I(inode)->io_tree, alloc_start, locked_end,
 		      GFP_NOFS);
 
-	btrfs_end_transaction(trans, BTRFS_I(inode)->root);
-out_free:
-	btrfs_free_reserved_data_space(root, inode, alloc_end - alloc_start);
+	btrfs_free_reserved_data_space(BTRFS_I(inode)->root, inode,
+				       alloc_end - alloc_start);
 out:
 	mutex_unlock(&inode->i_mutex);
 	return ret;
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index cdbb054102b9..645a17927a8f 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -237,7 +237,6 @@ static noinline int create_subvol(struct btrfs_root *root,
 	u64 objectid;
 	u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
 	u64 index = 0;
-	unsigned long nr = 1;
 
 	/*
 	 * 1 - inode item
@@ -290,7 +289,7 @@ static noinline int create_subvol(struct btrfs_root *root,
 	btrfs_set_root_generation(&root_item, trans->transid);
 	btrfs_set_root_level(&root_item, 0);
 	btrfs_set_root_refs(&root_item, 1);
-	btrfs_set_root_used(&root_item, 0);
+	btrfs_set_root_used(&root_item, leaf->len);
 	btrfs_set_root_last_snapshot(&root_item, 0);
 
 	memset(&root_item.drop_progress, 0, sizeof(root_item.drop_progress));
@@ -342,24 +341,21 @@ static noinline int create_subvol(struct btrfs_root *root,
 
 	d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
 fail:
-	nr = trans->blocks_used;
 	err = btrfs_commit_transaction(trans, root);
 	if (err && !ret)
 		ret = err;
 
 	btrfs_unreserve_metadata_space(root, 6);
-	btrfs_btree_balance_dirty(root, nr);
 	return ret;
 }
 
 static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 			   char *name, int namelen)
 {
+	struct inode *inode;
 	struct btrfs_pending_snapshot *pending_snapshot;
 	struct btrfs_trans_handle *trans;
-	int ret = 0;
-	int err;
-	unsigned long nr = 0;
+	int ret;
 
 	if (!root->ref_cows)
 		return -EINVAL;
@@ -372,20 +368,20 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 	 */
 	ret = btrfs_reserve_metadata_space(root, 6);
 	if (ret)
-		goto fail_unlock;
+		goto fail;
 
 	pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
 	if (!pending_snapshot) {
 		ret = -ENOMEM;
 		btrfs_unreserve_metadata_space(root, 6);
-		goto fail_unlock;
+		goto fail;
 	}
 	pending_snapshot->name = kmalloc(namelen + 1, GFP_NOFS);
 	if (!pending_snapshot->name) {
 		ret = -ENOMEM;
 		kfree(pending_snapshot);
 		btrfs_unreserve_metadata_space(root, 6);
-		goto fail_unlock;
+		goto fail;
 	}
 	memcpy(pending_snapshot->name, name, namelen);
 	pending_snapshot->name[namelen] = '\0';
@@ -395,10 +391,19 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
 	pending_snapshot->root = root;
 	list_add(&pending_snapshot->list,
 		 &trans->transaction->pending_snapshots);
-	err = btrfs_commit_transaction(trans, root);
+	ret = btrfs_commit_transaction(trans, root);
+	BUG_ON(ret);
+	btrfs_unreserve_metadata_space(root, 6);
 
-fail_unlock:
-	btrfs_btree_balance_dirty(root, nr);
+	inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
+	if (IS_ERR(inode)) {
+		ret = PTR_ERR(inode);
+		goto fail;
+	}
+	BUG_ON(!inode);
+	d_instantiate(dentry, inode);
+	ret = 0;
+fail:
 	return ret;
 }
 
@@ -1027,8 +1032,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
 	BUG_ON(!trans);
 
 	/* punch hole in destination first */
-	btrfs_drop_extents(trans, root, inode, off, off + len,
-			   off + len, 0, &hint_byte, 1);
+	btrfs_drop_extents(trans, inode, off, off + len, &hint_byte, 1);
 
 	/* clone data */
 	key.objectid = src->i_ino;
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 5799bc46a309..5c2a9e78a949 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -291,16 +291,16 @@ int btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry)
 
 /*
  * remove an ordered extent from the tree.  No references are dropped
- * but, anyone waiting on this extent is woken up.
+ * and you must wake_up entry->wait.  You must hold the tree mutex
+ * while you call this function.
  */
-int btrfs_remove_ordered_extent(struct inode *inode,
+static int __btrfs_remove_ordered_extent(struct inode *inode,
 				struct btrfs_ordered_extent *entry)
 {
 	struct btrfs_ordered_inode_tree *tree;
 	struct rb_node *node;
 
 	tree = &BTRFS_I(inode)->ordered_tree;
-	mutex_lock(&tree->mutex);
 	node = &entry->rb_node;
 	rb_erase(node, &tree->tree);
 	tree->last = NULL;
@@ -326,16 +326,34 @@ int btrfs_remove_ordered_extent(struct inode *inode,
 	}
 	spin_unlock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
 
+	return 0;
+}
+
+/*
+ * remove an ordered extent from the tree.  No references are dropped
+ * but any waiters are woken.
+ */
+int btrfs_remove_ordered_extent(struct inode *inode,
+				struct btrfs_ordered_extent *entry)
+{
+	struct btrfs_ordered_inode_tree *tree;
+	int ret;
+
+	tree = &BTRFS_I(inode)->ordered_tree;
+	mutex_lock(&tree->mutex);
+	ret = __btrfs_remove_ordered_extent(inode, entry);
 	mutex_unlock(&tree->mutex);
 	wake_up(&entry->wait);
-	return 0;
+
+	return ret;
 }
 
 /*
  * wait for all the ordered extents in a root.  This is done when balancing
  * space between drives.
  */
-int btrfs_wait_ordered_extents(struct btrfs_root *root, int nocow_only)
+int btrfs_wait_ordered_extents(struct btrfs_root *root,
+			       int nocow_only, int delay_iput)
 {
 	struct list_head splice;
 	struct list_head *cur;
@@ -372,7 +390,10 @@ int btrfs_wait_ordered_extents(struct btrfs_root *root, int nocow_only)
 		if (inode) {
 			btrfs_start_ordered_extent(inode, ordered, 1);
 			btrfs_put_ordered_extent(ordered);
-			iput(inode);
+			if (delay_iput)
+				btrfs_add_delayed_iput(inode);
+			else
+				iput(inode);
 		} else {
 			btrfs_put_ordered_extent(ordered);
 		}
@@ -430,7 +451,7 @@ again:
 				btrfs_wait_ordered_range(inode, 0, (u64)-1);
 			else
 				filemap_flush(inode->i_mapping);
-			iput(inode);
+			btrfs_add_delayed_iput(inode);
 		}
 
 		cond_resched();
@@ -589,7 +610,7 @@ out:
  * After an extent is done, call this to conditionally update the on disk
  * i_size.  i_size is updated to cover any fully written part of the file.
  */
-int btrfs_ordered_update_i_size(struct inode *inode,
+int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 				struct btrfs_ordered_extent *ordered)
 {
 	struct btrfs_ordered_inode_tree *tree = &BTRFS_I(inode)->ordered_tree;
@@ -597,18 +618,32 @@ int btrfs_ordered_update_i_size(struct inode *inode,
 	u64 disk_i_size;
 	u64 new_i_size;
 	u64 i_size_test;
+	u64 i_size = i_size_read(inode);
 	struct rb_node *node;
+	struct rb_node *prev = NULL;
 	struct btrfs_ordered_extent *test;
+	int ret = 1;
+
+	if (ordered)
+		offset = entry_end(ordered);
+	else
+		offset = ALIGN(offset, BTRFS_I(inode)->root->sectorsize);
 
 	mutex_lock(&tree->mutex);
 	disk_i_size = BTRFS_I(inode)->disk_i_size;
 
+	/* truncate file */
+	if (disk_i_size > i_size) {
+		BTRFS_I(inode)->disk_i_size = i_size;
+		ret = 0;
+		goto out;
+	}
+
 	/*
 	 * if the disk i_size is already at the inode->i_size, or
 	 * this ordered extent is inside the disk i_size, we're done
 	 */
-	if (disk_i_size >= inode->i_size ||
-	    ordered->file_offset + ordered->len <= disk_i_size) {
+	if (disk_i_size == i_size || offset <= disk_i_size) {
 		goto out;
 	}
 
@@ -616,8 +651,7 @@ int btrfs_ordered_update_i_size(struct inode *inode,
 	 * we can't update the disk_isize if there are delalloc bytes
 	 * between disk_i_size and  this ordered extent
 	 */
-	if (test_range_bit(io_tree, disk_i_size,
-			   ordered->file_offset + ordered->len - 1,
+	if (test_range_bit(io_tree, disk_i_size, offset - 1,
 			   EXTENT_DELALLOC, 0, NULL)) {
 		goto out;
 	}
@@ -626,20 +660,32 @@ int btrfs_ordered_update_i_size(struct inode *inode,
 	 * if we find an ordered extent then we can't update disk i_size
 	 * yet
 	 */
-	node = &ordered->rb_node;
-	while (1) {
-		node = rb_prev(node);
-		if (!node)
-			break;
+	if (ordered) {
+		node = rb_prev(&ordered->rb_node);
+	} else {
+		prev = tree_search(tree, offset);
+		/*
+		 * we insert file extents without involving ordered struct,
+		 * so there should be no ordered struct cover this offset
+		 */
+		if (prev) {
+			test = rb_entry(prev, struct btrfs_ordered_extent,
+					rb_node);
+			BUG_ON(offset_in_entry(test, offset));
+		}
+		node = prev;
+	}
+	while (node) {
 		test = rb_entry(node, struct btrfs_ordered_extent, rb_node);
 		if (test->file_offset + test->len <= disk_i_size)
 			break;
-		if (test->file_offset >= inode->i_size)
+		if (test->file_offset >= i_size)
 			break;
 		if (test->file_offset >= disk_i_size)
 			goto out;
+		node = rb_prev(node);
 	}
-	new_i_size = min_t(u64, entry_end(ordered), i_size_read(inode));
+	new_i_size = min_t(u64, offset, i_size);
 
 	/*
 	 * at this point, we know we can safely update i_size to at least
@@ -647,7 +693,14 @@ int btrfs_ordered_update_i_size(struct inode *inode,
 	 * walk forward and see if ios from higher up in the file have
 	 * finished.
 	 */
-	node = rb_next(&ordered->rb_node);
+	if (ordered) {
+		node = rb_next(&ordered->rb_node);
+	} else {
+		if (prev)
+			node = rb_next(prev);
+		else
+			node = rb_first(&tree->tree);
+	}
 	i_size_test = 0;
 	if (node) {
 		/*
@@ -655,10 +708,10 @@ int btrfs_ordered_update_i_size(struct inode *inode,
 		 * between our ordered extent and the next one.
 		 */
 		test = rb_entry(node, struct btrfs_ordered_extent, rb_node);
-		if (test->file_offset > entry_end(ordered))
+		if (test->file_offset > offset)
 			i_size_test = test->file_offset;
 	} else {
-		i_size_test = i_size_read(inode);
+		i_size_test = i_size;
 	}
 
 	/*
@@ -667,15 +720,25 @@ int btrfs_ordered_update_i_size(struct inode *inode,
 	 * are no delalloc bytes in this area, it is safe to update
 	 * disk_i_size to the end of the region.
 	 */
-	if (i_size_test > entry_end(ordered) &&
-	    !test_range_bit(io_tree, entry_end(ordered), i_size_test - 1,
-			   EXTENT_DELALLOC, 0, NULL)) {
-		new_i_size = min_t(u64, i_size_test, i_size_read(inode));
+	if (i_size_test > offset &&
+	    !test_range_bit(io_tree, offset, i_size_test - 1,
+			    EXTENT_DELALLOC, 0, NULL)) {
+		new_i_size = min_t(u64, i_size_test, i_size);
 	}
 	BTRFS_I(inode)->disk_i_size = new_i_size;
+	ret = 0;
 out:
+	/*
+	 * we need to remove the ordered extent with the tree lock held
+	 * so that other people calling this function don't find our fully
+	 * processed ordered entry and skip updating the i_size
+	 */
+	if (ordered)
+		__btrfs_remove_ordered_extent(inode, ordered);
 	mutex_unlock(&tree->mutex);
-	return 0;
+	if (ordered)
+		wake_up(&ordered->wait);
+	return ret;
 }
 
 /*
diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h
index f82e87488ca8..1fe1282ef47c 100644
--- a/fs/btrfs/ordered-data.h
+++ b/fs/btrfs/ordered-data.h
@@ -150,12 +150,13 @@ void btrfs_start_ordered_extent(struct inode *inode,
 int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len);
 struct btrfs_ordered_extent *
 btrfs_lookup_first_ordered_extent(struct inode * inode, u64 file_offset);
-int btrfs_ordered_update_i_size(struct inode *inode,
+int btrfs_ordered_update_i_size(struct inode *inode, u64 offset,
 				struct btrfs_ordered_extent *ordered);
 int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr, u32 *sum);
-int btrfs_wait_ordered_extents(struct btrfs_root *root, int nocow_only);
 int btrfs_run_ordered_operations(struct btrfs_root *root, int wait);
 int btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root,
 				struct inode *inode);
+int btrfs_wait_ordered_extents(struct btrfs_root *root,
+			       int nocow_only, int delay_iput);
 #endif
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index cfcc93c93a7b..ed3e4a2ec2c8 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -1561,6 +1561,20 @@ static int invalidate_extent_cache(struct btrfs_root *root,
 	return 0;
 }
 
+static void put_inodes(struct list_head *list)
+{
+	struct inodevec *ivec;
+	while (!list_empty(list)) {
+		ivec = list_entry(list->next, struct inodevec, list);
+		list_del(&ivec->list);
+		while (ivec->nr > 0) {
+			ivec->nr--;
+			iput(ivec->inode[ivec->nr]);
+		}
+		kfree(ivec);
+	}
+}
+
 static int find_next_key(struct btrfs_path *path, int level,
 			 struct btrfs_key *key)
 
@@ -1723,6 +1737,11 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
 
 		btrfs_btree_balance_dirty(root, nr);
 
+		/*
+		 * put inodes outside transaction, otherwise we may deadlock.
+		 */
+		put_inodes(&inode_list);
+
 		if (replaced && rc->stage == UPDATE_DATA_PTRS)
 			invalidate_extent_cache(root, &key, &next_key);
 	}
@@ -1752,19 +1771,7 @@ out:
 
 	btrfs_btree_balance_dirty(root, nr);
 
-	/*
-	 * put inodes while we aren't holding the tree locks
-	 */
-	while (!list_empty(&inode_list)) {
-		struct inodevec *ivec;
-		ivec = list_entry(inode_list.next, struct inodevec, list);
-		list_del(&ivec->list);
-		while (ivec->nr > 0) {
-			ivec->nr--;
-			iput(ivec->inode[ivec->nr]);
-		}
-		kfree(ivec);
-	}
+	put_inodes(&inode_list);
 
 	if (replaced && rc->stage == UPDATE_DATA_PTRS)
 		invalidate_extent_cache(root, &key, &next_key);
@@ -3274,8 +3281,10 @@ static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
 		return -ENOMEM;
 
 	path = btrfs_alloc_path();
-	if (!path)
+	if (!path) {
+		kfree(cluster);
 		return -ENOMEM;
+	}
 
 	rc->extents_found = 0;
 	rc->extents_skipped = 0;
@@ -3534,8 +3543,8 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
 	       (unsigned long long)rc->block_group->key.objectid,
 	       (unsigned long long)rc->block_group->flags);
 
-	btrfs_start_delalloc_inodes(fs_info->tree_root);
-	btrfs_wait_ordered_extents(fs_info->tree_root, 0);
+	btrfs_start_delalloc_inodes(fs_info->tree_root, 0);
+	btrfs_wait_ordered_extents(fs_info->tree_root, 0, 0);
 
 	while (1) {
 		rc->extents_found = 0;
@@ -3755,6 +3764,7 @@ out:
 				       BTRFS_DATA_RELOC_TREE_OBJECTID);
 		if (IS_ERR(fs_root))
 			err = PTR_ERR(fs_root);
+		btrfs_orphan_cleanup(fs_root);
 	}
 	return err;
 }
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 752a5463bf53..8a1ea6e64575 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -66,7 +66,8 @@ enum {
 	Opt_degraded, Opt_subvol, Opt_device, Opt_nodatasum, Opt_nodatacow,
 	Opt_max_extent, Opt_max_inline, Opt_alloc_start, Opt_nobarrier,
 	Opt_ssd, Opt_nossd, Opt_ssd_spread, Opt_thread_pool, Opt_noacl,
-	Opt_compress, Opt_notreelog, Opt_ratio, Opt_flushoncommit,
+	Opt_compress, Opt_compress_force, Opt_notreelog, Opt_ratio,
+	Opt_flushoncommit,
 	Opt_discard, Opt_err,
 };
 
@@ -82,6 +83,7 @@ static match_table_t tokens = {
 	{Opt_alloc_start, "alloc_start=%s"},
 	{Opt_thread_pool, "thread_pool=%d"},
 	{Opt_compress, "compress"},
+	{Opt_compress_force, "compress-force"},
 	{Opt_ssd, "ssd"},
 	{Opt_ssd_spread, "ssd_spread"},
 	{Opt_nossd, "nossd"},
@@ -128,6 +130,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 	substring_t args[MAX_OPT_ARGS];
 	char *p, *num;
 	int intarg;
+	int ret = 0;
 
 	if (!options)
 		return 0;
@@ -172,6 +175,11 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 			printk(KERN_INFO "btrfs: use compression\n");
 			btrfs_set_opt(info->mount_opt, COMPRESS);
 			break;
+		case Opt_compress_force:
+			printk(KERN_INFO "btrfs: forcing compression\n");
+			btrfs_set_opt(info->mount_opt, FORCE_COMPRESS);
+			btrfs_set_opt(info->mount_opt, COMPRESS);
+			break;
 		case Opt_ssd:
 			printk(KERN_INFO "btrfs: use ssd allocation scheme\n");
 			btrfs_set_opt(info->mount_opt, SSD);
@@ -262,12 +270,18 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
 		case Opt_discard:
 			btrfs_set_opt(info->mount_opt, DISCARD);
 			break;
+		case Opt_err:
+			printk(KERN_INFO "btrfs: unrecognized mount option "
+			       "'%s'\n", p);
+			ret = -EINVAL;
+			goto out;
 		default:
 			break;
 		}
 	}
+out:
 	kfree(options);
-	return 0;
+	return ret;
 }
 
 /*
@@ -405,8 +419,8 @@ int btrfs_sync_fs(struct super_block *sb, int wait)
 		return 0;
 	}
 
-	btrfs_start_delalloc_inodes(root);
-	btrfs_wait_ordered_extents(root, 0);
+	btrfs_start_delalloc_inodes(root, 0);
+	btrfs_wait_ordered_extents(root, 0, 0);
 
 	trans = btrfs_start_transaction(root, 1);
 	ret = btrfs_commit_transaction(trans, root);
@@ -450,6 +464,8 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
 		seq_puts(seq, ",notreelog");
 	if (btrfs_test_opt(root, FLUSHONCOMMIT))
 		seq_puts(seq, ",flushoncommit");
+	if (btrfs_test_opt(root, DISCARD))
+		seq_puts(seq, ",discard");
 	if (!(root->fs_info->sb->s_flags & MS_POSIXACL))
 		seq_puts(seq, ",noacl");
 	return 0;
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index c207e8c32c9b..b2acc79f1b34 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -333,6 +333,9 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans,
 	memset(trans, 0, sizeof(*trans));
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
 
+	if (throttle)
+		btrfs_run_delayed_iputs(root);
+
 	return 0;
 }
 
@@ -354,7 +357,7 @@ int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans,
  * those extents are sent to disk but does not wait on them
  */
 int btrfs_write_marked_extents(struct btrfs_root *root,
-			       struct extent_io_tree *dirty_pages)
+			       struct extent_io_tree *dirty_pages, int mark)
 {
 	int ret;
 	int err = 0;
@@ -367,7 +370,7 @@ int btrfs_write_marked_extents(struct btrfs_root *root,
 
 	while (1) {
 		ret = find_first_extent_bit(dirty_pages, start, &start, &end,
-					    EXTENT_DIRTY);
+					    mark);
 		if (ret)
 			break;
 		while (start <= end) {
@@ -413,7 +416,7 @@ int btrfs_write_marked_extents(struct btrfs_root *root,
  * on all the pages and clear them from the dirty pages state tree
  */
 int btrfs_wait_marked_extents(struct btrfs_root *root,
-			      struct extent_io_tree *dirty_pages)
+			      struct extent_io_tree *dirty_pages, int mark)
 {
 	int ret;
 	int err = 0;
@@ -425,12 +428,12 @@ int btrfs_wait_marked_extents(struct btrfs_root *root,
 	unsigned long index;
 
 	while (1) {
-		ret = find_first_extent_bit(dirty_pages, 0, &start, &end,
-					    EXTENT_DIRTY);
+		ret = find_first_extent_bit(dirty_pages, start, &start, &end,
+					    mark);
 		if (ret)
 			break;
 
-		clear_extent_dirty(dirty_pages, start, end, GFP_NOFS);
+		clear_extent_bits(dirty_pages, start, end, mark, GFP_NOFS);
 		while (start <= end) {
 			index = start >> PAGE_CACHE_SHIFT;
 			start = (u64)(index + 1) << PAGE_CACHE_SHIFT;
@@ -460,13 +463,13 @@ int btrfs_wait_marked_extents(struct btrfs_root *root,
  * those extents are on disk for transaction or log commit
  */
 int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
-					struct extent_io_tree *dirty_pages)
+				struct extent_io_tree *dirty_pages, int mark)
 {
 	int ret;
 	int ret2;
 
-	ret = btrfs_write_marked_extents(root, dirty_pages);
-	ret2 = btrfs_wait_marked_extents(root, dirty_pages);
+	ret = btrfs_write_marked_extents(root, dirty_pages, mark);
+	ret2 = btrfs_wait_marked_extents(root, dirty_pages, mark);
 	return ret || ret2;
 }
 
@@ -479,7 +482,8 @@ int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans,
 		return filemap_write_and_wait(btree_inode->i_mapping);
 	}
 	return btrfs_write_and_wait_marked_extents(root,
-					   &trans->transaction->dirty_pages);
+					   &trans->transaction->dirty_pages,
+					   EXTENT_DIRTY);
 }
 
 /*
@@ -497,13 +501,16 @@ static int update_cowonly_root(struct btrfs_trans_handle *trans,
 {
 	int ret;
 	u64 old_root_bytenr;
+	u64 old_root_used;
 	struct btrfs_root *tree_root = root->fs_info->tree_root;
 
+	old_root_used = btrfs_root_used(&root->root_item);
 	btrfs_write_dirty_block_groups(trans, root);
 
 	while (1) {
 		old_root_bytenr = btrfs_root_bytenr(&root->root_item);
-		if (old_root_bytenr == root->node->start)
+		if (old_root_bytenr == root->node->start &&
+		    old_root_used == btrfs_root_used(&root->root_item))
 			break;
 
 		btrfs_set_root_node(&root->root_item, root->node);
@@ -512,6 +519,7 @@ static int update_cowonly_root(struct btrfs_trans_handle *trans,
 					&root->root_item);
 		BUG_ON(ret);
 
+		old_root_used = btrfs_root_used(&root->root_item);
 		ret = btrfs_write_dirty_block_groups(trans, root);
 		BUG_ON(ret);
 	}
@@ -795,7 +803,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans,
 	memcpy(&pending->root_key, &key, sizeof(key));
 fail:
 	kfree(new_root_item);
-	btrfs_unreserve_metadata_space(root, 6);
 	return ret;
 }
 
@@ -807,7 +814,6 @@ static noinline int finish_pending_snapshot(struct btrfs_fs_info *fs_info,
 	u64 index = 0;
 	struct btrfs_trans_handle *trans;
 	struct inode *parent_inode;
-	struct inode *inode;
 	struct btrfs_root *parent_root;
 
 	parent_inode = pending->dentry->d_parent->d_inode;
@@ -839,8 +845,6 @@ static noinline int finish_pending_snapshot(struct btrfs_fs_info *fs_info,
 
 	BUG_ON(ret);
 
-	inode = btrfs_lookup_dentry(parent_inode, pending->dentry);
-	d_instantiate(pending->dentry, inode);
 fail:
 	btrfs_end_transaction(trans, fs_info->fs_root);
 	return ret;
@@ -994,11 +998,11 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 		mutex_unlock(&root->fs_info->trans_mutex);
 
 		if (flush_on_commit) {
-			btrfs_start_delalloc_inodes(root);
-			ret = btrfs_wait_ordered_extents(root, 0);
+			btrfs_start_delalloc_inodes(root, 1);
+			ret = btrfs_wait_ordered_extents(root, 0, 1);
 			BUG_ON(ret);
 		} else if (snap_pending) {
-			ret = btrfs_wait_ordered_extents(root, 1);
+			ret = btrfs_wait_ordered_extents(root, 0, 1);
 			BUG_ON(ret);
 		}
 
@@ -1116,6 +1120,10 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
 		current->journal_info = NULL;
 
 	kmem_cache_free(btrfs_trans_handle_cachep, trans);
+
+	if (current != root->fs_info->transaction_kthread)
+		btrfs_run_delayed_iputs(root);
+
 	return ret;
 }
 
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index d4e3e7a6938c..93c7ccb33118 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -107,10 +107,10 @@ void btrfs_throttle(struct btrfs_root *root);
 int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
 				struct btrfs_root *root);
 int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
-					struct extent_io_tree *dirty_pages);
+				struct extent_io_tree *dirty_pages, int mark);
 int btrfs_write_marked_extents(struct btrfs_root *root,
-					struct extent_io_tree *dirty_pages);
+				struct extent_io_tree *dirty_pages, int mark);
 int btrfs_wait_marked_extents(struct btrfs_root *root,
-					struct extent_io_tree *dirty_pages);
+				struct extent_io_tree *dirty_pages, int mark);
 int btrfs_transaction_in_commit(struct btrfs_fs_info *info);
 #endif
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 741666a7676a..4a9434b622ec 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -542,8 +542,8 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans,
 
 	saved_nbytes = inode_get_bytes(inode);
 	/* drop any overlapping extents */
-	ret = btrfs_drop_extents(trans, root, inode,
-			 start, extent_end, extent_end, start, &alloc_hint, 1);
+	ret = btrfs_drop_extents(trans, inode, start, extent_end,
+				 &alloc_hint, 1);
 	BUG_ON(ret);
 
 	if (found_type == BTRFS_FILE_EXTENT_REG ||
@@ -930,6 +930,17 @@ out_nowrite:
 	return 0;
 }
 
+static int insert_orphan_item(struct btrfs_trans_handle *trans,
+			      struct btrfs_root *root, u64 offset)
+{
+	int ret;
+	ret = btrfs_find_orphan_item(root, offset);
+	if (ret > 0)
+		ret = btrfs_insert_orphan_item(trans, root, offset);
+	return ret;
+}
+
+
 /*
  * There are a few corners where the link count of the file can't
  * be properly maintained during replay.  So, instead of adding
@@ -997,9 +1008,13 @@ static noinline int fixup_inode_link_count(struct btrfs_trans_handle *trans,
 	}
 	BTRFS_I(inode)->index_cnt = (u64)-1;
 
-	if (inode->i_nlink == 0 && S_ISDIR(inode->i_mode)) {
-		ret = replay_dir_deletes(trans, root, NULL, path,
-					 inode->i_ino, 1);
+	if (inode->i_nlink == 0) {
+		if (S_ISDIR(inode->i_mode)) {
+			ret = replay_dir_deletes(trans, root, NULL, path,
+						 inode->i_ino, 1);
+			BUG_ON(ret);
+		}
+		ret = insert_orphan_item(trans, root, inode->i_ino);
 		BUG_ON(ret);
 	}
 	btrfs_free_path(path);
@@ -1587,7 +1602,6 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 		/* inode keys are done during the first stage */
 		if (key.type == BTRFS_INODE_ITEM_KEY &&
 		    wc->stage == LOG_WALK_REPLAY_INODES) {
-			struct inode *inode;
 			struct btrfs_inode_item *inode_item;
 			u32 mode;
 
@@ -1603,31 +1617,16 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb,
 					     eb, i, &key);
 			BUG_ON(ret);
 
-			/* for regular files, truncate away
-			 * extents past the new EOF
+			/* for regular files, make sure corresponding
+			 * orhpan item exist. extents past the new EOF
+			 * will be truncated later by orphan cleanup.
 			 */
 			if (S_ISREG(mode)) {
-				inode = read_one_inode(root,
-						       key.objectid);
-				BUG_ON(!inode);
-
-				ret = btrfs_truncate_inode_items(wc->trans,
-					root, inode, inode->i_size,
-					BTRFS_EXTENT_DATA_KEY);
+				ret = insert_orphan_item(wc->trans, root,
+							 key.objectid);
 				BUG_ON(ret);
-
-				/* if the nlink count is zero here, the iput
-				 * will free the inode.  We bump it to make
-				 * sure it doesn't get freed until the link
-				 * count fixup is done
-				 */
-				if (inode->i_nlink == 0) {
-					btrfs_inc_nlink(inode);
-					btrfs_update_inode(wc->trans,
-							   root, inode);
-				}
-				iput(inode);
 			}
+
 			ret = link_to_fixup_dir(wc->trans, root,
 						path, key.objectid);
 			BUG_ON(ret);
@@ -1977,10 +1976,11 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 {
 	int index1;
 	int index2;
+	int mark;
 	int ret;
 	struct btrfs_root *log = root->log_root;
 	struct btrfs_root *log_root_tree = root->fs_info->log_root_tree;
-	u64 log_transid = 0;
+	unsigned long log_transid = 0;
 
 	mutex_lock(&root->log_mutex);
 	index1 = root->log_transid % 2;
@@ -2014,24 +2014,29 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 		goto out;
 	}
 
+	log_transid = root->log_transid;
+	if (log_transid % 2 == 0)
+		mark = EXTENT_DIRTY;
+	else
+		mark = EXTENT_NEW;
+
 	/* we start IO on  all the marked extents here, but we don't actually
 	 * wait for them until later.
 	 */
-	ret = btrfs_write_marked_extents(log, &log->dirty_log_pages);
+	ret = btrfs_write_marked_extents(log, &log->dirty_log_pages, mark);
 	BUG_ON(ret);
 
 	btrfs_set_root_node(&log->root_item, log->node);
 
 	root->log_batch = 0;
-	log_transid = root->log_transid;
 	root->log_transid++;
 	log->log_transid = root->log_transid;
 	root->log_start_pid = 0;
 	smp_mb();
 	/*
-	 * log tree has been flushed to disk, new modifications of
-	 * the log will be written to new positions. so it's safe to
-	 * allow log writers to go in.
+	 * IO has been started, blocks of the log tree have WRITTEN flag set
+	 * in their headers. new modifications of the log will be written to
+	 * new positions. so it's safe to allow log writers to go in.
 	 */
 	mutex_unlock(&root->log_mutex);
 
@@ -2052,7 +2057,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 
 	index2 = log_root_tree->log_transid % 2;
 	if (atomic_read(&log_root_tree->log_commit[index2])) {
-		btrfs_wait_marked_extents(log, &log->dirty_log_pages);
+		btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
 		wait_log_commit(trans, log_root_tree,
 				log_root_tree->log_transid);
 		mutex_unlock(&log_root_tree->log_mutex);
@@ -2072,16 +2077,17 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
 	 * check the full commit flag again
 	 */
 	if (root->fs_info->last_trans_log_full_commit == trans->transid) {
-		btrfs_wait_marked_extents(log, &log->dirty_log_pages);
+		btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
 		mutex_unlock(&log_root_tree->log_mutex);
 		ret = -EAGAIN;
 		goto out_wake_log_root;
 	}
 
 	ret = btrfs_write_and_wait_marked_extents(log_root_tree,
-				&log_root_tree->dirty_log_pages);
+				&log_root_tree->dirty_log_pages,
+				EXTENT_DIRTY | EXTENT_NEW);
 	BUG_ON(ret);
-	btrfs_wait_marked_extents(log, &log->dirty_log_pages);
+	btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
 
 	btrfs_set_super_log_root(&root->fs_info->super_for_commit,
 				log_root_tree->node->start);
@@ -2147,12 +2153,12 @@ int btrfs_free_log(struct btrfs_trans_handle *trans, struct btrfs_root *root)
 
 	while (1) {
 		ret = find_first_extent_bit(&log->dirty_log_pages,
-				    0, &start, &end, EXTENT_DIRTY);
+				0, &start, &end, EXTENT_DIRTY | EXTENT_NEW);
 		if (ret)
 			break;
 
-		clear_extent_dirty(&log->dirty_log_pages,
-				   start, end, GFP_NOFS);
+		clear_extent_bits(&log->dirty_log_pages, start, end,
+				  EXTENT_DIRTY | EXTENT_NEW, GFP_NOFS);
 	}
 
 	if (log->log_transid > 0) {
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 7eda483d7b5a..41ecbb2347f2 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1135,7 +1135,7 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
 		root->fs_info->avail_metadata_alloc_bits;
 
 	if ((all_avail & BTRFS_BLOCK_GROUP_RAID10) &&
-	    root->fs_info->fs_devices->rw_devices <= 4) {
+	    root->fs_info->fs_devices->num_devices <= 4) {
 		printk(KERN_ERR "btrfs: unable to go below four devices "
 		       "on raid10\n");
 		ret = -EINVAL;
@@ -1143,7 +1143,7 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path)
 	}
 
 	if ((all_avail & BTRFS_BLOCK_GROUP_RAID1) &&
-	    root->fs_info->fs_devices->rw_devices <= 2) {
+	    root->fs_info->fs_devices->num_devices <= 2) {
 		printk(KERN_ERR "btrfs: unable to go below two "
 		       "devices on raid1\n");
 		ret = -EINVAL;
@@ -1434,8 +1434,8 @@ int btrfs_init_new_device(struct btrfs_root *root, char *device_path)
 		return -EINVAL;
 
 	bdev = open_bdev_exclusive(device_path, 0, root->fs_info->bdev_holder);
-	if (!bdev)
-		return -EIO;
+	if (IS_ERR(bdev))
+		return PTR_ERR(bdev);
 
 	if (root->fs_info->fs_devices->seeding) {
 		seeding_dev = 1;
@@ -2209,7 +2209,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans,
 		max_chunk_size = 10 * calc_size;
 		min_stripe_size = 64 * 1024 * 1024;
 	} else if (type & BTRFS_BLOCK_GROUP_METADATA) {
-		max_chunk_size = 4 * calc_size;
+		max_chunk_size = 256 * 1024 * 1024;
 		min_stripe_size = 32 * 1024 * 1024;
 	} else if (type & BTRFS_BLOCK_GROUP_SYSTEM) {
 		calc_size = 8 * 1024 * 1024;
@@ -2538,6 +2538,11 @@ int btrfs_chunk_readonly(struct btrfs_root *root, u64 chunk_offset)
 	if (!em)
 		return 1;
 
+	if (btrfs_test_opt(root, DEGRADED)) {
+		free_extent_map(em);
+		return 0;
+	}
+
 	map = (struct map_lookup *)em->bdev;
 	for (i = 0; i < map->num_stripes; i++) {
 		if (!map->stripes[i].dev->writeable) {
@@ -2649,8 +2654,10 @@ again:
 	em = lookup_extent_mapping(em_tree, logical, *length);
 	read_unlock(&em_tree->lock);
 
-	if (!em && unplug_page)
+	if (!em && unplug_page) {
+		kfree(multi);
 		return 0;
+	}
 
 	if (!em) {
 		printk(KERN_CRIT "unable to find logical %llu len %llu\n",
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index b6dd5967c48a..193b58f7d3f3 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -85,22 +85,23 @@ out:
 	return ret;
 }
 
-int __btrfs_setxattr(struct inode *inode, const char *name,
-			    const void *value, size_t size, int flags)
+static int do_setxattr(struct btrfs_trans_handle *trans,
+		       struct inode *inode, const char *name,
+		       const void *value, size_t size, int flags)
 {
 	struct btrfs_dir_item *di;
 	struct btrfs_root *root = BTRFS_I(inode)->root;
-	struct btrfs_trans_handle *trans;
 	struct btrfs_path *path;
-	int ret = 0, mod = 0;
+	size_t name_len = strlen(name);
+	int ret = 0;
+
+	if (name_len + size > BTRFS_MAX_XATTR_SIZE(root))
+		return -ENOSPC;
 
 	path = btrfs_alloc_path();
 	if (!path)
 		return -ENOMEM;
 
-	trans = btrfs_join_transaction(root, 1);
-	btrfs_set_trans_block_group(trans, inode);
-
 	/* first lets see if we already have this xattr */
 	di = btrfs_lookup_xattr(trans, root, path, inode->i_ino, name,
 				strlen(name), -1);
@@ -118,15 +119,12 @@ int __btrfs_setxattr(struct inode *inode, const char *name,
 		}
 
 		ret = btrfs_delete_one_dir_name(trans, root, path, di);
-		if (ret)
-			goto out;
+		BUG_ON(ret);
 		btrfs_release_path(root, path);
 
 		/* if we don't have a value then we are removing the xattr */
-		if (!value) {
-			mod = 1;
+		if (!value)
 			goto out;
-		}
 	} else {
 		btrfs_release_path(root, path);
 
@@ -138,20 +136,45 @@ int __btrfs_setxattr(struct inode *inode, const char *name,
 	}
 
 	/* ok we have to create a completely new xattr */
-	ret = btrfs_insert_xattr_item(trans, root, name, strlen(name),
-				      value, size, inode->i_ino);
+	ret = btrfs_insert_xattr_item(trans, root, path, inode->i_ino,
+				      name, name_len, value, size);
+	BUG_ON(ret);
+out:
+	btrfs_free_path(path);
+	return ret;
+}
+
+int __btrfs_setxattr(struct btrfs_trans_handle *trans,
+		     struct inode *inode, const char *name,
+		     const void *value, size_t size, int flags)
+{
+	struct btrfs_root *root = BTRFS_I(inode)->root;
+	int ret;
+
+	if (trans)
+		return do_setxattr(trans, inode, name, value, size, flags);
+
+	ret = btrfs_reserve_metadata_space(root, 2);
 	if (ret)
-		goto out;
-	mod = 1;
+		return ret;
 
-out:
-	if (mod) {
-		inode->i_ctime = CURRENT_TIME;
-		ret = btrfs_update_inode(trans, root, inode);
+	trans = btrfs_start_transaction(root, 1);
+	if (!trans) {
+		ret = -ENOMEM;
+		goto out;
 	}
+	btrfs_set_trans_block_group(trans, inode);
 
-	btrfs_end_transaction(trans, root);
-	btrfs_free_path(path);
+	ret = do_setxattr(trans, inode, name, value, size, flags);
+	if (ret)
+		goto out;
+
+	inode->i_ctime = CURRENT_TIME;
+	ret = btrfs_update_inode(trans, root, inode);
+	BUG_ON(ret);
+out:
+	btrfs_end_transaction_throttle(trans, root);
+	btrfs_unreserve_metadata_space(root, 2);
 	return ret;
 }
 
@@ -314,7 +337,9 @@ int btrfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 
 	if (size == 0)
 		value = "";  /* empty EA, do not remove */
-	return __btrfs_setxattr(dentry->d_inode, name, value, size, flags);
+
+	return __btrfs_setxattr(NULL, dentry->d_inode, name, value, size,
+				flags);
 }
 
 int btrfs_removexattr(struct dentry *dentry, const char *name)
@@ -329,10 +354,13 @@ int btrfs_removexattr(struct dentry *dentry, const char *name)
 
 	if (!btrfs_is_valid_xattr(name))
 		return -EOPNOTSUPP;
-	return __btrfs_setxattr(dentry->d_inode, name, NULL, 0, XATTR_REPLACE);
+
+	return __btrfs_setxattr(NULL, dentry->d_inode, name, NULL, 0,
+				XATTR_REPLACE);
 }
 
-int btrfs_xattr_security_init(struct inode *inode, struct inode *dir)
+int btrfs_xattr_security_init(struct btrfs_trans_handle *trans,
+			      struct inode *inode, struct inode *dir)
 {
 	int err;
 	size_t len;
@@ -354,7 +382,7 @@ int btrfs_xattr_security_init(struct inode *inode, struct inode *dir)
 	} else {
 		strcpy(name, XATTR_SECURITY_PREFIX);
 		strcpy(name + XATTR_SECURITY_PREFIX_LEN, suffix);
-		err = __btrfs_setxattr(inode, name, value, len, 0);
+		err = __btrfs_setxattr(trans, inode, name, value, len, 0);
 		kfree(name);
 	}
 
diff --git a/fs/btrfs/xattr.h b/fs/btrfs/xattr.h
index c71e9c3cf3f7..721efa0346e0 100644
--- a/fs/btrfs/xattr.h
+++ b/fs/btrfs/xattr.h
@@ -27,15 +27,16 @@ extern struct xattr_handler *btrfs_xattr_handlers[];
 
 extern ssize_t __btrfs_getxattr(struct inode *inode, const char *name,
 		void *buffer, size_t size);
-extern int __btrfs_setxattr(struct inode *inode, const char *name,
-		const void *value, size_t size, int flags);
-
+extern int __btrfs_setxattr(struct btrfs_trans_handle *trans,
+			    struct inode *inode, const char *name,
+			    const void *value, size_t size, int flags);
 extern ssize_t btrfs_getxattr(struct dentry *dentry, const char *name,
 		void *buffer, size_t size);
 extern int btrfs_setxattr(struct dentry *dentry, const char *name,
 		const void *value, size_t size, int flags);
 extern int btrfs_removexattr(struct dentry *dentry, const char *name);
 
-extern int btrfs_xattr_security_init(struct inode *inode, struct inode *dir);
+extern int btrfs_xattr_security_init(struct btrfs_trans_handle *trans,
+				     struct inode *inode, struct inode *dir);
 
 #endif /* __XATTR__ */
diff --git a/fs/cachefiles/bind.c b/fs/cachefiles/bind.c
index 3797e0077b35..2906077ac798 100644
--- a/fs/cachefiles/bind.c
+++ b/fs/cachefiles/bind.c
@@ -84,7 +84,7 @@ int cachefiles_daemon_bind(struct cachefiles_cache *cache, char *args)
 static int cachefiles_daemon_add_cache(struct cachefiles_cache *cache)
 {
 	struct cachefiles_object *fsdef;
-	struct nameidata nd;
+	struct path path;
 	struct kstatfs stats;
 	struct dentry *graveyard, *cachedir, *root;
 	const struct cred *saved_cred;
@@ -114,15 +114,12 @@ static int cachefiles_daemon_add_cache(struct cachefiles_cache *cache)
 	_debug("- fsdef %p", fsdef);
 
 	/* look up the directory at the root of the cache */
-	memset(&nd, 0, sizeof(nd));
-
-	ret = path_lookup(cache->rootdirname, LOOKUP_DIRECTORY, &nd);
+	ret = kern_path(cache->rootdirname, LOOKUP_DIRECTORY, &path);
 	if (ret < 0)
 		goto error_open_root;
 
-	cache->mnt = mntget(nd.path.mnt);
-	root = dget(nd.path.dentry);
-	path_put(&nd.path);
+	cache->mnt = path.mnt;
+	root = path.dentry;
 
 	/* check parameters */
 	ret = -EOPNOTSUPP;
diff --git a/fs/cachefiles/daemon.c b/fs/cachefiles/daemon.c
index 4618516dd994..c2413561ea75 100644
--- a/fs/cachefiles/daemon.c
+++ b/fs/cachefiles/daemon.c
@@ -21,6 +21,7 @@
 #include <linux/mount.h>
 #include <linux/statfs.h>
 #include <linux/ctype.h>
+#include <linux/string.h>
 #include <linux/fs_struct.h>
 #include "internal.h"
 
@@ -257,8 +258,7 @@ static ssize_t cachefiles_daemon_write(struct file *file,
 		if (args == data)
 			goto error;
 		*args = '\0';
-		for (args++; isspace(*args); args++)
-			continue;
+		args = skip_spaces(++args);
 	}
 
 	/* run the appropriate command handler */
diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c
index a6c8c6fe8df9..1d8332563863 100644
--- a/fs/cachefiles/rdwr.c
+++ b/fs/cachefiles/rdwr.c
@@ -11,7 +11,6 @@
 
 #include <linux/mount.h>
 #include <linux/file.h>
-#include <linux/ima.h>
 #include "internal.h"
 
 /*
@@ -923,7 +922,6 @@ int cachefiles_write_page(struct fscache_storage *op, struct page *page)
 	if (IS_ERR(file)) {
 		ret = PTR_ERR(file);
 	} else {
-		ima_counts_get(file);
 		ret = -EIO;
 		if (file->f_op->write) {
 			pos = (loff_t) page->index << PAGE_SHIFT;
diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES
index 094ea65afc85..7b2600b380d7 100644
--- a/fs/cifs/CHANGES
+++ b/fs/cifs/CHANGES
@@ -5,7 +5,9 @@ have duplicated data). Fix oops in cifs_lookup. Workaround problem
 mounting to OS/400 Netserve. Fix oops in cifs_get_tcp_session.
 Disable use of server inode numbers when server only
 partially supports them (e.g. for one server querying inode numbers on
-FindFirst fails but QPathInfo queries works).
+FindFirst fails but QPathInfo queries works). Fix oops with dfs in 
+cifs_put_smb_ses. Fix mmap to work on directio mounts (needed
+for OpenOffice when on forcedirectio mount e.g.)
 
 Version 1.60
 -------------
diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c
index fea9e898c4ba..b44ce0a0711c 100644
--- a/fs/cifs/cifs_dfs_ref.c
+++ b/fs/cifs/cifs_dfs_ref.c
@@ -269,7 +269,7 @@ static int add_mount_helper(struct vfsmount *newmnt, struct nameidata *nd,
 	int err;
 
 	mntget(newmnt);
-	err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags, mntlist);
+	err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags | MNT_SHRINKABLE, mntlist);
 	switch (err) {
 	case 0:
 		path_put(&nd->path);
@@ -371,7 +371,6 @@ cifs_dfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
 	if (IS_ERR(mnt))
 		goto out_err;
 
-	nd->path.mnt->mnt_flags |= MNT_SHRINKABLE;
 	rc = add_mount_helper(mnt, nd, &cifs_dfs_automount_list);
 
 out:
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 29f1da761bbf..8c6a03627176 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -758,7 +758,7 @@ const struct file_operations cifs_file_ops = {
 };
 
 const struct file_operations cifs_file_direct_ops = {
-	/* no mmap, no aio, no readv -
+	/* no aio, no readv -
 	   BB reevaluate whether they can be done with directio, no cache */
 	.read = cifs_user_read,
 	.write = cifs_user_write,
@@ -767,6 +767,7 @@ const struct file_operations cifs_file_direct_ops = {
 	.lock = cifs_lock,
 	.fsync = cifs_fsync,
 	.flush = cifs_flush,
+	.mmap = cifs_file_mmap,
 	.splice_read = generic_file_splice_read,
 #ifdef CONFIG_CIFS_POSIX
 	.unlocked_ioctl  = cifs_ioctl,
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 63ea83ff687f..3bbcaa716b3c 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -2287,12 +2287,12 @@ int
 cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb,
 		char *mount_data_global, const char *devname)
 {
-	int rc = 0;
+	int rc;
 	int xid;
 	struct smb_vol *volume_info;
-	struct cifsSesInfo *pSesInfo = NULL;
-	struct cifsTconInfo *tcon = NULL;
-	struct TCP_Server_Info *srvTcp = NULL;
+	struct cifsSesInfo *pSesInfo;
+	struct cifsTconInfo *tcon;
+	struct TCP_Server_Info *srvTcp;
 	char   *full_path;
 	char *mount_data = mount_data_global;
 #ifdef CONFIG_CIFS_DFS_UPCALL
@@ -2301,6 +2301,10 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb,
 	int referral_walks_count = 0;
 try_mount_again:
 #endif
+	rc = 0;
+	tcon = NULL;
+	pSesInfo = NULL;
+	srvTcp = NULL;
 	full_path = NULL;
 
 	xid = GetXid();
@@ -2597,6 +2601,7 @@ remote_path_check:
 
 			cleanup_volume_info(&volume_info);
 			referral_walks_count++;
+			FreeXid(xid);
 			goto try_mount_again;
 		}
 #else /* No DFS support, return error on mount */
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 1f42f772865a..6ccf7262d1b7 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -214,7 +214,8 @@ int cifs_posix_open(char *full_path, struct inode **pinode,
 		posix_flags |= SMB_O_EXCL;
 	if (oflags & O_TRUNC)
 		posix_flags |= SMB_O_TRUNC;
-	if (oflags & O_SYNC)
+	/* be safe and imply O_SYNC for O_DSYNC */
+	if (oflags & O_DSYNC)
 		posix_flags |= SMB_O_SYNC;
 	if (oflags & O_DIRECTORY)
 		posix_flags |= SMB_O_DIRECTORY;
diff --git a/fs/cifs/export.c b/fs/cifs/export.c
index 75949d6a5f1b..6177f7cca16a 100644
--- a/fs/cifs/export.c
+++ b/fs/cifs/export.c
@@ -24,7 +24,7 @@
  */
 
  /*
-  * See Documentation/filesystems/Exporting
+  * See Documentation/filesystems/nfs/Exporting
   * and examples in fs/exportfs
   *
   * Since cifs is a network file system, an "fsid" must be included for
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 429337eb7afe..057e1dae12ab 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -76,8 +76,10 @@ static inline fmode_t cifs_posix_convert_flags(unsigned int flags)
 	   reopening a file.  They had their effect on the original open */
 	if (flags & O_APPEND)
 		posix_flags |= (fmode_t)O_APPEND;
-	if (flags & O_SYNC)
-		posix_flags |= (fmode_t)O_SYNC;
+	if (flags & O_DSYNC)
+		posix_flags |= (fmode_t)O_DSYNC;
+	if (flags & __O_SYNC)
+		posix_flags |= (fmode_t)__O_SYNC;
 	if (flags & O_DIRECTORY)
 		posix_flags |= (fmode_t)O_DIRECTORY;
 	if (flags & O_NOFOLLOW)
diff --git a/fs/compat.c b/fs/compat.c
index 6c19040ffeef..00d90c2e66f0 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -38,8 +38,6 @@
 #include <linux/dirent.h>
 #include <linux/fsnotify.h>
 #include <linux/highuid.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
 #include <linux/nfsd/syscall.h>
 #include <linux/personality.h>
 #include <linux/rwsem.h>
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 2346895b3a77..c5c45de1a2ee 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -111,43 +111,40 @@
 #include <linux/dvb/frontend.h>
 #include <linux/dvb/video.h>
 
+#include <linux/sort.h>
+
 #ifdef CONFIG_SPARC
 #include <asm/fbio.h>
 #endif
 
-static int do_ioctl32_pointer(unsigned int fd, unsigned int cmd,
-			      unsigned long arg, struct file *f)
-{
-	return sys_ioctl(fd, cmd, (unsigned long)compat_ptr(arg));
-}
-
-static int w_long(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int w_long(unsigned int fd, unsigned int cmd,
+		compat_ulong_t __user *argp)
 {
 	mm_segment_t old_fs = get_fs();
 	int err;
 	unsigned long val;
-	
+
 	set_fs (KERNEL_DS);
 	err = sys_ioctl(fd, cmd, (unsigned long)&val);
 	set_fs (old_fs);
-	if (!err && put_user(val, (u32 __user *)compat_ptr(arg)))
+	if (!err && put_user(val, argp))
 		return -EFAULT;
 	return err;
 }
- 
-static int rw_long(unsigned int fd, unsigned int cmd, unsigned long arg)
+
+static int rw_long(unsigned int fd, unsigned int cmd,
+		compat_ulong_t __user *argp)
 {
 	mm_segment_t old_fs = get_fs();
-	u32 __user *argptr = compat_ptr(arg);
 	int err;
 	unsigned long val;
-	
-	if(get_user(val, argptr))
+
+	if(get_user(val, argp))
 		return -EFAULT;
 	set_fs (KERNEL_DS);
 	err = sys_ioctl(fd, cmd, (unsigned long)&val);
 	set_fs (old_fs);
-	if (!err && put_user(val, argptr))
+	if (!err && put_user(val, argp))
 		return -EFAULT;
 	return err;
 }
@@ -161,7 +158,8 @@ struct compat_video_event {
 	} u;
 };
 
-static int do_video_get_event(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int do_video_get_event(unsigned int fd, unsigned int cmd,
+		struct compat_video_event __user *up)
 {
 	struct video_event kevent;
 	mm_segment_t old_fs = get_fs();
@@ -172,8 +170,6 @@ static int do_video_get_event(unsigned int fd, unsigned int cmd, unsigned long a
 	set_fs(old_fs);
 
 	if (!err) {
-		struct compat_video_event __user *up = compat_ptr(arg);
-
 		err  = put_user(kevent.type, &up->type);
 		err |= put_user(kevent.timestamp, &up->timestamp);
 		err |= put_user(kevent.u.size.w, &up->u.size.w);
@@ -192,15 +188,14 @@ struct compat_video_still_picture {
         int32_t size;
 };
 
-static int do_video_stillpicture(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int do_video_stillpicture(unsigned int fd, unsigned int cmd,
+	struct compat_video_still_picture __user *up)
 {
-	struct compat_video_still_picture __user *up;
 	struct video_still_picture __user *up_native;
 	compat_uptr_t fp;
 	int32_t size;
 	int err;
 
-	up = (struct compat_video_still_picture __user *) arg;
 	err  = get_user(fp, &up->iFrame);
 	err |= get_user(size, &up->size);
 	if (err)
@@ -224,14 +219,13 @@ struct compat_video_spu_palette {
 	compat_uptr_t palette;
 };
 
-static int do_video_set_spu_palette(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int do_video_set_spu_palette(unsigned int fd, unsigned int cmd,
+		struct compat_video_spu_palette __user *up)
 {
-	struct compat_video_spu_palette __user *up;
 	struct video_spu_palette __user *up_native;
 	compat_uptr_t palp;
 	int length, err;
 
-	up = (struct compat_video_spu_palette __user *) arg;
 	err  = get_user(palp, &up->palette);
 	err |= get_user(length, &up->length);
 
@@ -299,16 +293,15 @@ static int sg_build_iovec(sg_io_hdr_t __user *sgio, void __user *dxferp, u16 iov
 	return 0;
 }
 
-static int sg_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int sg_ioctl_trans(unsigned int fd, unsigned int cmd,
+			sg_io_hdr32_t __user *sgio32)
 {
 	sg_io_hdr_t __user *sgio;
-	sg_io_hdr32_t __user *sgio32;
 	u16 iovec_count;
 	u32 data;
 	void __user *dxferp;
 	int err;
 
-	sgio32 = compat_ptr(arg);
 	if (get_user(iovec_count, &sgio32->iovec_count))
 		return -EFAULT;
 
@@ -398,11 +391,11 @@ struct compat_sg_req_info { /* used by SG_GET_REQUEST_TABLE ioctl() */
 	int unused;
 };
 
-static int sg_grt_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int sg_grt_trans(unsigned int fd, unsigned int cmd, struct
+			compat_sg_req_info __user *o)
 {
 	int err, i;
 	sg_req_info_t __user *r;
-	struct compat_sg_req_info __user *o = (void __user *)arg;
 	r = compat_alloc_user_space(sizeof(sg_req_info_t)*SG_MAX_QUEUE);
 	err = sys_ioctl(fd,cmd,(unsigned long)r);
 	if (err < 0)
@@ -430,9 +423,9 @@ struct sock_fprog32 {
 #define PPPIOCSPASS32	_IOW('t', 71, struct sock_fprog32)
 #define PPPIOCSACTIVE32	_IOW('t', 70, struct sock_fprog32)
 
-static int ppp_sock_fprog_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int ppp_sock_fprog_ioctl_trans(unsigned int fd, unsigned int cmd,
+			struct sock_fprog32 __user *u_fprog32)
 {
-	struct sock_fprog32 __user *u_fprog32 = compat_ptr(arg);
 	struct sock_fprog __user *u_fprog64 = compat_alloc_user_space(sizeof(struct sock_fprog));
 	void __user *fptr64;
 	u32 fptr32;
@@ -469,15 +462,14 @@ struct ppp_idle32 {
 };
 #define PPPIOCGIDLE32		_IOR('t', 63, struct ppp_idle32)
 
-static int ppp_gidle(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int ppp_gidle(unsigned int fd, unsigned int cmd,
+		struct ppp_idle32 __user *idle32)
 {
 	struct ppp_idle __user *idle;
-	struct ppp_idle32 __user *idle32;
 	__kernel_time_t xmit, recv;
 	int err;
 
 	idle = compat_alloc_user_space(sizeof(*idle));
-	idle32 = compat_ptr(arg);
 
 	err = sys_ioctl(fd, PPPIOCGIDLE, (unsigned long) idle);
 
@@ -491,15 +483,14 @@ static int ppp_gidle(unsigned int fd, unsigned int cmd, unsigned long arg)
 	return err;
 }
 
-static int ppp_scompress(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int ppp_scompress(unsigned int fd, unsigned int cmd,
+	struct ppp_option_data32 __user *odata32)
 {
 	struct ppp_option_data __user *odata;
-	struct ppp_option_data32 __user *odata32;
 	__u32 data;
 	void __user *datap;
 
 	odata = compat_alloc_user_space(sizeof(*odata));
-	odata32 = compat_ptr(arg);
 
 	if (get_user(data, &odata32->ptr))
 		return -EFAULT;
@@ -515,35 +506,6 @@ static int ppp_scompress(unsigned int fd, unsigned int cmd, unsigned long arg)
 	return sys_ioctl(fd, PPPIOCSCOMPRESS, (unsigned long) odata);
 }
 
-static int ppp_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	int err;
-
-	switch (cmd) {
-	case PPPIOCGIDLE32:
-		err = ppp_gidle(fd, cmd, arg);
-		break;
-
-	case PPPIOCSCOMPRESS32:
-		err = ppp_scompress(fd, cmd, arg);
-		break;
-
-	default:
-		do {
-			static int count;
-			if (++count <= 20)
-				printk("ppp_ioctl: Unknown cmd fd(%d) "
-				       "cmd(%08x) arg(%08x)\n",
-				       (int)fd, (unsigned int)cmd, (unsigned int)arg);
-		} while(0);
-		err = -EINVAL;
-		break;
-	};
-
-	return err;
-}
-
-
 #ifdef CONFIG_BLOCK
 struct mtget32 {
 	compat_long_t	mt_type;
@@ -561,7 +523,7 @@ struct mtpos32 {
 };
 #define MTIOCPOS32	_IOR('m', 3, struct mtpos32)
 
-static int mt_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int mt_ioctl_trans(unsigned int fd, unsigned int cmd, void __user *argp)
 {
 	mm_segment_t old_fs = get_fs();
 	struct mtget get;
@@ -581,15 +543,6 @@ static int mt_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
 		kcmd = MTIOCGET;
 		karg = &get;
 		break;
-	default:
-		do {
-			static int count;
-			if (++count <= 20)
-				printk("mt_ioctl: Unknown cmd fd(%d) "
-				       "cmd(%08x) arg(%08x)\n",
-				       (int)fd, (unsigned int)cmd, (unsigned int)arg);
-		} while(0);
-		return -EINVAL;
 	}
 	set_fs (KERNEL_DS);
 	err = sys_ioctl (fd, kcmd, (unsigned long)karg);
@@ -598,11 +551,11 @@ static int mt_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
 		return err;
 	switch (cmd) {
 	case MTIOCPOS32:
-		upos32 = compat_ptr(arg);
+		upos32 = argp;
 		err = __put_user(pos.mt_blkno, &upos32->mt_blkno);
 		break;
 	case MTIOCGET32:
-		umget32 = compat_ptr(arg);
+		umget32 = argp;
 		err = __put_user(get.mt_type, &umget32->mt_type);
 		err |= __put_user(get.mt_resid, &umget32->mt_resid);
 		err |= __put_user(get.mt_dsreg, &umget32->mt_dsreg);
@@ -617,162 +570,8 @@ static int mt_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
 
 #endif /* CONFIG_BLOCK */
 
-#ifdef CONFIG_VT
-
-static int vt_check(struct file *file)
-{
-	struct tty_struct *tty;
-	struct inode *inode = file->f_path.dentry->d_inode;
-	struct vc_data *vc;
-	
-	if (file->f_op->unlocked_ioctl != tty_ioctl)
-		return -EINVAL;
-	                
-	tty = (struct tty_struct *)file->private_data;
-	if (tty_paranoia_check(tty, inode, "tty_ioctl"))
-		return -EINVAL;
-	                                                
-	if (tty->ops->ioctl != vt_ioctl)
-		return -EINVAL;
-
-	vc = (struct vc_data *)tty->driver_data;
-	if (!vc_cons_allocated(vc->vc_num)) 	/* impossible? */
-		return -ENOIOCTLCMD;
-
-	/*
-	 * To have permissions to do most of the vt ioctls, we either have
-	 * to be the owner of the tty, or have CAP_SYS_TTY_CONFIG.
-	 */
-	if (current->signal->tty == tty || capable(CAP_SYS_TTY_CONFIG))
-		return 1;
-	return 0;                                                    
-}
-
-struct consolefontdesc32 {
-	unsigned short charcount;       /* characters in font (256 or 512) */
-	unsigned short charheight;      /* scan lines per character (1-32) */
-	compat_caddr_t chardata;	/* font data in expanded form */
-};
-
-static int do_fontx_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg, struct file *file)
-{
-	struct consolefontdesc32 __user *user_cfd = compat_ptr(arg);
-	struct console_font_op op;
-	compat_caddr_t data;
-	int i, perm;
-
-	perm = vt_check(file);
-	if (perm < 0) return perm;
-	
-	switch (cmd) {
-	case PIO_FONTX:
-		if (!perm)
-			return -EPERM;
-		op.op = KD_FONT_OP_SET;
-		op.flags = 0;
-		op.width = 8;
-		if (get_user(op.height, &user_cfd->charheight) ||
-		    get_user(op.charcount, &user_cfd->charcount) ||
-		    get_user(data, &user_cfd->chardata))
-			return -EFAULT;
-		op.data = compat_ptr(data);
-		return con_font_op(vc_cons[fg_console].d, &op);
-	case GIO_FONTX:
-		op.op = KD_FONT_OP_GET;
-		op.flags = 0;
-		op.width = 8;
-		if (get_user(op.height, &user_cfd->charheight) ||
-		    get_user(op.charcount, &user_cfd->charcount) ||
-		    get_user(data, &user_cfd->chardata))
-			return -EFAULT;
-		if (!data)
-			return 0;
-		op.data = compat_ptr(data);
-		i = con_font_op(vc_cons[fg_console].d, &op);
-		if (i)
-			return i;
-		if (put_user(op.height, &user_cfd->charheight) ||
-		    put_user(op.charcount, &user_cfd->charcount) ||
-		    put_user((compat_caddr_t)(unsigned long)op.data,
-				&user_cfd->chardata))
-			return -EFAULT;
-		return 0;
-	}
-	return -EINVAL;
-}
-
-struct console_font_op32 {
-	compat_uint_t op;        /* operation code KD_FONT_OP_* */
-	compat_uint_t flags;     /* KD_FONT_FLAG_* */
-	compat_uint_t width, height;     /* font size */
-	compat_uint_t charcount;
-	compat_caddr_t data;    /* font data with height fixed to 32 */
-};
-                                        
-static int do_kdfontop_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg, struct file *file)
-{
-	struct console_font_op op;
-	struct console_font_op32 __user *fontop = compat_ptr(arg);
-	int perm = vt_check(file), i;
-	struct vc_data *vc;
-	
-	if (perm < 0) return perm;
-	
-	if (copy_from_user(&op, fontop, sizeof(struct console_font_op32)))
-		return -EFAULT;
-	if (!perm && op.op != KD_FONT_OP_GET)
-		return -EPERM;
-	op.data = compat_ptr(((struct console_font_op32 *)&op)->data);
-	op.flags |= KD_FONT_FLAG_OLD;
-	vc = ((struct tty_struct *)file->private_data)->driver_data;
-	i = con_font_op(vc, &op);
-	if (i)
-		return i;
-	((struct console_font_op32 *)&op)->data = (unsigned long)op.data;
-	if (copy_to_user(fontop, &op, sizeof(struct console_font_op32)))
-		return -EFAULT;
-	return 0;
-}
-
-struct unimapdesc32 {
-	unsigned short entry_ct;
-	compat_caddr_t entries;
-};
-
-static int do_unimap_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg, struct file *file)
-{
-	struct unimapdesc32 tmp;
-	struct unimapdesc32 __user *user_ud = compat_ptr(arg);
-	int perm = vt_check(file);
-	struct vc_data *vc;
-
-	if (perm < 0)
-		return perm;
-	if (copy_from_user(&tmp, user_ud, sizeof tmp))
-		return -EFAULT;
-	if (tmp.entries)
-		if (!access_ok(VERIFY_WRITE, compat_ptr(tmp.entries),
-				tmp.entry_ct*sizeof(struct unipair)))
-			return -EFAULT;
-	vc = ((struct tty_struct *)file->private_data)->driver_data;
-	switch (cmd) {
-	case PIO_UNIMAP:
-		if (!perm)
-			return -EPERM;
-		return con_set_unimap(vc, tmp.entry_ct,
-						compat_ptr(tmp.entries));
-	case GIO_UNIMAP:
-		if (!perm && fg_console != vc->vc_num)
-			return -EPERM;
-		return con_get_unimap(vc, tmp.entry_ct, &(user_ud->entry_ct),
-						compat_ptr(tmp.entries));
-	}
-	return 0;
-}
-
-#endif /* CONFIG_VT */
-
-static int do_smb_getmountuid(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int do_smb_getmountuid(unsigned int fd, unsigned int cmd,
+			compat_uid_t __user *argp)
 {
 	mm_segment_t old_fs = get_fs();
 	__kernel_uid_t kuid;
@@ -785,20 +584,15 @@ static int do_smb_getmountuid(unsigned int fd, unsigned int cmd, unsigned long a
 	set_fs(old_fs);
 
 	if (err >= 0)
-		err = put_user(kuid, (compat_uid_t __user *)compat_ptr(arg));
+		err = put_user(kuid, argp);
 
 	return err;
 }
 
-static __used int
-ret_einval(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	return -EINVAL;
-}
-
-static int ioc_settimeout(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int ioc_settimeout(unsigned int fd, unsigned int cmd,
+		compat_ulong_t __user *argp)
 {
-	return rw_long(fd, AUTOFS_IOC_SETTIMEOUT, arg);
+	return rw_long(fd, AUTOFS_IOC_SETTIMEOUT, argp);
 }
 
 /* Bluetooth ioctls */
@@ -856,7 +650,8 @@ static int set_raw32_request(struct raw_config_request *req, struct raw32_config
         return ret ? -EFAULT : 0;
 }
 
-static int raw_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
+static int raw_ioctl(unsigned fd, unsigned cmd,
+		struct raw32_config_request __user *user_req)
 {
         int ret;
 
@@ -864,7 +659,6 @@ static int raw_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
         case RAW_SETBIND:
         case RAW_GETBIND: {
                 struct raw_config_request req;
-                struct raw32_config_request __user *user_req = compat_ptr(arg);
                 mm_segment_t oldfs = get_fs();
 
                 if ((ret = get_raw32_request(&req, user_req)))
@@ -879,9 +673,6 @@ static int raw_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
                 }
                 break;
         }
-        default:
-                ret = sys_ioctl(fd, cmd, arg);
-                break;
         }
         return ret;
 }
@@ -909,11 +700,11 @@ struct serial_struct32 {
         compat_int_t    reserved[1];
 };
 
-static int serial_struct_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
+static int serial_struct_ioctl(unsigned fd, unsigned cmd,
+			struct serial_struct32 __user *ss32)
 {
         typedef struct serial_struct SS;
         typedef struct serial_struct32 SS32;
-        struct serial_struct32 __user *ss32 = compat_ptr(arg);
         int err;
         struct serial_struct ss;
         mm_segment_t oldseg = get_fs();
@@ -951,96 +742,6 @@ static int serial_struct_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
         return err;
 }
 
-struct usbdevfs_ctrltransfer32 {
-        u8 bRequestType;
-        u8 bRequest;
-        u16 wValue;
-        u16 wIndex;
-        u16 wLength;
-        u32 timeout;  /* in milliseconds */
-        compat_caddr_t data;
-};
-
-#define USBDEVFS_CONTROL32           _IOWR('U', 0, struct usbdevfs_ctrltransfer32)
-
-static int do_usbdevfs_control(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-        struct usbdevfs_ctrltransfer32 __user *p32 = compat_ptr(arg);
-        struct usbdevfs_ctrltransfer __user *p;
-        __u32 udata;
-        p = compat_alloc_user_space(sizeof(*p));
-        if (copy_in_user(p, p32, (sizeof(*p32) - sizeof(compat_caddr_t))) ||
-            get_user(udata, &p32->data) ||
-	    put_user(compat_ptr(udata), &p->data))
-		return -EFAULT;
-        return sys_ioctl(fd, USBDEVFS_CONTROL, (unsigned long)p);
-}
-
-
-struct usbdevfs_bulktransfer32 {
-        compat_uint_t ep;
-        compat_uint_t len;
-        compat_uint_t timeout; /* in milliseconds */
-        compat_caddr_t data;
-};
-
-#define USBDEVFS_BULK32              _IOWR('U', 2, struct usbdevfs_bulktransfer32)
-
-static int do_usbdevfs_bulk(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-        struct usbdevfs_bulktransfer32 __user *p32 = compat_ptr(arg);
-        struct usbdevfs_bulktransfer __user *p;
-        compat_uint_t n;
-        compat_caddr_t addr;
-
-        p = compat_alloc_user_space(sizeof(*p));
-
-        if (get_user(n, &p32->ep) || put_user(n, &p->ep) ||
-            get_user(n, &p32->len) || put_user(n, &p->len) ||
-            get_user(n, &p32->timeout) || put_user(n, &p->timeout) ||
-            get_user(addr, &p32->data) || put_user(compat_ptr(addr), &p->data))
-                return -EFAULT;
-
-        return sys_ioctl(fd, USBDEVFS_BULK, (unsigned long)p);
-}
-
-
-/*
- *  USBDEVFS_SUBMITURB, USBDEVFS_REAPURB and USBDEVFS_REAPURBNDELAY
- *  are handled in usbdevfs core.			-Christopher Li
- */
-
-struct usbdevfs_disconnectsignal32 {
-        compat_int_t signr;
-        compat_caddr_t context;
-};
-
-#define USBDEVFS_DISCSIGNAL32      _IOR('U', 14, struct usbdevfs_disconnectsignal32)
-
-static int do_usbdevfs_discsignal(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-        struct usbdevfs_disconnectsignal kdis;
-        struct usbdevfs_disconnectsignal32 __user *udis;
-        mm_segment_t old_fs;
-        u32 uctx;
-        int err;
-
-        udis = compat_ptr(arg);
-
-        if (get_user(kdis.signr, &udis->signr) ||
-            __get_user(uctx, &udis->context))
-                return -EFAULT;
-
-        kdis.context = compat_ptr(uctx);
-
-        old_fs = get_fs();
-        set_fs(KERNEL_DS);
-        err = sys_ioctl(fd, USBDEVFS_DISCSIGNAL, (unsigned long) &kdis);
-        set_fs(old_fs);
-
-        return err;
-}
-
 /*
  * I2C layer ioctls
  */
@@ -1069,9 +770,9 @@ struct i2c_rdwr_aligned {
 	struct i2c_msg msgs[0];
 };
 
-static int do_i2c_rdwr_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int do_i2c_rdwr_ioctl(unsigned int fd, unsigned int cmd,
+			struct i2c_rdwr_ioctl_data32    __user *udata)
 {
-	struct i2c_rdwr_ioctl_data32	__user *udata = compat_ptr(arg);
 	struct i2c_rdwr_aligned		__user *tdata;
 	struct i2c_msg			__user *tmsgs;
 	struct i2c_msg32		__user *umsgs;
@@ -1105,10 +806,10 @@ static int do_i2c_rdwr_ioctl(unsigned int fd, unsigned int cmd, unsigned long ar
 	return sys_ioctl(fd, cmd, (unsigned long)tdata);
 }
 
-static int do_i2c_smbus_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
+static int do_i2c_smbus_ioctl(unsigned int fd, unsigned int cmd,
+			struct i2c_smbus_ioctl_data32   __user *udata)
 {
 	struct i2c_smbus_ioctl_data	__user *tdata;
-	struct i2c_smbus_ioctl_data32	__user *udata;
 	compat_caddr_t			datap;
 
 	tdata = compat_alloc_user_space(sizeof(*tdata));
@@ -1117,7 +818,6 @@ static int do_i2c_smbus_ioctl(unsigned int fd, unsigned int cmd, unsigned long a
 	if (!access_ok(VERIFY_WRITE, tdata, sizeof(*tdata)))
 		return -EFAULT;
 
-	udata = compat_ptr(arg);
 	if (!access_ok(VERIFY_READ, udata, sizeof(*udata)))
 		return -EFAULT;
 
@@ -1137,7 +837,7 @@ static int do_i2c_smbus_ioctl(unsigned int fd, unsigned int cmd, unsigned long a
 #define RTC_EPOCH_READ32	_IOR('p', 0x0d, compat_ulong_t)
 #define RTC_EPOCH_SET32		_IOW('p', 0x0e, compat_ulong_t)
 
-static int rtc_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
+static int rtc_ioctl(unsigned fd, unsigned cmd, void __user *argp)
 {
 	mm_segment_t oldfs = get_fs();
 	compat_ulong_t val32;
@@ -1155,29 +855,14 @@ static int rtc_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
 		if (ret)
 			return ret;
 		val32 = kval;
-		return put_user(val32, (unsigned int __user *)arg);
+		return put_user(val32, (unsigned int __user *)argp);
 	case RTC_IRQP_SET32:
-		return sys_ioctl(fd, RTC_IRQP_SET, arg); 
+		return sys_ioctl(fd, RTC_IRQP_SET, (unsigned long)argp);
 	case RTC_EPOCH_SET32:
-		return sys_ioctl(fd, RTC_EPOCH_SET, arg);
-	default:
-		/* unreached */
-		return -ENOIOCTLCMD;
+		return sys_ioctl(fd, RTC_EPOCH_SET, (unsigned long)argp);
 	}
-}
 
-static int
-lp_timeout_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
-{
-	struct compat_timeval __user *tc = (struct compat_timeval __user *)arg;
-	struct timeval __user *tn = compat_alloc_user_space(sizeof(struct timeval));
-	struct timeval ts;
-	if (get_user(ts.tv_sec, &tc->tv_sec) ||
-	    get_user(ts.tv_usec, &tc->tv_usec) ||
-	    put_user(ts.tv_sec, &tn->tv_sec) ||
-	    put_user(ts.tv_usec, &tn->tv_usec))
-		return -EFAULT;
-	return sys_ioctl(fd, cmd, (unsigned long)tn);
+	return -ENOIOCTLCMD;
 }
 
 /* on ia32 l_start is on a 32-bit boundary */
@@ -1197,9 +882,9 @@ struct space_resv_32 {
 #define FS_IOC_RESVSP64_32	_IOW ('X', 42, struct space_resv_32)
 
 /* just account for different alignment */
-static int compat_ioctl_preallocate(struct file *file, unsigned long arg)
+static int compat_ioctl_preallocate(struct file *file,
+			struct space_resv_32    __user *p32)
 {
-	struct space_resv_32	__user *p32 = compat_ptr(arg);
 	struct space_resv	__user *p = compat_alloc_user_space(sizeof(*p));
 
 	if (copy_in_user(&p->l_type,	&p32->l_type,	sizeof(s16)) ||
@@ -1215,27 +900,13 @@ static int compat_ioctl_preallocate(struct file *file, unsigned long arg)
 }
 #endif
 
+/*
+ * simple reversible transform to make our table more evenly
+ * distributed after sorting.
+ */
+#define XFORM(i) (((i) ^ ((i) << 27) ^ ((i) << 17)) & 0xffffffff)
 
-typedef int (*ioctl_trans_handler_t)(unsigned int, unsigned int,
-					unsigned long, struct file *);
-
-struct ioctl_trans {
-	unsigned long cmd;
-	ioctl_trans_handler_t handler;
-	struct ioctl_trans *next;
-};
-
-#define HANDLE_IOCTL(cmd,handler) \
-	{ (cmd), (ioctl_trans_handler_t)(handler) },
-
-/* pointer to compatible structure or no argument */
-#define COMPATIBLE_IOCTL(cmd) \
-	{ (cmd), do_ioctl32_pointer },
-
-/* argument is an unsigned long integer, not a pointer */
-#define ULONG_IOCTL(cmd) \
-	{ (cmd), (ioctl_trans_handler_t)sys_ioctl },
-
+#define COMPATIBLE_IOCTL(cmd) XFORM(cmd),
 /* ioctl should not be warned about even if it's not implemented.
    Valid reasons to use this:
    - It is implemented with ->compat_ioctl on some device, but programs
@@ -1245,7 +916,7 @@ struct ioctl_trans {
    Most other reasons are not valid. */
 #define IGNORE_IOCTL(cmd) COMPATIBLE_IOCTL(cmd)
 
-static struct ioctl_trans ioctl_start[] = {
+static unsigned int ioctl_pointer[] = {
 /* compatible ioctls first */
 COMPATIBLE_IOCTL(0x4B50)   /* KDGHWCLK - not in the kernel, but don't complain */
 COMPATIBLE_IOCTL(0x4B51)   /* KDSHWCLK - not in the kernel, but don't complain */
@@ -1256,7 +927,6 @@ COMPATIBLE_IOCTL(TCSETA)
 COMPATIBLE_IOCTL(TCSETAW)
 COMPATIBLE_IOCTL(TCSETAF)
 COMPATIBLE_IOCTL(TCSBRK)
-ULONG_IOCTL(TCSBRKP)
 COMPATIBLE_IOCTL(TCXONC)
 COMPATIBLE_IOCTL(TCFLSH)
 COMPATIBLE_IOCTL(TCGETS)
@@ -1266,7 +936,6 @@ COMPATIBLE_IOCTL(TCSETSF)
 COMPATIBLE_IOCTL(TIOCLINUX)
 COMPATIBLE_IOCTL(TIOCSBRK)
 COMPATIBLE_IOCTL(TIOCCBRK)
-ULONG_IOCTL(TIOCMIWAIT)
 COMPATIBLE_IOCTL(TIOCGICOUNT)
 /* Little t */
 COMPATIBLE_IOCTL(TIOCGETD)
@@ -1288,7 +957,6 @@ COMPATIBLE_IOCTL(TIOCSTI)
 COMPATIBLE_IOCTL(TIOCOUTQ)
 COMPATIBLE_IOCTL(TIOCSPGRP)
 COMPATIBLE_IOCTL(TIOCGPGRP)
-ULONG_IOCTL(TIOCSCTTY)
 COMPATIBLE_IOCTL(TIOCGPTN)
 COMPATIBLE_IOCTL(TIOCSPTLCK)
 COMPATIBLE_IOCTL(TIOCSERGETLSR)
@@ -1311,44 +979,11 @@ COMPATIBLE_IOCTL(FIGETBSZ)
 /* 'X' - originally XFS but some now in the VFS */
 COMPATIBLE_IOCTL(FIFREEZE)
 COMPATIBLE_IOCTL(FITHAW)
-/* RAID */
-COMPATIBLE_IOCTL(RAID_VERSION)
-COMPATIBLE_IOCTL(GET_ARRAY_INFO)
-COMPATIBLE_IOCTL(GET_DISK_INFO)
-COMPATIBLE_IOCTL(PRINT_RAID_DEBUG)
-COMPATIBLE_IOCTL(RAID_AUTORUN)
-COMPATIBLE_IOCTL(CLEAR_ARRAY)
-COMPATIBLE_IOCTL(ADD_NEW_DISK)
-ULONG_IOCTL(HOT_REMOVE_DISK)
-COMPATIBLE_IOCTL(SET_ARRAY_INFO)
-COMPATIBLE_IOCTL(SET_DISK_INFO)
-COMPATIBLE_IOCTL(WRITE_RAID_INFO)
-COMPATIBLE_IOCTL(UNPROTECT_ARRAY)
-COMPATIBLE_IOCTL(PROTECT_ARRAY)
-ULONG_IOCTL(HOT_ADD_DISK)
-ULONG_IOCTL(SET_DISK_FAULTY)
-COMPATIBLE_IOCTL(RUN_ARRAY)
-COMPATIBLE_IOCTL(STOP_ARRAY)
-COMPATIBLE_IOCTL(STOP_ARRAY_RO)
-COMPATIBLE_IOCTL(RESTART_ARRAY_RW)
-COMPATIBLE_IOCTL(GET_BITMAP_FILE)
-ULONG_IOCTL(SET_BITMAP_FILE)
-/* Big K */
-COMPATIBLE_IOCTL(PIO_FONT)
-COMPATIBLE_IOCTL(GIO_FONT)
-COMPATIBLE_IOCTL(PIO_CMAP)
-COMPATIBLE_IOCTL(GIO_CMAP)
-ULONG_IOCTL(KDSIGACCEPT)
 COMPATIBLE_IOCTL(KDGETKEYCODE)
 COMPATIBLE_IOCTL(KDSETKEYCODE)
-ULONG_IOCTL(KIOCSOUND)
-ULONG_IOCTL(KDMKTONE)
 COMPATIBLE_IOCTL(KDGKBTYPE)
-ULONG_IOCTL(KDSETMODE)
 COMPATIBLE_IOCTL(KDGETMODE)
-ULONG_IOCTL(KDSKBMODE)
 COMPATIBLE_IOCTL(KDGKBMODE)
-ULONG_IOCTL(KDSKBMETA)
 COMPATIBLE_IOCTL(KDGKBMETA)
 COMPATIBLE_IOCTL(KDGKBENT)
 COMPATIBLE_IOCTL(KDSKBENT)
@@ -1358,15 +993,7 @@ COMPATIBLE_IOCTL(KDGKBDIACR)
 COMPATIBLE_IOCTL(KDSKBDIACR)
 COMPATIBLE_IOCTL(KDKBDREP)
 COMPATIBLE_IOCTL(KDGKBLED)
-ULONG_IOCTL(KDSKBLED)
 COMPATIBLE_IOCTL(KDGETLED)
-ULONG_IOCTL(KDSETLED)
-COMPATIBLE_IOCTL(GIO_SCRNMAP)
-COMPATIBLE_IOCTL(PIO_SCRNMAP)
-COMPATIBLE_IOCTL(GIO_UNISCRNMAP)
-COMPATIBLE_IOCTL(PIO_UNISCRNMAP)
-COMPATIBLE_IOCTL(PIO_FONTRESET)
-COMPATIBLE_IOCTL(PIO_UNIMAPCLR)
 #ifdef CONFIG_BLOCK
 /* Big S */
 COMPATIBLE_IOCTL(SCSI_IOCTL_GET_IDLUN)
@@ -1378,20 +1005,9 @@ COMPATIBLE_IOCTL(SCSI_IOCTL_SEND_COMMAND)
 COMPATIBLE_IOCTL(SCSI_IOCTL_PROBE_HOST)
 COMPATIBLE_IOCTL(SCSI_IOCTL_GET_PCI)
 #endif
-/* Big V */
-COMPATIBLE_IOCTL(VT_SETMODE)
-COMPATIBLE_IOCTL(VT_GETMODE)
-COMPATIBLE_IOCTL(VT_GETSTATE)
-COMPATIBLE_IOCTL(VT_OPENQRY)
-ULONG_IOCTL(VT_ACTIVATE)
-ULONG_IOCTL(VT_WAITACTIVE)
-ULONG_IOCTL(VT_RELDISP)
-ULONG_IOCTL(VT_DISALLOCATE)
-COMPATIBLE_IOCTL(VT_RESIZE)
-COMPATIBLE_IOCTL(VT_RESIZEX)
-COMPATIBLE_IOCTL(VT_LOCKSWITCH)
-COMPATIBLE_IOCTL(VT_UNLOCKSWITCH)
-COMPATIBLE_IOCTL(VT_GETHIFONTMASK)
+/* Big V (don't complain on serial console) */
+IGNORE_IOCTL(VT_OPENQRY)
+IGNORE_IOCTL(VT_GETMODE)
 /* Little p (/dev/rtc, /dev/envctrl, etc.) */
 COMPATIBLE_IOCTL(RTC_AIE_ON)
 COMPATIBLE_IOCTL(RTC_AIE_OFF)
@@ -1420,11 +1036,12 @@ COMPATIBLE_IOCTL(MTIOCTOP)
 /* Socket level stuff */
 COMPATIBLE_IOCTL(FIOQSIZE)
 #ifdef CONFIG_BLOCK
+/* loop */
+IGNORE_IOCTL(LOOP_CLR_FD)
 /* SG stuff */
 COMPATIBLE_IOCTL(SG_SET_TIMEOUT)
 COMPATIBLE_IOCTL(SG_GET_TIMEOUT)
 COMPATIBLE_IOCTL(SG_EMULATED_HOST)
-ULONG_IOCTL(SG_SET_TRANSFORM)
 COMPATIBLE_IOCTL(SG_GET_TRANSFORM)
 COMPATIBLE_IOCTL(SG_SET_RESERVED_SIZE)
 COMPATIBLE_IOCTL(SG_GET_RESERVED_SIZE)
@@ -1478,8 +1095,6 @@ COMPATIBLE_IOCTL(PPPIOCGCHAN)
 /* PPPOX */
 COMPATIBLE_IOCTL(PPPOEIOCSFWD)
 COMPATIBLE_IOCTL(PPPOEIOCDFWD)
-/* LP */
-COMPATIBLE_IOCTL(LPGETSTATUS)
 /* ppdev */
 COMPATIBLE_IOCTL(PPSETMODE)
 COMPATIBLE_IOCTL(PPRSTATUS)
@@ -1661,8 +1276,6 @@ COMPATIBLE_IOCTL(SOUND_MIXER_GETLEVELS)
 COMPATIBLE_IOCTL(SOUND_MIXER_SETLEVELS)
 COMPATIBLE_IOCTL(OSS_GETVERSION)
 /* AUTOFS */
-ULONG_IOCTL(AUTOFS_IOC_READY)
-ULONG_IOCTL(AUTOFS_IOC_FAIL)
 COMPATIBLE_IOCTL(AUTOFS_IOC_CATATONIC)
 COMPATIBLE_IOCTL(AUTOFS_IOC_PROTOVER)
 COMPATIBLE_IOCTL(AUTOFS_IOC_EXPIRE)
@@ -1755,30 +1368,11 @@ COMPATIBLE_IOCTL(PCIIOC_CONTROLLER)
 COMPATIBLE_IOCTL(PCIIOC_MMAP_IS_IO)
 COMPATIBLE_IOCTL(PCIIOC_MMAP_IS_MEM)
 COMPATIBLE_IOCTL(PCIIOC_WRITE_COMBINE)
-/* USB */
-COMPATIBLE_IOCTL(USBDEVFS_RESETEP)
-COMPATIBLE_IOCTL(USBDEVFS_SETINTERFACE)
-COMPATIBLE_IOCTL(USBDEVFS_SETCONFIGURATION)
-COMPATIBLE_IOCTL(USBDEVFS_GETDRIVER)
-COMPATIBLE_IOCTL(USBDEVFS_DISCARDURB)
-COMPATIBLE_IOCTL(USBDEVFS_CLAIMINTERFACE)
-COMPATIBLE_IOCTL(USBDEVFS_RELEASEINTERFACE)
-COMPATIBLE_IOCTL(USBDEVFS_CONNECTINFO)
-COMPATIBLE_IOCTL(USBDEVFS_HUB_PORTINFO)
-COMPATIBLE_IOCTL(USBDEVFS_RESET)
-COMPATIBLE_IOCTL(USBDEVFS_SUBMITURB32)
-COMPATIBLE_IOCTL(USBDEVFS_REAPURB32)
-COMPATIBLE_IOCTL(USBDEVFS_REAPURBNDELAY32)
-COMPATIBLE_IOCTL(USBDEVFS_CLEAR_HALT)
 /* NBD */
-ULONG_IOCTL(NBD_SET_SOCK)
-ULONG_IOCTL(NBD_SET_BLKSIZE)
-ULONG_IOCTL(NBD_SET_SIZE)
 COMPATIBLE_IOCTL(NBD_DO_IT)
 COMPATIBLE_IOCTL(NBD_CLEAR_SOCK)
 COMPATIBLE_IOCTL(NBD_CLEAR_QUE)
 COMPATIBLE_IOCTL(NBD_PRINT_DEBUG)
-ULONG_IOCTL(NBD_SET_SIZE_BLOCKS)
 COMPATIBLE_IOCTL(NBD_DISCONNECT)
 /* i2c */
 COMPATIBLE_IOCTL(I2C_SLAVE)
@@ -1878,42 +1472,6 @@ COMPATIBLE_IOCTL(JSIOCGAXES)
 COMPATIBLE_IOCTL(JSIOCGBUTTONS)
 COMPATIBLE_IOCTL(JSIOCGNAME(0))
 
-/* now things that need handlers */
-#ifdef CONFIG_BLOCK
-HANDLE_IOCTL(SG_IO,sg_ioctl_trans)
-HANDLE_IOCTL(SG_GET_REQUEST_TABLE, sg_grt_trans)
-#endif
-HANDLE_IOCTL(PPPIOCGIDLE32, ppp_ioctl_trans)
-HANDLE_IOCTL(PPPIOCSCOMPRESS32, ppp_ioctl_trans)
-HANDLE_IOCTL(PPPIOCSPASS32, ppp_sock_fprog_ioctl_trans)
-HANDLE_IOCTL(PPPIOCSACTIVE32, ppp_sock_fprog_ioctl_trans)
-#ifdef CONFIG_BLOCK
-HANDLE_IOCTL(MTIOCGET32, mt_ioctl_trans)
-HANDLE_IOCTL(MTIOCPOS32, mt_ioctl_trans)
-#endif
-#define AUTOFS_IOC_SETTIMEOUT32 _IOWR(0x93,0x64,unsigned int)
-HANDLE_IOCTL(AUTOFS_IOC_SETTIMEOUT32, ioc_settimeout)
-#ifdef CONFIG_VT
-HANDLE_IOCTL(PIO_FONTX, do_fontx_ioctl)
-HANDLE_IOCTL(GIO_FONTX, do_fontx_ioctl)
-HANDLE_IOCTL(PIO_UNIMAP, do_unimap_ioctl)
-HANDLE_IOCTL(GIO_UNIMAP, do_unimap_ioctl)
-HANDLE_IOCTL(KDFONTOP, do_kdfontop_ioctl)
-#endif
-/* One SMB ioctl needs translations. */
-#define SMB_IOC_GETMOUNTUID_32 _IOR('u', 1, compat_uid_t)
-HANDLE_IOCTL(SMB_IOC_GETMOUNTUID_32, do_smb_getmountuid)
-/* block stuff */
-#ifdef CONFIG_BLOCK
-/* loop */
-IGNORE_IOCTL(LOOP_CLR_FD)
-/* Raw devices */
-HANDLE_IOCTL(RAW_SETBIND, raw_ioctl)
-HANDLE_IOCTL(RAW_GETBIND, raw_ioctl)
-#endif
-/* Serial */
-HANDLE_IOCTL(TIOCGSERIAL, serial_struct_ioctl)
-HANDLE_IOCTL(TIOCSSERIAL, serial_struct_ioctl)
 #ifdef TIOCGLTC
 COMPATIBLE_IOCTL(TIOCGLTC)
 COMPATIBLE_IOCTL(TIOCSLTC)
@@ -1928,39 +1486,6 @@ COMPATIBLE_IOCTL(TIOCSLTC)
 COMPATIBLE_IOCTL(TIOCSTART)
 COMPATIBLE_IOCTL(TIOCSTOP)
 #endif
-/* Usbdevfs */
-HANDLE_IOCTL(USBDEVFS_CONTROL32, do_usbdevfs_control)
-HANDLE_IOCTL(USBDEVFS_BULK32, do_usbdevfs_bulk)
-HANDLE_IOCTL(USBDEVFS_DISCSIGNAL32, do_usbdevfs_discsignal)
-COMPATIBLE_IOCTL(USBDEVFS_IOCTL32)
-/* i2c */
-HANDLE_IOCTL(I2C_FUNCS, w_long)
-HANDLE_IOCTL(I2C_RDWR, do_i2c_rdwr_ioctl)
-HANDLE_IOCTL(I2C_SMBUS, do_i2c_smbus_ioctl)
-/* Not implemented in the native kernel */
-HANDLE_IOCTL(RTC_IRQP_READ32, rtc_ioctl)
-HANDLE_IOCTL(RTC_IRQP_SET32, rtc_ioctl)
-HANDLE_IOCTL(RTC_EPOCH_READ32, rtc_ioctl)
-HANDLE_IOCTL(RTC_EPOCH_SET32, rtc_ioctl)
-
-/* dvb */
-HANDLE_IOCTL(VIDEO_GET_EVENT, do_video_get_event)
-HANDLE_IOCTL(VIDEO_STILLPICTURE, do_video_stillpicture)
-HANDLE_IOCTL(VIDEO_SET_SPU_PALETTE, do_video_set_spu_palette)
-
-/* parport */
-COMPATIBLE_IOCTL(LPTIME)
-COMPATIBLE_IOCTL(LPCHAR)
-COMPATIBLE_IOCTL(LPABORTOPEN)
-COMPATIBLE_IOCTL(LPCAREFUL)
-COMPATIBLE_IOCTL(LPWAIT)
-COMPATIBLE_IOCTL(LPSETIRQ)
-COMPATIBLE_IOCTL(LPGETSTATUS)
-COMPATIBLE_IOCTL(LPGETSTATUS)
-COMPATIBLE_IOCTL(LPRESET)
-/*LPGETSTATS not implemented, but no kernels seem to compile it in anyways*/
-COMPATIBLE_IOCTL(LPGETFLAGS)
-HANDLE_IOCTL(LPSETTIMEOUT, lp_timeout_trans)
 
 /* fat 'r' ioctls. These are handled by fat with ->compat_ioctl,
    but we don't want warnings on other file systems. So declare
@@ -1988,12 +1513,108 @@ IGNORE_IOCTL(FBIOGCURSOR32)
 #endif
 };
 
-#define IOCTL_HASHSIZE 256
-static struct ioctl_trans *ioctl32_hash_table[IOCTL_HASHSIZE];
-
-static inline unsigned long ioctl32_hash(unsigned long cmd)
+/*
+ * Convert common ioctl arguments based on their command number
+ *
+ * Please do not add any code in here. Instead, implement
+ * a compat_ioctl operation in the place that handleѕ the
+ * ioctl for the native case.
+ */
+static long do_ioctl_trans(int fd, unsigned int cmd,
+		 unsigned long arg, struct file *file)
 {
-	return (((cmd >> 6) ^ (cmd >> 4) ^ cmd)) % IOCTL_HASHSIZE;
+	void __user *argp = compat_ptr(arg);
+
+	switch (cmd) {
+	case PPPIOCGIDLE32:
+		return ppp_gidle(fd, cmd, argp);
+	case PPPIOCSCOMPRESS32:
+		return ppp_scompress(fd, cmd, argp);
+	case PPPIOCSPASS32:
+	case PPPIOCSACTIVE32:
+		return ppp_sock_fprog_ioctl_trans(fd, cmd, argp);
+#ifdef CONFIG_BLOCK
+	case SG_IO:
+		return sg_ioctl_trans(fd, cmd, argp);
+	case SG_GET_REQUEST_TABLE:
+		return sg_grt_trans(fd, cmd, argp);
+	case MTIOCGET32:
+	case MTIOCPOS32:
+		return mt_ioctl_trans(fd, cmd, argp);
+	/* Raw devices */
+	case RAW_SETBIND:
+	case RAW_GETBIND:
+		return raw_ioctl(fd, cmd, argp);
+#endif
+#define AUTOFS_IOC_SETTIMEOUT32 _IOWR(0x93,0x64,unsigned int)
+	case AUTOFS_IOC_SETTIMEOUT32:
+		return ioc_settimeout(fd, cmd, argp);
+	/* One SMB ioctl needs translations. */
+#define SMB_IOC_GETMOUNTUID_32 _IOR('u', 1, compat_uid_t)
+	case SMB_IOC_GETMOUNTUID_32:
+		return do_smb_getmountuid(fd, cmd, argp);
+	/* Serial */
+	case TIOCGSERIAL:
+	case TIOCSSERIAL:
+		return serial_struct_ioctl(fd, cmd, argp);
+	/* i2c */
+	case I2C_FUNCS:
+		return w_long(fd, cmd, argp);
+	case I2C_RDWR:
+		return do_i2c_rdwr_ioctl(fd, cmd, argp);
+	case I2C_SMBUS:
+		return do_i2c_smbus_ioctl(fd, cmd, argp);
+	/* Not implemented in the native kernel */
+	case RTC_IRQP_READ32:
+	case RTC_IRQP_SET32:
+	case RTC_EPOCH_READ32:
+	case RTC_EPOCH_SET32:
+		return rtc_ioctl(fd, cmd, argp);
+
+	/* dvb */
+	case VIDEO_GET_EVENT:
+		return do_video_get_event(fd, cmd, argp);
+	case VIDEO_STILLPICTURE:
+		return do_video_stillpicture(fd, cmd, argp);
+	case VIDEO_SET_SPU_PALETTE:
+		return do_video_set_spu_palette(fd, cmd, argp);
+	}
+
+	/*
+	 * These take an integer instead of a pointer as 'arg',
+	 * so we must not do a compat_ptr() translation.
+	 */
+	switch (cmd) {
+	/* Big T */
+	case TCSBRKP:
+	case TIOCMIWAIT:
+	case TIOCSCTTY:
+	/* RAID */
+	case HOT_REMOVE_DISK:
+	case HOT_ADD_DISK:
+	case SET_DISK_FAULTY:
+	case SET_BITMAP_FILE:
+	/* Big K */
+	case KDSIGACCEPT:
+	case KIOCSOUND:
+	case KDMKTONE:
+	case KDSETMODE:
+	case KDSKBMODE:
+	case KDSKBMETA:
+	case KDSKBLED:
+	case KDSETLED:
+	/* AUTOFS */
+	case AUTOFS_IOC_READY:
+	case AUTOFS_IOC_FAIL:
+	/* NBD */
+	case NBD_SET_SOCK:
+	case NBD_SET_BLKSIZE:
+	case NBD_SET_SIZE:
+	case NBD_SET_SIZE_BLOCKS:
+		return do_vfs_ioctl(file, fd, cmd, arg);
+	}
+
+	return -ENOIOCTLCMD;
 }
 
 static void compat_ioctl_error(struct file *filp, unsigned int fd,
@@ -2025,12 +1646,33 @@ static void compat_ioctl_error(struct file *filp, unsigned int fd,
 		free_page((unsigned long)path);
 }
 
+static int compat_ioctl_check_table(unsigned int xcmd)
+{
+	int i;
+	const int max = ARRAY_SIZE(ioctl_pointer) - 1;
+
+	BUILD_BUG_ON(max >= (1 << 16));
+
+	/* guess initial offset into table, assuming a
+	   normalized distribution */
+	i = ((xcmd >> 16) * max) >> 16;
+
+	/* do linear search up first, until greater or equal */
+	while (ioctl_pointer[i] < xcmd && i < max)
+		i++;
+
+	/* then do linear search down */
+	while (ioctl_pointer[i] > xcmd && i > 0)
+		i--;
+
+	return ioctl_pointer[i] == xcmd;
+}
+
 asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd,
 				unsigned long arg)
 {
 	struct file *filp;
 	int error = -EBADF;
-	struct ioctl_trans *t;
 	int fput_needed;
 
 	filp = fget_light(fd, &fput_needed);
@@ -2058,7 +1700,7 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd,
 #if defined(CONFIG_IA64) || defined(CONFIG_X86_64)
 	case FS_IOC_RESVSP_32:
 	case FS_IOC_RESVSP64_32:
-		error = compat_ioctl_preallocate(filp, arg);
+		error = compat_ioctl_preallocate(filp, compat_ptr(arg));
 		goto out_fput;
 #else
 	case FS_IOC_RESVSP:
@@ -2087,12 +1729,11 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd,
 		break;
 	}
 
-	for (t = ioctl32_hash_table[ioctl32_hash(cmd)]; t; t = t->next) {
-		if (t->cmd == cmd)
-			goto found_handler;
-	}
+	if (compat_ioctl_check_table(XFORM(cmd)))
+		goto found_handler;
 
-	{
+	error = do_ioctl_trans(fd, cmd, arg, filp);
+	if (error == -ENOIOCTLCMD) {
 		static int count;
 
 		if (++count <= 50)
@@ -2103,13 +1744,7 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd,
 	goto out_fput;
 
  found_handler:
-	if (t->handler) {
-		lock_kernel();
-		error = t->handler(fd, cmd, arg, filp);
-		unlock_kernel();
-		goto out_fput;
-	}
-
+	arg = (unsigned long)compat_ptr(arg);
  do_ioctl:
 	error = do_vfs_ioctl(filp, fd, cmd, arg);
  out_fput:
@@ -2118,35 +1753,22 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd,
 	return error;
 }
 
-static void ioctl32_insert_translation(struct ioctl_trans *trans)
+static int __init init_sys32_ioctl_cmp(const void *p, const void *q)
 {
-	unsigned long hash;
-	struct ioctl_trans *t;
-
-	hash = ioctl32_hash (trans->cmd);
-	if (!ioctl32_hash_table[hash])
-		ioctl32_hash_table[hash] = trans;
-	else {
-		t = ioctl32_hash_table[hash];
-		while (t->next)
-			t = t->next;
-		trans->next = NULL;
-		t->next = trans;
-	}
+	unsigned int a, b;
+	a = *(unsigned int *)p;
+	b = *(unsigned int *)q;
+	if (a > b)
+		return 1;
+	if (a < b)
+		return -1;
+	return 0;
 }
 
 static int __init init_sys32_ioctl(void)
 {
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(ioctl_start); i++) {
-		if (ioctl_start[i].next) {
-			printk("ioctl translation %d bad\n",i);
-			return -1;
-		}
-
-		ioctl32_insert_translation(&ioctl_start[i]);
-	}
+	sort(ioctl_pointer, ARRAY_SIZE(ioctl_pointer), sizeof(*ioctl_pointer),
+		init_sys32_ioctl_cmp, NULL);
 	return 0;
 }
 __initcall(init_sys32_ioctl);
diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c
index c8afa6b1d91d..32a5f46b1157 100644
--- a/fs/configfs/symlink.c
+++ b/fs/configfs/symlink.c
@@ -121,8 +121,10 @@ static int get_target(const char *symname, struct path *path,
 				ret = -ENOENT;
 				path_put(path);
 			}
-		} else
+		} else {
 			ret = -EPERM;
+			path_put(path);
+		}
 	}
 
 	return ret;
diff --git a/fs/dcache.c b/fs/dcache.c
index a100fa35a48f..953173a293a9 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -978,6 +978,7 @@ struct dentry *d_alloc_name(struct dentry *parent, const char *name)
 	q.hash = full_name_hash(q.name, q.len);
 	return d_alloc(parent, &q);
 }
+EXPORT_SYMBOL(d_alloc_name);
 
 /* the caller must hold dcache_lock */
 static void __d_instantiate(struct dentry *dentry, struct inode *inode)
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 0d23b52dd22c..274ac865bae8 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -32,7 +32,9 @@ static struct vfsmount *debugfs_mount;
 static int debugfs_mount_count;
 static bool debugfs_registered;
 
-static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t dev)
+static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t dev,
+				       void *data, const struct file_operations *fops)
+
 {
 	struct inode *inode = new_inode(sb);
 
@@ -44,14 +46,18 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d
 			init_special_inode(inode, mode, dev);
 			break;
 		case S_IFREG:
-			inode->i_fop = &debugfs_file_operations;
+			inode->i_fop = fops ? fops : &debugfs_file_operations;
+			inode->i_private = data;
 			break;
 		case S_IFLNK:
 			inode->i_op = &debugfs_link_operations;
+			inode->i_fop = fops;
+			inode->i_private = data;
 			break;
 		case S_IFDIR:
 			inode->i_op = &simple_dir_inode_operations;
-			inode->i_fop = &simple_dir_operations;
+			inode->i_fop = fops ? fops : &simple_dir_operations;
+			inode->i_private = data;
 
 			/* directory inodes start off with i_nlink == 2
 			 * (for "." entry) */
@@ -64,7 +70,8 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d
 
 /* SMP-safe */
 static int debugfs_mknod(struct inode *dir, struct dentry *dentry,
-			 int mode, dev_t dev)
+			 int mode, dev_t dev, void *data,
+			 const struct file_operations *fops)
 {
 	struct inode *inode;
 	int error = -EPERM;
@@ -72,7 +79,7 @@ static int debugfs_mknod(struct inode *dir, struct dentry *dentry,
 	if (dentry->d_inode)
 		return -EEXIST;
 
-	inode = debugfs_get_inode(dir->i_sb, mode, dev);
+	inode = debugfs_get_inode(dir->i_sb, mode, dev, data, fops);
 	if (inode) {
 		d_instantiate(dentry, inode);
 		dget(dentry);
@@ -81,12 +88,13 @@ static int debugfs_mknod(struct inode *dir, struct dentry *dentry,
 	return error;
 }
 
-static int debugfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+static int debugfs_mkdir(struct inode *dir, struct dentry *dentry, int mode,
+			 void *data, const struct file_operations *fops)
 {
 	int res;
 
 	mode = (mode & (S_IRWXUGO | S_ISVTX)) | S_IFDIR;
-	res = debugfs_mknod(dir, dentry, mode, 0);
+	res = debugfs_mknod(dir, dentry, mode, 0, data, fops);
 	if (!res) {
 		inc_nlink(dir);
 		fsnotify_mkdir(dir, dentry);
@@ -94,18 +102,20 @@ static int debugfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	return res;
 }
 
-static int debugfs_link(struct inode *dir, struct dentry *dentry, int mode)
+static int debugfs_link(struct inode *dir, struct dentry *dentry, int mode,
+			void *data, const struct file_operations *fops)
 {
 	mode = (mode & S_IALLUGO) | S_IFLNK;
-	return debugfs_mknod(dir, dentry, mode, 0);
+	return debugfs_mknod(dir, dentry, mode, 0, data, fops);
 }
 
-static int debugfs_create(struct inode *dir, struct dentry *dentry, int mode)
+static int debugfs_create(struct inode *dir, struct dentry *dentry, int mode,
+			  void *data, const struct file_operations *fops)
 {
 	int res;
 
 	mode = (mode & S_IALLUGO) | S_IFREG;
-	res = debugfs_mknod(dir, dentry, mode, 0);
+	res = debugfs_mknod(dir, dentry, mode, 0, data, fops);
 	if (!res)
 		fsnotify_create(dir, dentry);
 	return res;
@@ -139,7 +149,9 @@ static struct file_system_type debug_fs_type = {
 
 static int debugfs_create_by_name(const char *name, mode_t mode,
 				  struct dentry *parent,
-				  struct dentry **dentry)
+				  struct dentry **dentry,
+				  void *data,
+				  const struct file_operations *fops)
 {
 	int error = 0;
 
@@ -148,15 +160,8 @@ static int debugfs_create_by_name(const char *name, mode_t mode,
 	 * block. A pointer to that is in the struct vfsmount that we
 	 * have around.
 	 */
-	if (!parent) {
-		if (debugfs_mount && debugfs_mount->mnt_sb) {
-			parent = debugfs_mount->mnt_sb->s_root;
-		}
-	}
-	if (!parent) {
-		pr_debug("debugfs: Ah! can not find a parent!\n");
-		return -EFAULT;
-	}
+	if (!parent)
+		parent = debugfs_mount->mnt_sb->s_root;
 
 	*dentry = NULL;
 	mutex_lock(&parent->d_inode->i_mutex);
@@ -164,13 +169,16 @@ static int debugfs_create_by_name(const char *name, mode_t mode,
 	if (!IS_ERR(*dentry)) {
 		switch (mode & S_IFMT) {
 		case S_IFDIR:
-			error = debugfs_mkdir(parent->d_inode, *dentry, mode);
+			error = debugfs_mkdir(parent->d_inode, *dentry, mode,
+					      data, fops);
 			break;
 		case S_IFLNK:
-			error = debugfs_link(parent->d_inode, *dentry, mode);
+			error = debugfs_link(parent->d_inode, *dentry, mode,
+					     data, fops);
 			break;
 		default:
-			error = debugfs_create(parent->d_inode, *dentry, mode);
+			error = debugfs_create(parent->d_inode, *dentry, mode,
+					       data, fops);
 			break;
 		}
 		dput(*dentry);
@@ -221,19 +229,13 @@ struct dentry *debugfs_create_file(const char *name, mode_t mode,
 	if (error)
 		goto exit;
 
-	error = debugfs_create_by_name(name, mode, parent, &dentry);
+	error = debugfs_create_by_name(name, mode, parent, &dentry,
+				       data, fops);
 	if (error) {
 		dentry = NULL;
 		simple_release_fs(&debugfs_mount, &debugfs_mount_count);
 		goto exit;
 	}
-
-	if (dentry->d_inode) {
-		if (data)
-			dentry->d_inode->i_private = data;
-		if (fops)
-			dentry->d_inode->i_fop = fops;
-	}
 exit:
 	return dentry;
 }
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index d5f8c96964be..8882ecc0f1bf 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -517,11 +517,23 @@ int devpts_pty_new(struct inode *ptmx_inode, struct tty_struct *tty)
 
 struct tty_struct *devpts_get_tty(struct inode *pts_inode, int number)
 {
+	struct dentry *dentry;
+	struct tty_struct *tty;
+
 	BUG_ON(pts_inode->i_rdev == MKDEV(TTYAUX_MAJOR, PTMX_MINOR));
 
+	/* Ensure dentry has not been deleted by devpts_pty_kill() */
+	dentry = d_find_alias(pts_inode);
+	if (!dentry)
+		return NULL;
+
+	tty = NULL;
 	if (pts_inode->i_sb->s_magic == DEVPTS_SUPER_MAGIC)
-		return (struct tty_struct *)pts_inode->i_private;
-	return NULL;
+		tty = (struct tty_struct *)pts_inode->i_private;
+
+	dput(dentry);
+
+	return tty;
 }
 
 void devpts_pty_kill(struct tty_struct *tty)
diff --git a/fs/direct-io.c b/fs/direct-io.c
index b912270942fa..e82adc2debb7 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -53,13 +53,6 @@
  *
  * If blkfactor is zero then the user's request was aligned to the filesystem's
  * blocksize.
- *
- * lock_type is DIO_LOCKING for regular files on direct-IO-naive filesystems.
- * This determines whether we need to do the fancy locking which prevents
- * direct-IO from being able to read uninitialised disk blocks.  If its zero
- * (blockdev) this locking is not done, and if it is DIO_OWN_LOCKING i_mutex is
- * not held for the entire direct write (taken briefly, initially, during a
- * direct read though, but its never held for the duration of a direct-IO).
  */
 
 struct dio {
@@ -68,7 +61,7 @@ struct dio {
 	struct inode *inode;
 	int rw;
 	loff_t i_size;			/* i_size when submitted */
-	int lock_type;			/* doesn't change */
+	int flags;			/* doesn't change */
 	unsigned blkbits;		/* doesn't change */
 	unsigned blkfactor;		/* When we're using an alignment which
 					   is finer than the filesystem's soft
@@ -104,6 +97,18 @@ struct dio {
 	unsigned cur_page_len;		/* Nr of bytes at cur_page_offset */
 	sector_t cur_page_block;	/* Where it starts */
 
+	/* BIO completion state */
+	spinlock_t bio_lock;		/* protects BIO fields below */
+	unsigned long refcount;		/* direct_io_worker() and bios */
+	struct bio *bio_list;		/* singly linked via bi_private */
+	struct task_struct *waiter;	/* waiting task (NULL if none) */
+
+	/* AIO related stuff */
+	struct kiocb *iocb;		/* kiocb */
+	int is_async;			/* is IO async ? */
+	int io_error;			/* IO error in completion path */
+	ssize_t result;                 /* IO result */
+
 	/*
 	 * Page fetching state. These variables belong to dio_refill_pages().
 	 */
@@ -115,22 +120,16 @@ struct dio {
 	 * Page queue.  These variables belong to dio_refill_pages() and
 	 * dio_get_page().
 	 */
-	struct page *pages[DIO_PAGES];	/* page buffer */
 	unsigned head;			/* next page to process */
 	unsigned tail;			/* last valid page + 1 */
 	int page_errors;		/* errno from get_user_pages() */
 
-	/* BIO completion state */
-	spinlock_t bio_lock;		/* protects BIO fields below */
-	unsigned long refcount;		/* direct_io_worker() and bios */
-	struct bio *bio_list;		/* singly linked via bi_private */
-	struct task_struct *waiter;	/* waiting task (NULL if none) */
-
-	/* AIO related stuff */
-	struct kiocb *iocb;		/* kiocb */
-	int is_async;			/* is IO async ? */
-	int io_error;			/* IO error in completion path */
-	ssize_t result;                 /* IO result */
+	/*
+	 * pages[] (and any fields placed after it) are not zeroed out at
+	 * allocation time.  Don't add new fields after pages[] unless you
+	 * wish that they not be zeroed.
+	 */
+	struct page *pages[DIO_PAGES];	/* page buffer */
 };
 
 /*
@@ -240,7 +239,8 @@ static int dio_complete(struct dio *dio, loff_t offset, int ret)
 	if (dio->end_io && dio->result)
 		dio->end_io(dio->iocb, offset, transferred,
 			    dio->map_bh.b_private);
-	if (dio->lock_type == DIO_LOCKING)
+
+	if (dio->flags & DIO_LOCKING)
 		/* lockdep: non-owner release */
 		up_read_non_owner(&dio->inode->i_alloc_sem);
 
@@ -515,21 +515,24 @@ static int get_more_blocks(struct dio *dio)
 		map_bh->b_state = 0;
 		map_bh->b_size = fs_count << dio->inode->i_blkbits;
 
+		/*
+		 * For writes inside i_size on a DIO_SKIP_HOLES filesystem we
+		 * forbid block creations: only overwrites are permitted.
+		 * We will return early to the caller once we see an
+		 * unmapped buffer head returned, and the caller will fall
+		 * back to buffered I/O.
+		 *
+		 * Otherwise the decision is left to the get_blocks method,
+		 * which may decide to handle it or also return an unmapped
+		 * buffer head.
+		 */
 		create = dio->rw & WRITE;
-		if (dio->lock_type == DIO_LOCKING) {
+		if (dio->flags & DIO_SKIP_HOLES) {
 			if (dio->block_in_file < (i_size_read(dio->inode) >>
 							dio->blkbits))
 				create = 0;
-		} else if (dio->lock_type == DIO_NO_LOCKING) {
-			create = 0;
 		}
 
-		/*
-		 * For writes inside i_size we forbid block creations: only
-		 * overwrites are permitted.  We fall back to buffered writes
-		 * at a higher level for inside-i_size block-instantiating
-		 * writes.
-		 */
 		ret = (*dio->get_block)(dio->inode, fs_startblk,
 						map_bh, create);
 	}
@@ -1039,7 +1042,7 @@ direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode,
 	 * we can let i_mutex go now that its achieved its purpose
 	 * of protecting us from looking up uninitialized blocks.
 	 */
-	if ((rw == READ) && (dio->lock_type == DIO_LOCKING))
+	if (rw == READ && (dio->flags & DIO_LOCKING))
 		mutex_unlock(&dio->inode->i_mutex);
 
 	/*
@@ -1086,30 +1089,28 @@ direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode,
 
 /*
  * This is a library function for use by filesystem drivers.
- * The locking rules are governed by the dio_lock_type parameter.
  *
- * DIO_NO_LOCKING (no locking, for raw block device access)
- * For writes, i_mutex is not held on entry; it is never taken.
+ * The locking rules are governed by the flags parameter:
+ *  - if the flags value contains DIO_LOCKING we use a fancy locking
+ *    scheme for dumb filesystems.
+ *    For writes this function is called under i_mutex and returns with
+ *    i_mutex held, for reads, i_mutex is not held on entry, but it is
+ *    taken and dropped again before returning.
+ *    For reads and writes i_alloc_sem is taken in shared mode and released
+ *    on I/O completion (which may happen asynchronously after returning to
+ *    the caller).
  *
- * DIO_LOCKING (simple locking for regular files)
- * For writes we are called under i_mutex and return with i_mutex held, even
- * though it is internally dropped.
- * For reads, i_mutex is not held on entry, but it is taken and dropped before
- * returning.
- *
- * DIO_OWN_LOCKING (filesystem provides synchronisation and handling of
- *	uninitialised data, allowing parallel direct readers and writers)
- * For writes we are called without i_mutex, return without it, never touch it.
- * For reads we are called under i_mutex and return with i_mutex held, even
- * though it may be internally dropped.
- *
- * Additional i_alloc_sem locking requirements described inline below.
+ *  - if the flags value does NOT contain DIO_LOCKING we don't use any
+ *    internal locking but rather rely on the filesystem to synchronize
+ *    direct I/O reads/writes versus each other and truncate.
+ *    For reads and writes both i_mutex and i_alloc_sem are not held on
+ *    entry and are never taken.
  */
 ssize_t
 __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 	struct block_device *bdev, const struct iovec *iov, loff_t offset, 
 	unsigned long nr_segs, get_block_t get_block, dio_iodone_t end_io,
-	int dio_lock_type)
+	int flags)
 {
 	int seg;
 	size_t size;
@@ -1120,8 +1121,6 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 	ssize_t retval = -EINVAL;
 	loff_t end = offset;
 	struct dio *dio;
-	int release_i_mutex = 0;
-	int acquire_i_mutex = 0;
 
 	if (rw & WRITE)
 		rw = WRITE_ODIRECT_PLUG;
@@ -1151,48 +1150,41 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 		}
 	}
 
-	dio = kzalloc(sizeof(*dio), GFP_KERNEL);
+	dio = kmalloc(sizeof(*dio), GFP_KERNEL);
 	retval = -ENOMEM;
 	if (!dio)
 		goto out;
-
 	/*
-	 * For block device access DIO_NO_LOCKING is used,
-	 *	neither readers nor writers do any locking at all
-	 * For regular files using DIO_LOCKING,
-	 *	readers need to grab i_mutex and i_alloc_sem
-	 *	writers need to grab i_alloc_sem only (i_mutex is already held)
-	 * For regular files using DIO_OWN_LOCKING,
-	 *	neither readers nor writers take any locks here
+	 * Believe it or not, zeroing out the page array caused a .5%
+	 * performance regression in a database benchmark.  So, we take
+	 * care to only zero out what's needed.
 	 */
-	dio->lock_type = dio_lock_type;
-	if (dio_lock_type != DIO_NO_LOCKING) {
+	memset(dio, 0, offsetof(struct dio, pages));
+
+	dio->flags = flags;
+	if (dio->flags & DIO_LOCKING) {
 		/* watch out for a 0 len io from a tricksy fs */
 		if (rw == READ && end > offset) {
-			struct address_space *mapping;
+			struct address_space *mapping =
+					iocb->ki_filp->f_mapping;
 
-			mapping = iocb->ki_filp->f_mapping;
-			if (dio_lock_type != DIO_OWN_LOCKING) {
-				mutex_lock(&inode->i_mutex);
-				release_i_mutex = 1;
-			}
+			/* will be released by direct_io_worker */
+			mutex_lock(&inode->i_mutex);
 
 			retval = filemap_write_and_wait_range(mapping, offset,
 							      end - 1);
 			if (retval) {
+				mutex_unlock(&inode->i_mutex);
 				kfree(dio);
 				goto out;
 			}
-
-			if (dio_lock_type == DIO_OWN_LOCKING) {
-				mutex_unlock(&inode->i_mutex);
-				acquire_i_mutex = 1;
-			}
 		}
 
-		if (dio_lock_type == DIO_LOCKING)
-			/* lockdep: not the owner will release it */
-			down_read_non_owner(&inode->i_alloc_sem);
+		/*
+		 * Will be released at I/O completion, possibly in a
+		 * different thread.
+		 */
+		down_read_non_owner(&inode->i_alloc_sem);
 	}
 
 	/*
@@ -1210,24 +1202,19 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode,
 	/*
 	 * In case of error extending write may have instantiated a few
 	 * blocks outside i_size. Trim these off again for DIO_LOCKING.
-	 * NOTE: DIO_NO_LOCK/DIO_OWN_LOCK callers have to handle this by
-	 * it's own meaner.
+	 *
+	 * NOTE: filesystems with their own locking have to handle this
+	 * on their own.
 	 */
-	if (unlikely(retval < 0 && (rw & WRITE))) {
-		loff_t isize = i_size_read(inode);
-
-		if (end > isize && dio_lock_type == DIO_LOCKING)
-			vmtruncate(inode, isize);
+	if (flags & DIO_LOCKING) {
+		if (unlikely((rw & WRITE) && retval < 0)) {
+			loff_t isize = i_size_read(inode);
+			if (end > isize)
+				vmtruncate(inode, isize);
+		}
 	}
 
-	if (rw == READ && dio_lock_type == DIO_LOCKING)
-		release_i_mutex = 0;
-
 out:
-	if (release_i_mutex)
-		mutex_unlock(&inode->i_mutex);
-	else if (acquire_i_mutex)
-		mutex_lock(&inode->i_mutex);
 	return retval;
 }
 EXPORT_SYMBOL(__blockdev_direct_IO);
diff --git a/fs/dlm/config.c b/fs/dlm/config.c
index fd9859f92fad..0df243850818 100644
--- a/fs/dlm/config.c
+++ b/fs/dlm/config.c
@@ -410,10 +410,10 @@ static struct config_group *make_cluster(struct config_group *g,
 	struct dlm_comms *cms = NULL;
 	void *gps = NULL;
 
-	cl = kzalloc(sizeof(struct dlm_cluster), GFP_KERNEL);
-	gps = kcalloc(3, sizeof(struct config_group *), GFP_KERNEL);
-	sps = kzalloc(sizeof(struct dlm_spaces), GFP_KERNEL);
-	cms = kzalloc(sizeof(struct dlm_comms), GFP_KERNEL);
+	cl = kzalloc(sizeof(struct dlm_cluster), GFP_NOFS);
+	gps = kcalloc(3, sizeof(struct config_group *), GFP_NOFS);
+	sps = kzalloc(sizeof(struct dlm_spaces), GFP_NOFS);
+	cms = kzalloc(sizeof(struct dlm_comms), GFP_NOFS);
 
 	if (!cl || !gps || !sps || !cms)
 		goto fail;
@@ -482,9 +482,9 @@ static struct config_group *make_space(struct config_group *g, const char *name)
 	struct dlm_nodes *nds = NULL;
 	void *gps = NULL;
 
-	sp = kzalloc(sizeof(struct dlm_space), GFP_KERNEL);
-	gps = kcalloc(2, sizeof(struct config_group *), GFP_KERNEL);
-	nds = kzalloc(sizeof(struct dlm_nodes), GFP_KERNEL);
+	sp = kzalloc(sizeof(struct dlm_space), GFP_NOFS);
+	gps = kcalloc(2, sizeof(struct config_group *), GFP_NOFS);
+	nds = kzalloc(sizeof(struct dlm_nodes), GFP_NOFS);
 
 	if (!sp || !gps || !nds)
 		goto fail;
@@ -536,7 +536,7 @@ static struct config_item *make_comm(struct config_group *g, const char *name)
 {
 	struct dlm_comm *cm;
 
-	cm = kzalloc(sizeof(struct dlm_comm), GFP_KERNEL);
+	cm = kzalloc(sizeof(struct dlm_comm), GFP_NOFS);
 	if (!cm)
 		return ERR_PTR(-ENOMEM);
 
@@ -569,7 +569,7 @@ static struct config_item *make_node(struct config_group *g, const char *name)
 	struct dlm_space *sp = config_item_to_space(g->cg_item.ci_parent);
 	struct dlm_node *nd;
 
-	nd = kzalloc(sizeof(struct dlm_node), GFP_KERNEL);
+	nd = kzalloc(sizeof(struct dlm_node), GFP_NOFS);
 	if (!nd)
 		return ERR_PTR(-ENOMEM);
 
@@ -705,7 +705,7 @@ static ssize_t comm_addr_write(struct dlm_comm *cm, const char *buf, size_t len)
 	if (cm->addr_count >= DLM_MAX_ADDR_COUNT)
 		return -ENOSPC;
 
-	addr = kzalloc(sizeof(*addr), GFP_KERNEL);
+	addr = kzalloc(sizeof(*addr), GFP_NOFS);
 	if (!addr)
 		return -ENOMEM;
 
@@ -868,7 +868,7 @@ int dlm_nodeid_list(char *lsname, int **ids_out, int *ids_count_out,
 
 	ids_count = sp->members_count;
 
-	ids = kcalloc(ids_count, sizeof(int), GFP_KERNEL);
+	ids = kcalloc(ids_count, sizeof(int), GFP_NOFS);
 	if (!ids) {
 		rv = -ENOMEM;
 		goto out;
@@ -886,7 +886,7 @@ int dlm_nodeid_list(char *lsname, int **ids_out, int *ids_count_out,
 	if (!new_count)
 		goto out_ids;
 
-	new = kcalloc(new_count, sizeof(int), GFP_KERNEL);
+	new = kcalloc(new_count, sizeof(int), GFP_NOFS);
 	if (!new) {
 		kfree(ids);
 		rv = -ENOMEM;
diff --git a/fs/dlm/debug_fs.c b/fs/dlm/debug_fs.c
index 1c8bb8c3a82e..375a2359b3bf 100644
--- a/fs/dlm/debug_fs.c
+++ b/fs/dlm/debug_fs.c
@@ -404,7 +404,7 @@ static void *table_seq_start(struct seq_file *seq, loff_t *pos)
 	if (bucket >= ls->ls_rsbtbl_size)
 		return NULL;
 
-	ri = kzalloc(sizeof(struct rsbtbl_iter), GFP_KERNEL);
+	ri = kzalloc(sizeof(struct rsbtbl_iter), GFP_NOFS);
 	if (!ri)
 		return NULL;
 	if (n == 0)
diff --git a/fs/dlm/dir.c b/fs/dlm/dir.c
index c4dfa1dcc86f..7b84c1dbc82e 100644
--- a/fs/dlm/dir.c
+++ b/fs/dlm/dir.c
@@ -49,8 +49,7 @@ static struct dlm_direntry *get_free_de(struct dlm_ls *ls, int len)
 	spin_unlock(&ls->ls_recover_list_lock);
 
 	if (!found)
-		de = kzalloc(sizeof(struct dlm_direntry) + len,
-			     ls->ls_allocation);
+		de = kzalloc(sizeof(struct dlm_direntry) + len, GFP_NOFS);
 	return de;
 }
 
@@ -212,7 +211,7 @@ int dlm_recover_directory(struct dlm_ls *ls)
 
 	dlm_dir_clear(ls);
 
-	last_name = kmalloc(DLM_RESNAME_MAXLEN, ls->ls_allocation);
+	last_name = kmalloc(DLM_RESNAME_MAXLEN, GFP_NOFS);
 	if (!last_name)
 		goto out;
 
@@ -323,7 +322,7 @@ static int get_entry(struct dlm_ls *ls, int nodeid, char *name,
 	if (namelen > DLM_RESNAME_MAXLEN)
 		return -EINVAL;
 
-	de = kzalloc(sizeof(struct dlm_direntry) + namelen, ls->ls_allocation);
+	de = kzalloc(sizeof(struct dlm_direntry) + namelen, GFP_NOFS);
 	if (!de)
 		return -ENOMEM;
 
diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h
index d01ca0a711db..826d3dc6e0ab 100644
--- a/fs/dlm/dlm_internal.h
+++ b/fs/dlm/dlm_internal.h
@@ -473,7 +473,6 @@ struct dlm_ls {
 	int			ls_low_nodeid;
 	int			ls_total_weight;
 	int			*ls_node_array;
-	gfp_t			ls_allocation;
 
 	struct dlm_rsb		ls_stub_rsb;	/* for returning errors */
 	struct dlm_lkb		ls_stub_lkb;	/* for returning errors */
diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c
index eb507c453c5f..9c0c1db1e105 100644
--- a/fs/dlm/lock.c
+++ b/fs/dlm/lock.c
@@ -2689,7 +2689,7 @@ static int _create_message(struct dlm_ls *ls, int mb_len,
 	   pass into lowcomms_commit and a message buffer (mb) that we
 	   write our data into */
 
-	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, ls->ls_allocation, &mb);
+	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, GFP_NOFS, &mb);
 	if (!mh)
 		return -ENOBUFS;
 
@@ -4512,7 +4512,7 @@ int dlm_user_request(struct dlm_ls *ls, struct dlm_user_args *ua,
 	}
 
 	if (flags & DLM_LKF_VALBLK) {
-		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_KERNEL);
+		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_NOFS);
 		if (!ua->lksb.sb_lvbptr) {
 			kfree(ua);
 			__put_lkb(ls, lkb);
@@ -4582,7 +4582,7 @@ int dlm_user_convert(struct dlm_ls *ls, struct dlm_user_args *ua_tmp,
 	ua = lkb->lkb_ua;
 
 	if (flags & DLM_LKF_VALBLK && !ua->lksb.sb_lvbptr) {
-		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_KERNEL);
+		ua->lksb.sb_lvbptr = kzalloc(DLM_USER_LVB_LEN, GFP_NOFS);
 		if (!ua->lksb.sb_lvbptr) {
 			error = -ENOMEM;
 			goto out_put;
diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index d489fcc86713..c010ecfc0d29 100644
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -430,7 +430,7 @@ static int new_lockspace(const char *name, int namelen, void **lockspace,
 
 	error = -ENOMEM;
 
-	ls = kzalloc(sizeof(struct dlm_ls) + namelen, GFP_KERNEL);
+	ls = kzalloc(sizeof(struct dlm_ls) + namelen, GFP_NOFS);
 	if (!ls)
 		goto out;
 	memcpy(ls->ls_name, name, namelen);
@@ -443,11 +443,6 @@ static int new_lockspace(const char *name, int namelen, void **lockspace,
 	if (flags & DLM_LSFL_TIMEWARN)
 		set_bit(LSFL_TIMEWARN, &ls->ls_flags);
 
-	if (flags & DLM_LSFL_FS)
-		ls->ls_allocation = GFP_NOFS;
-	else
-		ls->ls_allocation = GFP_KERNEL;
-
 	/* ls_exflags are forced to match among nodes, and we don't
 	   need to require all nodes to have some flags set */
 	ls->ls_exflags = (flags & ~(DLM_LSFL_TIMEWARN | DLM_LSFL_FS |
@@ -456,7 +451,7 @@ static int new_lockspace(const char *name, int namelen, void **lockspace,
 	size = dlm_config.ci_rsbtbl_size;
 	ls->ls_rsbtbl_size = size;
 
-	ls->ls_rsbtbl = kmalloc(sizeof(struct dlm_rsbtable) * size, GFP_KERNEL);
+	ls->ls_rsbtbl = kmalloc(sizeof(struct dlm_rsbtable) * size, GFP_NOFS);
 	if (!ls->ls_rsbtbl)
 		goto out_lsfree;
 	for (i = 0; i < size; i++) {
@@ -468,7 +463,7 @@ static int new_lockspace(const char *name, int namelen, void **lockspace,
 	size = dlm_config.ci_lkbtbl_size;
 	ls->ls_lkbtbl_size = size;
 
-	ls->ls_lkbtbl = kmalloc(sizeof(struct dlm_lkbtable) * size, GFP_KERNEL);
+	ls->ls_lkbtbl = kmalloc(sizeof(struct dlm_lkbtable) * size, GFP_NOFS);
 	if (!ls->ls_lkbtbl)
 		goto out_rsbfree;
 	for (i = 0; i < size; i++) {
@@ -480,7 +475,7 @@ static int new_lockspace(const char *name, int namelen, void **lockspace,
 	size = dlm_config.ci_dirtbl_size;
 	ls->ls_dirtbl_size = size;
 
-	ls->ls_dirtbl = kmalloc(sizeof(struct dlm_dirtable) * size, GFP_KERNEL);
+	ls->ls_dirtbl = kmalloc(sizeof(struct dlm_dirtable) * size, GFP_NOFS);
 	if (!ls->ls_dirtbl)
 		goto out_lkbfree;
 	for (i = 0; i < size; i++) {
@@ -527,7 +522,7 @@ static int new_lockspace(const char *name, int namelen, void **lockspace,
 	mutex_init(&ls->ls_requestqueue_mutex);
 	mutex_init(&ls->ls_clear_proc_locks);
 
-	ls->ls_recover_buf = kmalloc(dlm_config.ci_buffer_size, GFP_KERNEL);
+	ls->ls_recover_buf = kmalloc(dlm_config.ci_buffer_size, GFP_NOFS);
 	if (!ls->ls_recover_buf)
 		goto out_dirfree;
 
diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c
index 70736eb4b516..52cab160893c 100644
--- a/fs/dlm/lowcomms.c
+++ b/fs/dlm/lowcomms.c
@@ -1060,7 +1060,7 @@ static void init_local(void)
 		if (dlm_our_addr(&sas, i))
 			break;
 
-		addr = kmalloc(sizeof(*addr), GFP_KERNEL);
+		addr = kmalloc(sizeof(*addr), GFP_NOFS);
 		if (!addr)
 			break;
 		memcpy(addr, &sas, sizeof(*addr));
@@ -1099,7 +1099,7 @@ static int sctp_listen_for_all(void)
 	struct sockaddr_storage localaddr;
 	struct sctp_event_subscribe subscribe;
 	int result = -EINVAL, num = 1, i, addr_len;
-	struct connection *con = nodeid2con(0, GFP_KERNEL);
+	struct connection *con = nodeid2con(0, GFP_NOFS);
 	int bufsize = NEEDED_RMEM;
 
 	if (!con)
@@ -1171,7 +1171,7 @@ out:
 static int tcp_listen_for_all(void)
 {
 	struct socket *sock = NULL;
-	struct connection *con = nodeid2con(0, GFP_KERNEL);
+	struct connection *con = nodeid2con(0, GFP_NOFS);
 	int result = -EINVAL;
 
 	if (!con)
diff --git a/fs/dlm/member.c b/fs/dlm/member.c
index b128775913b2..84f70bfb0baf 100644
--- a/fs/dlm/member.c
+++ b/fs/dlm/member.c
@@ -48,7 +48,7 @@ static int dlm_add_member(struct dlm_ls *ls, int nodeid)
 	struct dlm_member *memb;
 	int w, error;
 
-	memb = kzalloc(sizeof(struct dlm_member), ls->ls_allocation);
+	memb = kzalloc(sizeof(struct dlm_member), GFP_NOFS);
 	if (!memb)
 		return -ENOMEM;
 
@@ -143,7 +143,7 @@ static void make_member_array(struct dlm_ls *ls)
 
 	ls->ls_total_weight = total;
 
-	array = kmalloc(sizeof(int) * total, ls->ls_allocation);
+	array = kmalloc(sizeof(int) * total, GFP_NOFS);
 	if (!array)
 		return;
 
@@ -226,7 +226,7 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out)
 			continue;
 		log_debug(ls, "new nodeid %d is a re-added member", rv->new[i]);
 
-		memb = kzalloc(sizeof(struct dlm_member), ls->ls_allocation);
+		memb = kzalloc(sizeof(struct dlm_member), GFP_NOFS);
 		if (!memb)
 			return -ENOMEM;
 		memb->nodeid = rv->new[i];
@@ -341,7 +341,7 @@ int dlm_ls_start(struct dlm_ls *ls)
 	int *ids = NULL, *new = NULL;
 	int error, ids_count = 0, new_count = 0;
 
-	rv = kzalloc(sizeof(struct dlm_recover), ls->ls_allocation);
+	rv = kzalloc(sizeof(struct dlm_recover), GFP_NOFS);
 	if (!rv)
 		return -ENOMEM;
 
diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c
index c1775b84ebab..8e0d00db004f 100644
--- a/fs/dlm/memory.c
+++ b/fs/dlm/memory.c
@@ -39,7 +39,7 @@ char *dlm_allocate_lvb(struct dlm_ls *ls)
 {
 	char *p;
 
-	p = kzalloc(ls->ls_lvblen, ls->ls_allocation);
+	p = kzalloc(ls->ls_lvblen, GFP_NOFS);
 	return p;
 }
 
@@ -57,7 +57,7 @@ struct dlm_rsb *dlm_allocate_rsb(struct dlm_ls *ls, int namelen)
 
 	DLM_ASSERT(namelen <= DLM_RESNAME_MAXLEN,);
 
-	r = kzalloc(sizeof(*r) + namelen, ls->ls_allocation);
+	r = kzalloc(sizeof(*r) + namelen, GFP_NOFS);
 	return r;
 }
 
@@ -72,7 +72,7 @@ struct dlm_lkb *dlm_allocate_lkb(struct dlm_ls *ls)
 {
 	struct dlm_lkb *lkb;
 
-	lkb = kmem_cache_zalloc(lkb_cache, ls->ls_allocation);
+	lkb = kmem_cache_zalloc(lkb_cache, GFP_NOFS);
 	return lkb;
 }
 
diff --git a/fs/dlm/netlink.c b/fs/dlm/netlink.c
index 55ea369f43a9..052095cd592f 100644
--- a/fs/dlm/netlink.c
+++ b/fs/dlm/netlink.c
@@ -26,7 +26,7 @@ static int prepare_data(u8 cmd, struct sk_buff **skbp, size_t size)
 	struct sk_buff *skb;
 	void *data;
 
-	skb = genlmsg_new(size, GFP_KERNEL);
+	skb = genlmsg_new(size, GFP_NOFS);
 	if (!skb)
 		return -ENOMEM;
 
diff --git a/fs/dlm/plock.c b/fs/dlm/plock.c
index b540aa5d1f61..b5f89aef3b29 100644
--- a/fs/dlm/plock.c
+++ b/fs/dlm/plock.c
@@ -82,7 +82,7 @@ int dlm_posix_lock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 	if (!ls)
 		return -EINVAL;
 
-	xop = kzalloc(sizeof(*xop), GFP_KERNEL);
+	xop = kzalloc(sizeof(*xop), GFP_NOFS);
 	if (!xop) {
 		rv = -ENOMEM;
 		goto out;
@@ -211,7 +211,7 @@ int dlm_posix_unlock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 	if (!ls)
 		return -EINVAL;
 
-	op = kzalloc(sizeof(*op), GFP_KERNEL);
+	op = kzalloc(sizeof(*op), GFP_NOFS);
 	if (!op) {
 		rv = -ENOMEM;
 		goto out;
@@ -266,7 +266,7 @@ int dlm_posix_get(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 	if (!ls)
 		return -EINVAL;
 
-	op = kzalloc(sizeof(*op), GFP_KERNEL);
+	op = kzalloc(sizeof(*op), GFP_NOFS);
 	if (!op) {
 		rv = -ENOMEM;
 		goto out;
diff --git a/fs/dlm/rcom.c b/fs/dlm/rcom.c
index 67522c268c14..3c83a49a48a3 100644
--- a/fs/dlm/rcom.c
+++ b/fs/dlm/rcom.c
@@ -38,7 +38,7 @@ static int create_rcom(struct dlm_ls *ls, int to_nodeid, int type, int len,
 	char *mb;
 	int mb_len = sizeof(struct dlm_rcom) + len;
 
-	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, ls->ls_allocation, &mb);
+	mh = dlm_lowcomms_get_buffer(to_nodeid, mb_len, GFP_NOFS, &mb);
 	if (!mh) {
 		log_print("create_rcom to %d type %d len %d ENOBUFS",
 			  to_nodeid, type, len);
diff --git a/fs/dlm/requestqueue.c b/fs/dlm/requestqueue.c
index 7a2307c08911..a44fa22890e1 100644
--- a/fs/dlm/requestqueue.c
+++ b/fs/dlm/requestqueue.c
@@ -35,7 +35,7 @@ void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_message *ms)
 	struct rq_entry *e;
 	int length = ms->m_header.h_length - sizeof(struct dlm_message);
 
-	e = kmalloc(sizeof(struct rq_entry) + length, ls->ls_allocation);
+	e = kmalloc(sizeof(struct rq_entry) + length, GFP_NOFS);
 	if (!e) {
 		log_print("dlm_add_requestqueue: out of memory len %d", length);
 		return;
diff --git a/fs/dlm/user.c b/fs/dlm/user.c
index ebce994ab0b7..e73a4bb572aa 100644
--- a/fs/dlm/user.c
+++ b/fs/dlm/user.c
@@ -267,7 +267,7 @@ static int device_user_lock(struct dlm_user_proc *proc,
 		goto out;
 	}
 
-	ua = kzalloc(sizeof(struct dlm_user_args), GFP_KERNEL);
+	ua = kzalloc(sizeof(struct dlm_user_args), GFP_NOFS);
 	if (!ua)
 		goto out;
 	ua->proc = proc;
@@ -307,7 +307,7 @@ static int device_user_unlock(struct dlm_user_proc *proc,
 	if (!ls)
 		return -ENOENT;
 
-	ua = kzalloc(sizeof(struct dlm_user_args), GFP_KERNEL);
+	ua = kzalloc(sizeof(struct dlm_user_args), GFP_NOFS);
 	if (!ua)
 		goto out;
 	ua->proc = proc;
@@ -352,7 +352,7 @@ static int dlm_device_register(struct dlm_ls *ls, char *name)
 
 	error = -ENOMEM;
 	len = strlen(name) + strlen(name_prefix) + 2;
-	ls->ls_device.name = kzalloc(len, GFP_KERNEL);
+	ls->ls_device.name = kzalloc(len, GFP_NOFS);
 	if (!ls->ls_device.name)
 		goto fail;
 
@@ -520,7 +520,7 @@ static ssize_t device_write(struct file *file, const char __user *buf,
 #endif
 		return -EINVAL;
 
-	kbuf = kzalloc(count + 1, GFP_KERNEL);
+	kbuf = kzalloc(count + 1, GFP_NOFS);
 	if (!kbuf)
 		return -ENOMEM;
 
@@ -546,7 +546,7 @@ static ssize_t device_write(struct file *file, const char __user *buf,
 
 		/* add 1 after namelen so that the name string is terminated */
 		kbuf = kzalloc(sizeof(struct dlm_write_request) + namelen + 1,
-			       GFP_KERNEL);
+			       GFP_NOFS);
 		if (!kbuf) {
 			kfree(k32buf);
 			return -ENOMEM;
@@ -648,7 +648,7 @@ static int device_open(struct inode *inode, struct file *file)
 	if (!ls)
 		return -ENOENT;
 
-	proc = kzalloc(sizeof(struct dlm_user_proc), GFP_KERNEL);
+	proc = kzalloc(sizeof(struct dlm_user_proc), GFP_NOFS);
 	if (!proc) {
 		dlm_put_lockspace(ls);
 		return -ENOMEM;
diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index fbb6e5eed697..7cb0a59f4b9d 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -1748,7 +1748,7 @@ ecryptfs_process_key_cipher(struct crypto_blkcipher **key_tfm,
 			    char *cipher_name, size_t *key_size)
 {
 	char dummy_key[ECRYPTFS_MAX_KEY_BYTES];
-	char *full_alg_name;
+	char *full_alg_name = NULL;
 	int rc;
 
 	*key_tfm = NULL;
@@ -1763,7 +1763,6 @@ ecryptfs_process_key_cipher(struct crypto_blkcipher **key_tfm,
 	if (rc)
 		goto out;
 	*key_tfm = crypto_alloc_blkcipher(full_alg_name, 0, CRYPTO_ALG_ASYNC);
-	kfree(full_alg_name);
 	if (IS_ERR(*key_tfm)) {
 		rc = PTR_ERR(*key_tfm);
 		printk(KERN_ERR "Unable to allocate crypto cipher with name "
@@ -1786,6 +1785,7 @@ ecryptfs_process_key_cipher(struct crypto_blkcipher **key_tfm,
 		goto out;
 	}
 out:
+	kfree(full_alg_name);
 	return rc;
 }
 
diff --git a/fs/ecryptfs/dentry.c b/fs/ecryptfs/dentry.c
index 2dda5ade75bc..8f006a0d6076 100644
--- a/fs/ecryptfs/dentry.c
+++ b/fs/ecryptfs/dentry.c
@@ -62,7 +62,7 @@ static int ecryptfs_d_revalidate(struct dentry *dentry, struct nameidata *nd)
 		struct inode *lower_inode =
 			ecryptfs_inode_to_lower(dentry->d_inode);
 
-		fsstack_copy_attr_all(dentry->d_inode, lower_inode, NULL);
+		fsstack_copy_attr_all(dentry->d_inode, lower_inode);
 	}
 out:
 	return rc;
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index 9e944057001b..678172b61be2 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -158,7 +158,7 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 	struct dentry *ecryptfs_dentry = file->f_path.dentry;
 	/* Private value of ecryptfs_dentry allocated in
 	 * ecryptfs_lookup() */
-	struct dentry *lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry);
+	struct dentry *lower_dentry;
 	struct ecryptfs_file_info *file_info;
 
 	mount_crypt_stat = &ecryptfs_superblock_to_private(
@@ -191,13 +191,6 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 				      | ECRYPTFS_ENCRYPTED);
 	}
 	mutex_unlock(&crypt_stat->cs_mutex);
-	if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_RDONLY)
-	    && !(file->f_flags & O_RDONLY)) {
-		rc = -EPERM;
-		printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs "
-		       "file must hence be opened RO\n", __func__);
-		goto out;
-	}
 	if (!ecryptfs_inode_to_private(inode)->lower_file) {
 		rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
 		if (rc) {
@@ -208,6 +201,13 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 			goto out;
 		}
 	}
+	if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_RDONLY)
+	    && !(file->f_flags & O_RDONLY)) {
+		rc = -EPERM;
+		printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs "
+		       "file must hence be opened RO\n", __func__);
+		goto out;
+	}
 	ecryptfs_set_file_lower(
 		file, ecryptfs_inode_to_private(inode)->lower_file);
 	if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) {
@@ -299,7 +299,6 @@ static int ecryptfs_ioctl(struct inode *inode, struct file *file,
 const struct file_operations ecryptfs_dir_fops = {
 	.readdir = ecryptfs_readdir,
 	.ioctl = ecryptfs_ioctl,
-	.mmap = generic_file_mmap,
 	.open = ecryptfs_open,
 	.flush = ecryptfs_flush,
 	.release = ecryptfs_release,
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 056fed62d0de..4a430ab4115c 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -282,7 +282,8 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry,
 		goto out;
 	}
 	rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry,
-				ecryptfs_dir_inode->i_sb, 1);
+				ecryptfs_dir_inode->i_sb,
+				ECRYPTFS_INTERPOSE_FLAG_D_ADD);
 	if (rc) {
 		printk(KERN_ERR "%s: Error interposing; rc = [%d]\n",
 		       __func__, rc);
@@ -463,9 +464,6 @@ out_lock:
 	unlock_dir(lower_dir_dentry);
 	dput(lower_new_dentry);
 	dput(lower_old_dentry);
-	d_drop(lower_old_dentry);
-	d_drop(new_dentry);
-	d_drop(old_dentry);
 	return rc;
 }
 
@@ -614,6 +612,7 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct dentry *lower_new_dentry;
 	struct dentry *lower_old_dir_dentry;
 	struct dentry *lower_new_dir_dentry;
+	struct dentry *trap = NULL;
 
 	lower_old_dentry = ecryptfs_dentry_to_lower(old_dentry);
 	lower_new_dentry = ecryptfs_dentry_to_lower(new_dentry);
@@ -621,14 +620,24 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	dget(lower_new_dentry);
 	lower_old_dir_dentry = dget_parent(lower_old_dentry);
 	lower_new_dir_dentry = dget_parent(lower_new_dentry);
-	lock_rename(lower_old_dir_dentry, lower_new_dir_dentry);
+	trap = lock_rename(lower_old_dir_dentry, lower_new_dir_dentry);
+	/* source should not be ancestor of target */
+	if (trap == lower_old_dentry) {
+		rc = -EINVAL;
+		goto out_lock;
+	}
+	/* target should not be ancestor of source */
+	if (trap == lower_new_dentry) {
+		rc = -ENOTEMPTY;
+		goto out_lock;
+	}
 	rc = vfs_rename(lower_old_dir_dentry->d_inode, lower_old_dentry,
 			lower_new_dir_dentry->d_inode, lower_new_dentry);
 	if (rc)
 		goto out_lock;
-	fsstack_copy_attr_all(new_dir, lower_new_dir_dentry->d_inode, NULL);
+	fsstack_copy_attr_all(new_dir, lower_new_dir_dentry->d_inode);
 	if (new_dir != old_dir)
-		fsstack_copy_attr_all(old_dir, lower_old_dir_dentry->d_inode, NULL);
+		fsstack_copy_attr_all(old_dir, lower_old_dir_dentry->d_inode);
 out_lock:
 	unlock_rename(lower_old_dir_dentry, lower_new_dir_dentry);
 	dput(lower_new_dentry->d_parent);
@@ -715,31 +724,31 @@ static void *ecryptfs_follow_link(struct dentry *dentry, struct nameidata *nd)
 	/* Released in ecryptfs_put_link(); only release here on error */
 	buf = kmalloc(len, GFP_KERNEL);
 	if (!buf) {
-		rc = -ENOMEM;
+		buf = ERR_PTR(-ENOMEM);
 		goto out;
 	}
 	old_fs = get_fs();
 	set_fs(get_ds());
 	rc = dentry->d_inode->i_op->readlink(dentry, (char __user *)buf, len);
 	set_fs(old_fs);
-	if (rc < 0)
-		goto out_free;
-	else
+	if (rc < 0) {
+		kfree(buf);
+		buf = ERR_PTR(rc);
+	} else
 		buf[rc] = '\0';
-	rc = 0;
-	nd_set_link(nd, buf);
-	goto out;
-out_free:
-	kfree(buf);
 out:
-	return ERR_PTR(rc);
+	nd_set_link(nd, buf);
+	return NULL;
 }
 
 static void
 ecryptfs_put_link(struct dentry *dentry, struct nameidata *nd, void *ptr)
 {
-	/* Free the char* */
-	kfree(nd_get_link(nd));
+	char *buf = nd_get_link(nd);
+	if (!IS_ERR(buf)) {
+		/* Free the char* */
+		kfree(buf);
+	}
 }
 
 /**
@@ -772,18 +781,23 @@ upper_size_to_lower_size(struct ecryptfs_crypt_stat *crypt_stat,
 }
 
 /**
- * ecryptfs_truncate
+ * truncate_upper
  * @dentry: The ecryptfs layer dentry
- * @new_length: The length to expand the file to
+ * @ia: Address of the ecryptfs inode's attributes
+ * @lower_ia: Address of the lower inode's attributes
  *
  * Function to handle truncations modifying the size of the file. Note
  * that the file sizes are interpolated. When expanding, we are simply
- * writing strings of 0's out. When truncating, we need to modify the
- * underlying file size according to the page index interpolations.
+ * writing strings of 0's out. When truncating, we truncate the upper
+ * inode and update the lower_ia according to the page index
+ * interpolations. If ATTR_SIZE is set in lower_ia->ia_valid upon return,
+ * the caller must use lower_ia in a call to notify_change() to perform
+ * the truncation of the lower inode.
  *
  * Returns zero on success; non-zero otherwise
  */
-int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
+static int truncate_upper(struct dentry *dentry, struct iattr *ia,
+			  struct iattr *lower_ia)
 {
 	int rc = 0;
 	struct inode *inode = dentry->d_inode;
@@ -794,8 +808,10 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 	loff_t lower_size_before_truncate;
 	loff_t lower_size_after_truncate;
 
-	if (unlikely((new_length == i_size)))
+	if (unlikely((ia->ia_size == i_size))) {
+		lower_ia->ia_valid &= ~ATTR_SIZE;
 		goto out;
+	}
 	crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat;
 	/* Set up a fake ecryptfs file, this is used to interface with
 	 * the file in the underlying filesystem so that the
@@ -815,28 +831,30 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 		&fake_ecryptfs_file,
 		ecryptfs_inode_to_private(dentry->d_inode)->lower_file);
 	/* Switch on growing or shrinking file */
-	if (new_length > i_size) {
+	if (ia->ia_size > i_size) {
 		char zero[] = { 0x00 };
 
+		lower_ia->ia_valid &= ~ATTR_SIZE;
 		/* Write a single 0 at the last position of the file;
 		 * this triggers code that will fill in 0's throughout
 		 * the intermediate portion of the previous end of the
 		 * file and the new and of the file */
 		rc = ecryptfs_write(&fake_ecryptfs_file, zero,
-				    (new_length - 1), 1);
-	} else { /* new_length < i_size_read(inode) */
-		/* We're chopping off all the pages down do the page
-		 * in which new_length is located. Fill in the end of
-		 * that page from (new_length & ~PAGE_CACHE_MASK) to
+				    (ia->ia_size - 1), 1);
+	} else { /* ia->ia_size < i_size_read(inode) */
+		/* We're chopping off all the pages down to the page
+		 * in which ia->ia_size is located. Fill in the end of
+		 * that page from (ia->ia_size & ~PAGE_CACHE_MASK) to
 		 * PAGE_CACHE_SIZE with zeros. */
 		size_t num_zeros = (PAGE_CACHE_SIZE
-				    - (new_length & ~PAGE_CACHE_MASK));
+				    - (ia->ia_size & ~PAGE_CACHE_MASK));
 
 		if (!(crypt_stat->flags & ECRYPTFS_ENCRYPTED)) {
-			rc = vmtruncate(inode, new_length);
+			rc = vmtruncate(inode, ia->ia_size);
 			if (rc)
 				goto out_free;
-			rc = vmtruncate(lower_dentry->d_inode, new_length);
+			lower_ia->ia_size = ia->ia_size;
+			lower_ia->ia_valid |= ATTR_SIZE;
 			goto out_free;
 		}
 		if (num_zeros) {
@@ -848,7 +866,7 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 				goto out_free;
 			}
 			rc = ecryptfs_write(&fake_ecryptfs_file, zeros_virt,
-					    new_length, num_zeros);
+					    ia->ia_size, num_zeros);
 			kfree(zeros_virt);
 			if (rc) {
 				printk(KERN_ERR "Error attempting to zero out "
@@ -857,7 +875,7 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 				goto out_free;
 			}
 		}
-		vmtruncate(inode, new_length);
+		vmtruncate(inode, ia->ia_size);
 		rc = ecryptfs_write_inode_size_to_metadata(inode);
 		if (rc) {
 			printk(KERN_ERR	"Problem with "
@@ -870,10 +888,12 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
 		lower_size_before_truncate =
 		    upper_size_to_lower_size(crypt_stat, i_size);
 		lower_size_after_truncate =
-		    upper_size_to_lower_size(crypt_stat, new_length);
-		if (lower_size_after_truncate < lower_size_before_truncate)
-			vmtruncate(lower_dentry->d_inode,
-				   lower_size_after_truncate);
+		    upper_size_to_lower_size(crypt_stat, ia->ia_size);
+		if (lower_size_after_truncate < lower_size_before_truncate) {
+			lower_ia->ia_size = lower_size_after_truncate;
+			lower_ia->ia_valid |= ATTR_SIZE;
+		} else
+			lower_ia->ia_valid &= ~ATTR_SIZE;
 	}
 out_free:
 	if (ecryptfs_file_to_private(&fake_ecryptfs_file))
@@ -883,6 +903,33 @@ out:
 	return rc;
 }
 
+/**
+ * ecryptfs_truncate
+ * @dentry: The ecryptfs layer dentry
+ * @new_length: The length to expand the file to
+ *
+ * Simple function that handles the truncation of an eCryptfs inode and
+ * its corresponding lower inode.
+ *
+ * Returns zero on success; non-zero otherwise
+ */
+int ecryptfs_truncate(struct dentry *dentry, loff_t new_length)
+{
+	struct iattr ia = { .ia_valid = ATTR_SIZE, .ia_size = new_length };
+	struct iattr lower_ia = { .ia_valid = 0 };
+	int rc;
+
+	rc = truncate_upper(dentry, &ia, &lower_ia);
+	if (!rc && lower_ia.ia_valid & ATTR_SIZE) {
+		struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry);
+
+		mutex_lock(&lower_dentry->d_inode->i_mutex);
+		rc = notify_change(lower_dentry, &lower_ia);
+		mutex_unlock(&lower_dentry->d_inode->i_mutex);
+	}
+	return rc;
+}
+
 static int
 ecryptfs_permission(struct inode *inode, int mask)
 {
@@ -905,6 +952,7 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia)
 {
 	int rc = 0;
 	struct dentry *lower_dentry;
+	struct iattr lower_ia;
 	struct inode *inode;
 	struct inode *lower_inode;
 	struct ecryptfs_crypt_stat *crypt_stat;
@@ -943,15 +991,11 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia)
 		}
 	}
 	mutex_unlock(&crypt_stat->cs_mutex);
+	memcpy(&lower_ia, ia, sizeof(lower_ia));
+	if (ia->ia_valid & ATTR_FILE)
+		lower_ia.ia_file = ecryptfs_file_to_lower(ia->ia_file);
 	if (ia->ia_valid & ATTR_SIZE) {
-		ecryptfs_printk(KERN_DEBUG,
-				"ia->ia_valid = [0x%x] ATTR_SIZE" " = [0x%x]\n",
-				ia->ia_valid, ATTR_SIZE);
-		rc = ecryptfs_truncate(dentry, ia->ia_size);
-		/* ecryptfs_truncate handles resizing of the lower file */
-		ia->ia_valid &= ~ATTR_SIZE;
-		ecryptfs_printk(KERN_DEBUG, "ia->ia_valid = [%x]\n",
-				ia->ia_valid);
+		rc = truncate_upper(dentry, ia, &lower_ia);
 		if (rc < 0)
 			goto out;
 	}
@@ -960,14 +1004,29 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia)
 	 * mode change is for clearing setuid/setgid bits. Allow lower fs
 	 * to interpret this in its own way.
 	 */
-	if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
-		ia->ia_valid &= ~ATTR_MODE;
+	if (lower_ia.ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
+		lower_ia.ia_valid &= ~ATTR_MODE;
 
 	mutex_lock(&lower_dentry->d_inode->i_mutex);
-	rc = notify_change(lower_dentry, ia);
+	rc = notify_change(lower_dentry, &lower_ia);
 	mutex_unlock(&lower_dentry->d_inode->i_mutex);
 out:
-	fsstack_copy_attr_all(inode, lower_inode, NULL);
+	fsstack_copy_attr_all(inode, lower_inode);
+	return rc;
+}
+
+int ecryptfs_getattr(struct vfsmount *mnt, struct dentry *dentry,
+		     struct kstat *stat)
+{
+	struct kstat lower_stat;
+	int rc;
+
+	rc = vfs_getattr(ecryptfs_dentry_to_lower_mnt(dentry),
+			 ecryptfs_dentry_to_lower(dentry), &lower_stat);
+	if (!rc) {
+		generic_fillattr(dentry->d_inode, stat);
+		stat->blocks = lower_stat.blocks;
+	}
 	return rc;
 }
 
@@ -1100,6 +1159,7 @@ const struct inode_operations ecryptfs_dir_iops = {
 const struct inode_operations ecryptfs_main_iops = {
 	.permission = ecryptfs_permission,
 	.setattr = ecryptfs_setattr,
+	.getattr = ecryptfs_getattr,
 	.setxattr = ecryptfs_setxattr,
 	.getxattr = ecryptfs_getxattr,
 	.listxattr = ecryptfs_listxattr,
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index c6ac85d6c701..ea2f92101dfe 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -35,7 +35,6 @@
 #include <linux/key.h>
 #include <linux/parser.h>
 #include <linux/fs_stack.h>
-#include <linux/ima.h>
 #include "ecryptfs_kernel.h"
 
 /**
@@ -119,7 +118,6 @@ int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
 	const struct cred *cred = current_cred();
 	struct ecryptfs_inode_info *inode_info =
 		ecryptfs_inode_to_private(ecryptfs_dentry->d_inode);
-	int opened_lower_file = 0;
 	int rc = 0;
 
 	mutex_lock(&inode_info->lower_file_mutex);
@@ -136,12 +134,9 @@ int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
 			       "for lower_dentry [0x%p] and lower_mnt [0x%p]; "
 			       "rc = [%d]\n", lower_dentry, lower_mnt, rc);
 			inode_info->lower_file = NULL;
-		} else
-			opened_lower_file = 1;
+		}
 	}
 	mutex_unlock(&inode_info->lower_file_mutex);
-	if (opened_lower_file)
-		ima_counts_get(inode_info->lower_file);
 	return rc;
 }
 
@@ -194,7 +189,7 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry,
 		init_special_inode(inode, lower_inode->i_mode,
 				   lower_inode->i_rdev);
 	dentry->d_op = &ecryptfs_dops;
-	fsstack_copy_attr_all(inode, lower_inode, NULL);
+	fsstack_copy_attr_all(inode, lower_inode);
 	/* This size will be overwritten for real files w/ headers and
 	 * other metadata */
 	fsstack_copy_inode_size(inode, lower_inode);
@@ -590,8 +585,8 @@ out:
  *                        with as much information as it can before needing
  *                        the lower filesystem.
  * ecryptfs_read_super(): this accesses the lower filesystem and uses
- *                        ecryptfs_interpolate to perform most of the linking
- * ecryptfs_interpolate(): links the lower filesystem into ecryptfs
+ *                        ecryptfs_interpose to perform most of the linking
+ * ecryptfs_interpose(): links the lower filesystem into ecryptfs (inode.c)
  */
 static int ecryptfs_get_sb(struct file_system_type *fs_type, int flags,
 			const char *dev_name, void *raw_data,
diff --git a/fs/eventfd.c b/fs/eventfd.c
index 8b47e4200e65..7758cc382ef0 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -135,26 +135,71 @@ static unsigned int eventfd_poll(struct file *file, poll_table *wait)
 	return events;
 }
 
-static ssize_t eventfd_read(struct file *file, char __user *buf, size_t count,
-			    loff_t *ppos)
+static void eventfd_ctx_do_read(struct eventfd_ctx *ctx, __u64 *cnt)
+{
+	*cnt = (ctx->flags & EFD_SEMAPHORE) ? 1 : ctx->count;
+	ctx->count -= *cnt;
+}
+
+/**
+ * eventfd_ctx_remove_wait_queue - Read the current counter and removes wait queue.
+ * @ctx: [in] Pointer to eventfd context.
+ * @wait: [in] Wait queue to be removed.
+ * @cnt: [out] Pointer to the 64bit conter value.
+ *
+ * Returns zero if successful, or the following error codes:
+ *
+ * -EAGAIN      : The operation would have blocked.
+ *
+ * This is used to atomically remove a wait queue entry from the eventfd wait
+ * queue head, and read/reset the counter value.
+ */
+int eventfd_ctx_remove_wait_queue(struct eventfd_ctx *ctx, wait_queue_t *wait,
+				  __u64 *cnt)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&ctx->wqh.lock, flags);
+	eventfd_ctx_do_read(ctx, cnt);
+	__remove_wait_queue(&ctx->wqh, wait);
+	if (*cnt != 0 && waitqueue_active(&ctx->wqh))
+		wake_up_locked_poll(&ctx->wqh, POLLOUT);
+	spin_unlock_irqrestore(&ctx->wqh.lock, flags);
+
+	return *cnt != 0 ? 0 : -EAGAIN;
+}
+EXPORT_SYMBOL_GPL(eventfd_ctx_remove_wait_queue);
+
+/**
+ * eventfd_ctx_read - Reads the eventfd counter or wait if it is zero.
+ * @ctx: [in] Pointer to eventfd context.
+ * @no_wait: [in] Different from zero if the operation should not block.
+ * @cnt: [out] Pointer to the 64bit conter value.
+ *
+ * Returns zero if successful, or the following error codes:
+ *
+ * -EAGAIN      : The operation would have blocked but @no_wait was nonzero.
+ * -ERESTARTSYS : A signal interrupted the wait operation.
+ *
+ * If @no_wait is zero, the function might sleep until the eventfd internal
+ * counter becomes greater than zero.
+ */
+ssize_t eventfd_ctx_read(struct eventfd_ctx *ctx, int no_wait, __u64 *cnt)
 {
-	struct eventfd_ctx *ctx = file->private_data;
 	ssize_t res;
-	__u64 ucnt = 0;
 	DECLARE_WAITQUEUE(wait, current);
 
-	if (count < sizeof(ucnt))
-		return -EINVAL;
 	spin_lock_irq(&ctx->wqh.lock);
+	*cnt = 0;
 	res = -EAGAIN;
 	if (ctx->count > 0)
-		res = sizeof(ucnt);
-	else if (!(file->f_flags & O_NONBLOCK)) {
+		res = 0;
+	else if (!no_wait) {
 		__add_wait_queue(&ctx->wqh, &wait);
-		for (res = 0;;) {
+		for (;;) {
 			set_current_state(TASK_INTERRUPTIBLE);
 			if (ctx->count > 0) {
-				res = sizeof(ucnt);
+				res = 0;
 				break;
 			}
 			if (signal_pending(current)) {
@@ -168,18 +213,32 @@ static ssize_t eventfd_read(struct file *file, char __user *buf, size_t count,
 		__remove_wait_queue(&ctx->wqh, &wait);
 		__set_current_state(TASK_RUNNING);
 	}
-	if (likely(res > 0)) {
-		ucnt = (ctx->flags & EFD_SEMAPHORE) ? 1 : ctx->count;
-		ctx->count -= ucnt;
+	if (likely(res == 0)) {
+		eventfd_ctx_do_read(ctx, cnt);
 		if (waitqueue_active(&ctx->wqh))
 			wake_up_locked_poll(&ctx->wqh, POLLOUT);
 	}
 	spin_unlock_irq(&ctx->wqh.lock);
-	if (res > 0 && put_user(ucnt, (__u64 __user *) buf))
-		return -EFAULT;
 
 	return res;
 }
+EXPORT_SYMBOL_GPL(eventfd_ctx_read);
+
+static ssize_t eventfd_read(struct file *file, char __user *buf, size_t count,
+			    loff_t *ppos)
+{
+	struct eventfd_ctx *ctx = file->private_data;
+	ssize_t res;
+	__u64 cnt;
+
+	if (count < sizeof(cnt))
+		return -EINVAL;
+	res = eventfd_ctx_read(ctx, file->f_flags & O_NONBLOCK, &cnt);
+	if (res < 0)
+		return res;
+
+	return put_user(cnt, (__u64 __user *) buf) ? -EFAULT : sizeof(cnt);
+}
 
 static ssize_t eventfd_write(struct file *file, const char __user *buf, size_t count,
 			     loff_t *ppos)
@@ -339,7 +398,7 @@ struct file *eventfd_file_create(unsigned int count, int flags)
 	ctx->flags = flags;
 
 	file = anon_inode_getfile("[eventfd]", &eventfd_fops, ctx,
-				  flags & EFD_SHARED_FCNTL_FLAGS);
+				  O_RDWR | (flags & EFD_SHARED_FCNTL_FLAGS));
 	if (IS_ERR(file))
 		eventfd_free_ctx(ctx);
 
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 366c503f9657..bd056a5b4efc 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1206,7 +1206,7 @@ SYSCALL_DEFINE1(epoll_create1, int, flags)
 	 * a file structure and a free file descriptor.
 	 */
 	error = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep,
-				 flags & O_CLOEXEC);
+				 O_RDWR | (flags & O_CLOEXEC));
 	if (error < 0)
 		ep_free(ep);
 
diff --git a/fs/exec.c b/fs/exec.c
index c0c636e34f60..0790a107ff7e 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -826,7 +826,9 @@ static int de_thread(struct task_struct *tsk)
 		attach_pid(tsk, PIDTYPE_PID,  task_pid(leader));
 		transfer_pid(leader, tsk, PIDTYPE_PGID);
 		transfer_pid(leader, tsk, PIDTYPE_SID);
+
 		list_replace_rcu(&leader->tasks, &tsk->tasks);
+		list_replace_init(&leader->sibling, &tsk->sibling);
 
 		tsk->group_leader = tsk;
 		leader->group_leader = tsk;
@@ -923,6 +925,15 @@ char *get_task_comm(char *buf, struct task_struct *tsk)
 void set_task_comm(struct task_struct *tsk, char *buf)
 {
 	task_lock(tsk);
+
+	/*
+	 * Threads may access current->comm without holding
+	 * the task lock, so write the string carefully.
+	 * Readers without a lock may see incomplete new
+	 * names but are safe from non-terminating string reads.
+	 */
+	memset(tsk->comm, 0, TASK_COMM_LEN);
+	wmb();
 	strlcpy(tsk->comm, buf, sizeof(tsk->comm));
 	task_unlock(tsk);
 	perf_event_comm(tsk);
@@ -930,9 +941,7 @@ void set_task_comm(struct task_struct *tsk, char *buf)
 
 int flush_old_exec(struct linux_binprm * bprm)
 {
-	char * name;
-	int i, ch, retval;
-	char tcomm[sizeof(current->comm)];
+	int retval;
 
 	/*
 	 * Make sure we have a private signal table and that
@@ -953,6 +962,25 @@ int flush_old_exec(struct linux_binprm * bprm)
 
 	bprm->mm = NULL;		/* We're using it now */
 
+	current->flags &= ~PF_RANDOMIZE;
+	flush_thread();
+	current->personality &= ~bprm->per_clear;
+
+	return 0;
+
+out:
+	return retval;
+}
+EXPORT_SYMBOL(flush_old_exec);
+
+void setup_new_exec(struct linux_binprm * bprm)
+{
+	int i, ch;
+	char * name;
+	char tcomm[sizeof(current->comm)];
+
+	arch_pick_mmap_layout(current->mm);
+
 	/* This is the point of no return */
 	current->sas_ss_sp = current->sas_ss_size = 0;
 
@@ -974,9 +1002,6 @@ int flush_old_exec(struct linux_binprm * bprm)
 	tcomm[i] = '\0';
 	set_task_comm(current, tcomm);
 
-	current->flags &= ~PF_RANDOMIZE;
-	flush_thread();
-
 	/* Set the new mm task size. We have to do that late because it may
 	 * depend on TIF_32BIT which is only updated in flush_thread() on
 	 * some architectures like powerpc
@@ -992,8 +1017,6 @@ int flush_old_exec(struct linux_binprm * bprm)
 		set_dumpable(current->mm, suid_dumpable);
 	}
 
-	current->personality &= ~bprm->per_clear;
-
 	/*
 	 * Flush performance counters when crossing a
 	 * security domain:
@@ -1008,14 +1031,8 @@ int flush_old_exec(struct linux_binprm * bprm)
 			
 	flush_signal_handlers(current, 0);
 	flush_old_files(current->files);
-
-	return 0;
-
-out:
-	return retval;
 }
-
-EXPORT_SYMBOL(flush_old_exec);
+EXPORT_SYMBOL(setup_new_exec);
 
 /*
  * Prepare credentials and lock ->cred_guard_mutex.
@@ -1752,17 +1769,20 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs)
 	struct mm_struct *mm = current->mm;
 	struct linux_binfmt * binfmt;
 	struct inode * inode;
-	struct file * file;
 	const struct cred *old_cred;
 	struct cred *cred;
 	int retval = 0;
 	int flag = 0;
 	int ispipe = 0;
-	unsigned long core_limit = current->signal->rlim[RLIMIT_CORE].rlim_cur;
 	char **helper_argv = NULL;
 	int helper_argc = 0;
 	int dump_count = 0;
 	static atomic_t core_dump_count = ATOMIC_INIT(0);
+	struct coredump_params cprm = {
+		.signr = signr,
+		.regs = regs,
+		.limit = current->signal->rlim[RLIMIT_CORE].rlim_cur,
+	};
 
 	audit_core_dumps(signr);
 
@@ -1818,15 +1838,15 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs)
 	ispipe = format_corename(corename, signr);
 	unlock_kernel();
 
-	if ((!ispipe) && (core_limit < binfmt->min_coredump))
+	if ((!ispipe) && (cprm.limit < binfmt->min_coredump))
 		goto fail_unlock;
 
  	if (ispipe) {
-		if (core_limit == 0) {
+		if (cprm.limit == 0) {
 			/*
 			 * Normally core limits are irrelevant to pipes, since
 			 * we're not writing to the file system, but we use
-			 * core_limit of 0 here as a speacial value. Any
+			 * cprm.limit of 0 here as a speacial value. Any
 			 * non-zero limit gets set to RLIM_INFINITY below, but
 			 * a limit of 0 skips the dump.  This is a consistent
 			 * way to catch recursive crashes.  We can still crash
@@ -1859,25 +1879,25 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs)
 			goto fail_dropcount;
 		}
 
-		core_limit = RLIM_INFINITY;
+		cprm.limit = RLIM_INFINITY;
 
 		/* SIGPIPE can happen, but it's just never processed */
 		if (call_usermodehelper_pipe(helper_argv[0], helper_argv, NULL,
-				&file)) {
+				&cprm.file)) {
  			printk(KERN_INFO "Core dump to %s pipe failed\n",
 			       corename);
 			goto fail_dropcount;
  		}
  	} else
- 		file = filp_open(corename,
+		cprm.file = filp_open(corename,
 				 O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag,
 				 0600);
-	if (IS_ERR(file))
+	if (IS_ERR(cprm.file))
 		goto fail_dropcount;
-	inode = file->f_path.dentry->d_inode;
+	inode = cprm.file->f_path.dentry->d_inode;
 	if (inode->i_nlink > 1)
 		goto close_fail;	/* multiple links - don't dump */
-	if (!ispipe && d_unhashed(file->f_path.dentry))
+	if (!ispipe && d_unhashed(cprm.file->f_path.dentry))
 		goto close_fail;
 
 	/* AK: actually i see no reason to not allow this for named pipes etc.,
@@ -1890,21 +1910,22 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs)
 	 */
 	if (inode->i_uid != current_fsuid())
 		goto close_fail;
-	if (!file->f_op)
+	if (!cprm.file->f_op)
 		goto close_fail;
-	if (!file->f_op->write)
+	if (!cprm.file->f_op->write)
 		goto close_fail;
-	if (!ispipe && do_truncate(file->f_path.dentry, 0, 0, file) != 0)
+	if (!ispipe &&
+	    do_truncate(cprm.file->f_path.dentry, 0, 0, cprm.file) != 0)
 		goto close_fail;
 
-	retval = binfmt->core_dump(signr, regs, file, core_limit);
+	retval = binfmt->core_dump(&cprm);
 
 	if (retval)
 		current->signal->group_exit_code |= 0x80;
 close_fail:
 	if (ispipe && core_pipe_limit)
-		wait_for_dump_helpers(file);
-	filp_close(file, NULL);
+		wait_for_dump_helpers(cprm.file);
+	filp_close(cprm.file, NULL);
 fail_dropcount:
 	if (dump_count)
 		atomic_dec(&core_dump_count);
diff --git a/fs/exofs/Kbuild b/fs/exofs/Kbuild
index cc2d22db119c..2d0f757fda3e 100644
--- a/fs/exofs/Kbuild
+++ b/fs/exofs/Kbuild
@@ -12,5 +12,5 @@
 # Kbuild - Gets included from the Kernels Makefile and build system
 #
 
-exofs-y := osd.o inode.o file.o symlink.o namei.o dir.o super.o
+exofs-y := ios.o inode.o file.o symlink.o namei.o dir.o super.o
 obj-$(CONFIG_EXOFS_FS) += exofs.o
diff --git a/fs/exofs/common.h b/fs/exofs/common.h
index c6718e4817fe..b1b178e61718 100644
--- a/fs/exofs/common.h
+++ b/fs/exofs/common.h
@@ -49,6 +49,7 @@
 #define EXOFS_MIN_PID   0x10000	/* Smallest partition ID */
 #define EXOFS_OBJ_OFF	0x10000	/* offset for objects */
 #define EXOFS_SUPER_ID	0x10000	/* object ID for on-disk superblock */
+#define EXOFS_DEVTABLE_ID 0x10001 /* object ID for on-disk device table */
 #define EXOFS_ROOT_ID	0x10002	/* object ID for root directory */
 
 /* exofs Application specific page/attribute */
@@ -78,17 +79,67 @@ enum {
 #define EXOFS_SUPER_MAGIC	0x5DF5
 
 /*
- * The file system control block - stored in an object's data (mainly, the one
- * with ID EXOFS_SUPER_ID).  This is where the in-memory superblock is stored
- * on disk.  Right now it just has a magic value, which is basically a sanity
- * check on our ability to communicate with the object store.
+ * The file system control block - stored in object EXOFS_SUPER_ID's data.
+ * This is where the in-memory superblock is stored on disk.
  */
+enum {EXOFS_FSCB_VER = 1, EXOFS_DT_VER = 1};
 struct exofs_fscb {
 	__le64  s_nextid;	/* Highest object ID used */
-	__le32  s_numfiles;	/* Number of files on fs */
+	__le64  s_numfiles;	/* Number of files on fs */
+	__le32	s_version;	/* == EXOFS_FSCB_VER */
 	__le16  s_magic;	/* Magic signature */
 	__le16  s_newfs;	/* Non-zero if this is a new fs */
-};
+
+	/* From here on it's a static part, only written by mkexofs */
+	__le64	s_dev_table_oid;   /* Resurved, not used */
+	__le64	s_dev_table_count; /* == 0 means no dev_table */
+} __packed;
+
+/*
+ * Describes the raid used in the FS. It is part of the device table.
+ * This here is taken from the pNFS-objects definition. In exofs we
+ * use one raid policy through-out the filesystem. (NOTE: the funny
+ * alignment at begining. We take care of it at exofs_device_table.
+ */
+struct exofs_dt_data_map {
+	__le32	cb_num_comps;
+	__le64	cb_stripe_unit;
+	__le32	cb_group_width;
+	__le32	cb_group_depth;
+	__le32	cb_mirror_cnt;
+	__le32	cb_raid_algorithm;
+} __packed;
+
+/*
+ * This is an osd device information descriptor. It is a single entry in
+ * the exofs device table. It describes an osd target lun which
+ * contains data belonging to this FS. (Same partition_id on all devices)
+ */
+struct exofs_dt_device_info {
+	__le32	systemid_len;
+	u8	systemid[OSD_SYSTEMID_LEN];
+	__le64	long_name_offset;	/* If !0 then offset-in-file */
+	__le32	osdname_len;		/* */
+	u8	osdname[44];		/* Embbeded, Ususally an asci uuid */
+} __packed;
+
+/*
+ * The EXOFS device table - stored in object EXOFS_DEVTABLE_ID's data.
+ * It contains the raid used for this multy-device FS and an array of
+ * participating devices.
+ */
+struct exofs_device_table {
+	__le32				dt_version;	/* == EXOFS_DT_VER */
+	struct exofs_dt_data_map	dt_data_map;	/* Raid policy to use */
+
+	/* Resurved space For future use. Total includeing this:
+	 * (8 * sizeof(le64))
+	 */
+	__le64				__Resurved[4];
+
+	__le64				dt_num_devices;	/* Array size */
+	struct exofs_dt_device_info	dt_dev_table[];	/* Array of devices */
+} __packed;
 
 /****************************************************************************
  * inode-related things
@@ -155,22 +206,4 @@ enum {
 	(((name_len) + offsetof(struct exofs_dir_entry, name)  + \
 	  EXOFS_DIR_ROUND) & ~EXOFS_DIR_ROUND)
 
-/*************************
- * function declarations *
- *************************/
-/* osd.c                 */
-void exofs_make_credential(u8 cred_a[OSD_CAP_LEN],
-			   const struct osd_obj_id *obj);
-
-int exofs_check_ok_resid(struct osd_request *or, u64 *in_resid, u64 *out_resid);
-static inline int exofs_check_ok(struct osd_request *or)
-{
-	return exofs_check_ok_resid(or, NULL, NULL);
-}
-int exofs_sync_op(struct osd_request *or, int timeout, u8 *cred);
-int exofs_async_op(struct osd_request *or,
-	osd_req_done_fn *async_done, void *caller_context, u8 *cred);
-
-int extract_attr_from_req(struct osd_request *or, struct osd_attr *attr);
-
 #endif /*ifndef __EXOFS_COM_H__*/
diff --git a/fs/exofs/exofs.h b/fs/exofs/exofs.h
index 5ec72e020b22..c35fd4623986 100644
--- a/fs/exofs/exofs.h
+++ b/fs/exofs/exofs.h
@@ -30,13 +30,17 @@
  * along with exofs; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
+#ifndef __EXOFS_H__
+#define __EXOFS_H__
 
 #include <linux/fs.h>
 #include <linux/time.h>
 #include "common.h"
 
-#ifndef __EXOFS_H__
-#define __EXOFS_H__
+/* FIXME: Remove once pnfs hits mainline
+ * #include <linux/exportfs/pnfs_osd_xdr.h>
+ */
+#include "pnfs.h"
 
 #define EXOFS_ERR(fmt, a...) printk(KERN_ERR "exofs: " fmt, ##a)
 
@@ -55,7 +59,7 @@
  * our extension to the in-memory superblock
  */
 struct exofs_sb_info {
-	struct osd_dev	*s_dev;			/* returned by get_osd_dev    */
+	struct exofs_fscb s_fscb;		/* Written often, pre-allocate*/
 	osd_id		s_pid;			/* partition ID of file system*/
 	int		s_timeout;		/* timeout for OSD operations */
 	uint64_t	s_nextid;		/* highest object ID used     */
@@ -63,7 +67,11 @@ struct exofs_sb_info {
 	spinlock_t	s_next_gen_lock;	/* spinlock for gen # update  */
 	u32		s_next_generation;	/* next gen # to use          */
 	atomic_t	s_curr_pending;		/* number of pending commands */
-	uint8_t		s_cred[OSD_CAP_LEN];	/* all-powerful credential    */
+	uint8_t		s_cred[OSD_CAP_LEN];	/* credential for the fscb    */
+
+	struct pnfs_osd_data_map data_map;	/* Default raid to use        */
+	unsigned	s_numdevs;		/* Num of devices in array    */
+	struct osd_dev	*s_ods[1];		/* Variable length, minimum 1 */
 };
 
 /*
@@ -79,6 +87,50 @@ struct exofs_i_info {
 	struct inode   vfs_inode;          /* normal in-memory inode          */
 };
 
+static inline osd_id exofs_oi_objno(struct exofs_i_info *oi)
+{
+	return oi->vfs_inode.i_ino + EXOFS_OBJ_OFF;
+}
+
+struct exofs_io_state;
+typedef void (*exofs_io_done_fn)(struct exofs_io_state *or, void *private);
+
+struct exofs_io_state {
+	struct kref		kref;
+
+	void			*private;
+	exofs_io_done_fn	done;
+
+	struct exofs_sb_info	*sbi;
+	struct osd_obj_id	obj;
+	u8			*cred;
+
+	/* Global read/write IO*/
+	loff_t			offset;
+	unsigned long		length;
+	void			*kern_buff;
+	struct bio		*bio;
+
+	/* Attributes */
+	unsigned		in_attr_len;
+	struct osd_attr 	*in_attr;
+	unsigned		out_attr_len;
+	struct osd_attr 	*out_attr;
+
+	/* Variable array of size numdevs */
+	unsigned numdevs;
+	struct exofs_per_dev_state {
+		struct osd_request *or;
+		struct bio *bio;
+	} per_dev[];
+};
+
+static inline unsigned exofs_io_state_size(unsigned numdevs)
+{
+	return sizeof(struct exofs_io_state) +
+		sizeof(struct exofs_per_dev_state) * numdevs;
+}
+
 /*
  * our inode flags
  */
@@ -130,6 +182,42 @@ static inline struct exofs_i_info *exofs_i(struct inode *inode)
 /*************************
  * function declarations *
  *************************/
+
+/* ios.c */
+void exofs_make_credential(u8 cred_a[OSD_CAP_LEN],
+			   const struct osd_obj_id *obj);
+int exofs_read_kern(struct osd_dev *od, u8 *cred, struct osd_obj_id *obj,
+		    u64 offset, void *p, unsigned length);
+
+int  exofs_get_io_state(struct exofs_sb_info *sbi, struct exofs_io_state** ios);
+void exofs_put_io_state(struct exofs_io_state *ios);
+
+int exofs_check_io(struct exofs_io_state *ios, u64 *resid);
+
+int exofs_sbi_create(struct exofs_io_state *ios);
+int exofs_sbi_remove(struct exofs_io_state *ios);
+int exofs_sbi_write(struct exofs_io_state *ios);
+int exofs_sbi_read(struct exofs_io_state *ios);
+
+int extract_attr_from_ios(struct exofs_io_state *ios, struct osd_attr *attr);
+
+int exofs_oi_truncate(struct exofs_i_info *oi, u64 new_len);
+static inline int exofs_oi_write(struct exofs_i_info *oi,
+				 struct exofs_io_state *ios)
+{
+	ios->obj.id = exofs_oi_objno(oi);
+	ios->cred = oi->i_cred;
+	return exofs_sbi_write(ios);
+}
+
+static inline int exofs_oi_read(struct exofs_i_info *oi,
+				struct exofs_io_state *ios)
+{
+	ios->obj.id = exofs_oi_objno(oi);
+	ios->cred = oi->i_cred;
+	return exofs_sbi_read(ios);
+}
+
 /* inode.c               */
 void exofs_truncate(struct inode *inode);
 int exofs_setattr(struct dentry *, struct iattr *);
@@ -169,6 +257,7 @@ extern const struct file_operations exofs_file_operations;
 
 /* inode.c           */
 extern const struct address_space_operations exofs_aops;
+extern const struct osd_attr g_attr_logical_length;
 
 /* namei.c           */
 extern const struct inode_operations exofs_dir_inode_operations;
diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c
index 6c10f7476699..2afbcebeda71 100644
--- a/fs/exofs/inode.c
+++ b/fs/exofs/inode.c
@@ -37,15 +37,18 @@
 
 #include "exofs.h"
 
-#ifdef CONFIG_EXOFS_DEBUG
-#  define EXOFS_DEBUG_OBJ_ISIZE 1
-#endif
+#define EXOFS_DBGMSG2(M...) do {} while (0)
+
+enum { BIO_MAX_PAGES_KMALLOC =
+		(PAGE_SIZE - sizeof(struct bio)) / sizeof(struct bio_vec),
+};
 
 struct page_collect {
 	struct exofs_sb_info *sbi;
 	struct request_queue *req_q;
 	struct inode *inode;
 	unsigned expected_pages;
+	struct exofs_io_state *ios;
 
 	struct bio *bio;
 	unsigned nr_pages;
@@ -54,22 +57,23 @@ struct page_collect {
 };
 
 static void _pcol_init(struct page_collect *pcol, unsigned expected_pages,
-		struct inode *inode)
+		       struct inode *inode)
 {
 	struct exofs_sb_info *sbi = inode->i_sb->s_fs_info;
 
 	pcol->sbi = sbi;
-	pcol->req_q = osd_request_queue(sbi->s_dev);
+	/* Create master bios on first Q, later on cloning, each clone will be
+	 * allocated on it's destination Q
+	 */
+	pcol->req_q = osd_request_queue(sbi->s_ods[0]);
 	pcol->inode = inode;
 	pcol->expected_pages = expected_pages;
 
+	pcol->ios = NULL;
 	pcol->bio = NULL;
 	pcol->nr_pages = 0;
 	pcol->length = 0;
 	pcol->pg_first = -1;
-
-	EXOFS_DBGMSG("_pcol_init ino=0x%lx expected_pages=%u\n", inode->i_ino,
-		     expected_pages);
 }
 
 static void _pcol_reset(struct page_collect *pcol)
@@ -80,35 +84,49 @@ static void _pcol_reset(struct page_collect *pcol)
 	pcol->nr_pages = 0;
 	pcol->length = 0;
 	pcol->pg_first = -1;
-	EXOFS_DBGMSG("_pcol_reset ino=0x%lx expected_pages=%u\n",
-		     pcol->inode->i_ino, pcol->expected_pages);
+	pcol->ios = NULL;
 
 	/* this is probably the end of the loop but in writes
 	 * it might not end here. don't be left with nothing
 	 */
 	if (!pcol->expected_pages)
-		pcol->expected_pages = 128;
+		pcol->expected_pages = BIO_MAX_PAGES_KMALLOC;
 }
 
 static int pcol_try_alloc(struct page_collect *pcol)
 {
-	int pages = min_t(unsigned, pcol->expected_pages, BIO_MAX_PAGES);
+	int pages = min_t(unsigned, pcol->expected_pages,
+			  BIO_MAX_PAGES_KMALLOC);
+
+	if (!pcol->ios) { /* First time allocate io_state */
+		int ret = exofs_get_io_state(pcol->sbi, &pcol->ios);
+
+		if (ret)
+			return ret;
+	}
 
 	for (; pages; pages >>= 1) {
-		pcol->bio = bio_alloc(GFP_KERNEL, pages);
+		pcol->bio = bio_kmalloc(GFP_KERNEL, pages);
 		if (likely(pcol->bio))
 			return 0;
 	}
 
-	EXOFS_ERR("Failed to kcalloc expected_pages=%u\n",
+	EXOFS_ERR("Failed to bio_kmalloc expected_pages=%u\n",
 		  pcol->expected_pages);
 	return -ENOMEM;
 }
 
 static void pcol_free(struct page_collect *pcol)
 {
-	bio_put(pcol->bio);
-	pcol->bio = NULL;
+	if (pcol->bio) {
+		bio_put(pcol->bio);
+		pcol->bio = NULL;
+	}
+
+	if (pcol->ios) {
+		exofs_put_io_state(pcol->ios);
+		pcol->ios = NULL;
+	}
 }
 
 static int pcol_add_page(struct page_collect *pcol, struct page *page,
@@ -161,22 +179,17 @@ static void update_write_page(struct page *page, int ret)
 /* Called at the end of reads, to optionally unlock pages and update their
  * status.
  */
-static int __readpages_done(struct osd_request *or, struct page_collect *pcol,
-			    bool do_unlock)
+static int __readpages_done(struct page_collect *pcol, bool do_unlock)
 {
 	struct bio_vec *bvec;
 	int i;
 	u64 resid;
 	u64 good_bytes;
 	u64 length = 0;
-	int ret = exofs_check_ok_resid(or, &resid, NULL);
-
-	osd_end_request(or);
+	int ret = exofs_check_io(pcol->ios, &resid);
 
 	if (likely(!ret))
 		good_bytes = pcol->length;
-	else if (!resid)
-		good_bytes = 0;
 	else
 		good_bytes = pcol->length - resid;
 
@@ -198,7 +211,7 @@ static int __readpages_done(struct osd_request *or, struct page_collect *pcol,
 		else
 			page_stat = ret;
 
-		EXOFS_DBGMSG("    readpages_done(0x%lx, 0x%lx) %s\n",
+		EXOFS_DBGMSG2("    readpages_done(0x%lx, 0x%lx) %s\n",
 			  inode->i_ino, page->index,
 			  page_stat ? "bad_bytes" : "good_bytes");
 
@@ -214,13 +227,13 @@ static int __readpages_done(struct osd_request *or, struct page_collect *pcol,
 }
 
 /* callback of async reads */
-static void readpages_done(struct osd_request *or, void *p)
+static void readpages_done(struct exofs_io_state *ios, void *p)
 {
 	struct page_collect *pcol = p;
 
-	__readpages_done(or, pcol, true);
+	__readpages_done(pcol, true);
 	atomic_dec(&pcol->sbi->s_curr_pending);
-	kfree(p);
+	kfree(pcol);
 }
 
 static void _unlock_pcol_pages(struct page_collect *pcol, int ret, int rw)
@@ -238,17 +251,13 @@ static void _unlock_pcol_pages(struct page_collect *pcol, int ret, int rw)
 
 		unlock_page(page);
 	}
-	pcol_free(pcol);
 }
 
 static int read_exec(struct page_collect *pcol, bool is_sync)
 {
 	struct exofs_i_info *oi = exofs_i(pcol->inode);
-	struct osd_obj_id obj = {pcol->sbi->s_pid,
-					pcol->inode->i_ino + EXOFS_OBJ_OFF};
-	struct osd_request *or = NULL;
+	struct exofs_io_state *ios = pcol->ios;
 	struct page_collect *pcol_copy = NULL;
-	loff_t i_start = pcol->pg_first << PAGE_CACHE_SHIFT;
 	int ret;
 
 	if (!pcol->bio)
@@ -257,17 +266,13 @@ static int read_exec(struct page_collect *pcol, bool is_sync)
 	/* see comment in _readpage() about sync reads */
 	WARN_ON(is_sync && (pcol->nr_pages != 1));
 
-	or = osd_start_request(pcol->sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		ret = -ENOMEM;
-		goto err;
-	}
-
-	osd_req_read(or, &obj, i_start, pcol->bio, pcol->length);
+	ios->bio = pcol->bio;
+	ios->length = pcol->length;
+	ios->offset = pcol->pg_first << PAGE_CACHE_SHIFT;
 
 	if (is_sync) {
-		exofs_sync_op(or, pcol->sbi->s_timeout, oi->i_cred);
-		return __readpages_done(or, pcol, false);
+		exofs_oi_read(oi, pcol->ios);
+		return __readpages_done(pcol, false);
 	}
 
 	pcol_copy = kmalloc(sizeof(*pcol_copy), GFP_KERNEL);
@@ -277,14 +282,16 @@ static int read_exec(struct page_collect *pcol, bool is_sync)
 	}
 
 	*pcol_copy = *pcol;
-	ret = exofs_async_op(or, readpages_done, pcol_copy, oi->i_cred);
+	ios->done = readpages_done;
+	ios->private = pcol_copy;
+	ret = exofs_oi_read(oi, ios);
 	if (unlikely(ret))
 		goto err;
 
 	atomic_inc(&pcol->sbi->s_curr_pending);
 
 	EXOFS_DBGMSG("read_exec obj=0x%llx start=0x%llx length=0x%lx\n",
-		  obj.id, _LLU(i_start), pcol->length);
+		  ios->obj.id, _LLU(ios->offset), pcol->length);
 
 	/* pages ownership was passed to pcol_copy */
 	_pcol_reset(pcol);
@@ -293,12 +300,10 @@ static int read_exec(struct page_collect *pcol, bool is_sync)
 err:
 	if (!is_sync)
 		_unlock_pcol_pages(pcol, ret, READ);
-	else /* Pages unlocked by caller in sync mode only free bio */
-		pcol_free(pcol);
+
+	pcol_free(pcol);
 
 	kfree(pcol_copy);
-	if (or)
-		osd_end_request(or);
 	return ret;
 }
 
@@ -370,12 +375,12 @@ try_again:
 	if (len != PAGE_CACHE_SIZE)
 		zero_user(page, len, PAGE_CACHE_SIZE - len);
 
-	EXOFS_DBGMSG("    readpage_strip(0x%lx, 0x%lx) len=0x%zx\n",
+	EXOFS_DBGMSG2("    readpage_strip(0x%lx, 0x%lx) len=0x%zx\n",
 		     inode->i_ino, page->index, len);
 
 	ret = pcol_add_page(pcol, page, len);
 	if (ret) {
-		EXOFS_DBGMSG("Failed pcol_add_page pages[i]=%p "
+		EXOFS_DBGMSG2("Failed pcol_add_page pages[i]=%p "
 			  "this_len=0x%zx nr_pages=%u length=0x%lx\n",
 			  page, len, pcol->nr_pages, pcol->length);
 
@@ -419,9 +424,8 @@ static int _readpage(struct page *page, bool is_sync)
 
 	_pcol_init(&pcol, 1, page->mapping->host);
 
-	/* readpage_strip might call read_exec(,async) inside at several places
-	 * but this is safe for is_async=0 since read_exec will not do anything
-	 * when we have a single page.
+	/* readpage_strip might call read_exec(,is_sync==false) at several
+	 * places but not if we have a single page.
 	 */
 	ret = readpage_strip(&pcol, page);
 	if (ret) {
@@ -440,8 +444,8 @@ static int exofs_readpage(struct file *file, struct page *page)
 	return _readpage(page, false);
 }
 
-/* Callback for osd_write. All writes are asynchronouse */
-static void writepages_done(struct osd_request *or, void *p)
+/* Callback for osd_write. All writes are asynchronous */
+static void writepages_done(struct exofs_io_state *ios, void *p)
 {
 	struct page_collect *pcol = p;
 	struct bio_vec *bvec;
@@ -449,16 +453,12 @@ static void writepages_done(struct osd_request *or, void *p)
 	u64 resid;
 	u64  good_bytes;
 	u64  length = 0;
+	int ret = exofs_check_io(ios, &resid);
 
-	int ret = exofs_check_ok_resid(or, NULL, &resid);
-
-	osd_end_request(or);
 	atomic_dec(&pcol->sbi->s_curr_pending);
 
 	if (likely(!ret))
 		good_bytes = pcol->length;
-	else if (!resid)
-		good_bytes = 0;
 	else
 		good_bytes = pcol->length - resid;
 
@@ -482,7 +482,7 @@ static void writepages_done(struct osd_request *or, void *p)
 
 		update_write_page(page, page_stat);
 		unlock_page(page);
-		EXOFS_DBGMSG("    writepages_done(0x%lx, 0x%lx) status=%d\n",
+		EXOFS_DBGMSG2("    writepages_done(0x%lx, 0x%lx) status=%d\n",
 			     inode->i_ino, page->index, page_stat);
 
 		length += bvec->bv_len;
@@ -496,23 +496,13 @@ static void writepages_done(struct osd_request *or, void *p)
 static int write_exec(struct page_collect *pcol)
 {
 	struct exofs_i_info *oi = exofs_i(pcol->inode);
-	struct osd_obj_id obj = {pcol->sbi->s_pid,
-					pcol->inode->i_ino + EXOFS_OBJ_OFF};
-	struct osd_request *or = NULL;
+	struct exofs_io_state *ios = pcol->ios;
 	struct page_collect *pcol_copy = NULL;
-	loff_t i_start = pcol->pg_first << PAGE_CACHE_SHIFT;
 	int ret;
 
 	if (!pcol->bio)
 		return 0;
 
-	or = osd_start_request(pcol->sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		EXOFS_ERR("write_exec: Faild to osd_start_request()\n");
-		ret = -ENOMEM;
-		goto err;
-	}
-
 	pcol_copy = kmalloc(sizeof(*pcol_copy), GFP_KERNEL);
 	if (!pcol_copy) {
 		EXOFS_ERR("write_exec: Faild to kmalloc(pcol)\n");
@@ -523,16 +513,22 @@ static int write_exec(struct page_collect *pcol)
 	*pcol_copy = *pcol;
 
 	pcol_copy->bio->bi_rw |= (1 << BIO_RW); /* FIXME: bio_set_dir() */
-	osd_req_write(or, &obj, i_start, pcol_copy->bio, pcol_copy->length);
-	ret = exofs_async_op(or, writepages_done, pcol_copy, oi->i_cred);
+
+	ios->bio = pcol_copy->bio;
+	ios->offset = pcol_copy->pg_first << PAGE_CACHE_SHIFT;
+	ios->length = pcol_copy->length;
+	ios->done = writepages_done;
+	ios->private = pcol_copy;
+
+	ret = exofs_oi_write(oi, ios);
 	if (unlikely(ret)) {
-		EXOFS_ERR("write_exec: exofs_async_op() Faild\n");
+		EXOFS_ERR("write_exec: exofs_oi_write() Faild\n");
 		goto err;
 	}
 
 	atomic_inc(&pcol->sbi->s_curr_pending);
 	EXOFS_DBGMSG("write_exec(0x%lx, 0x%llx) start=0x%llx length=0x%lx\n",
-		  pcol->inode->i_ino, pcol->pg_first, _LLU(i_start),
+		  pcol->inode->i_ino, pcol->pg_first, _LLU(ios->offset),
 		  pcol->length);
 	/* pages ownership was passed to pcol_copy */
 	_pcol_reset(pcol);
@@ -540,9 +536,9 @@ static int write_exec(struct page_collect *pcol)
 
 err:
 	_unlock_pcol_pages(pcol, ret, WRITE);
+	pcol_free(pcol);
 	kfree(pcol_copy);
-	if (or)
-		osd_end_request(or);
+
 	return ret;
 }
 
@@ -586,6 +582,9 @@ static int writepage_strip(struct page *page,
 			if (PageError(page))
 				ClearPageError(page);
 			unlock_page(page);
+			EXOFS_DBGMSG("writepage_strip(0x%lx, 0x%lx) "
+				     "outside the limits\n",
+				     inode->i_ino, page->index);
 			return 0;
 		}
 	}
@@ -600,6 +599,9 @@ try_again:
 		ret = write_exec(pcol);
 		if (unlikely(ret))
 			goto fail;
+
+		EXOFS_DBGMSG("writepage_strip(0x%lx, 0x%lx) Discontinuity\n",
+			     inode->i_ino, page->index);
 		goto try_again;
 	}
 
@@ -609,7 +611,7 @@ try_again:
 			goto fail;
 	}
 
-	EXOFS_DBGMSG("    writepage_strip(0x%lx, 0x%lx) len=0x%zx\n",
+	EXOFS_DBGMSG2("    writepage_strip(0x%lx, 0x%lx) len=0x%zx\n",
 		     inode->i_ino, page->index, len);
 
 	ret = pcol_add_page(pcol, page, len);
@@ -634,6 +636,8 @@ try_again:
 	return 0;
 
 fail:
+	EXOFS_DBGMSG("Error: writepage_strip(0x%lx, 0x%lx)=>%d\n",
+		     inode->i_ino, page->index, ret);
 	set_bit(AS_EIO, &page->mapping->flags);
 	unlock_page(page);
 	return ret;
@@ -652,14 +656,17 @@ static int exofs_writepages(struct address_space *mapping,
 			wbc->range_end >> PAGE_CACHE_SHIFT;
 
 	if (start || end)
-		expected_pages = min(end - start + 1, 32L);
+		expected_pages = end - start + 1;
 	else
 		expected_pages = mapping->nrpages;
 
-	EXOFS_DBGMSG("inode(0x%lx) wbc->start=0x%llx wbc->end=0x%llx"
-		     " m->nrpages=%lu start=0x%lx end=0x%lx\n",
+	if (expected_pages < 32L)
+		expected_pages = 32L;
+
+	EXOFS_DBGMSG("inode(0x%lx) wbc->start=0x%llx wbc->end=0x%llx "
+		     "nrpages=%lu start=0x%lx end=0x%lx expected_pages=%ld\n",
 		     mapping->host->i_ino, wbc->range_start, wbc->range_end,
-		     mapping->nrpages, start, end);
+		     mapping->nrpages, start, end, expected_pages);
 
 	_pcol_init(&pcol, expected_pages, mapping->host);
 
@@ -731,13 +738,28 @@ static int exofs_write_begin_export(struct file *file,
 					fsdata);
 }
 
+static int exofs_write_end(struct file *file, struct address_space *mapping,
+			loff_t pos, unsigned len, unsigned copied,
+			struct page *page, void *fsdata)
+{
+	struct inode *inode = mapping->host;
+	/* According to comment in simple_write_end i_mutex is held */
+	loff_t i_size = inode->i_size;
+	int ret;
+
+	ret = simple_write_end(file, mapping,pos, len, copied, page, fsdata);
+	if (i_size != inode->i_size)
+		mark_inode_dirty(inode);
+	return ret;
+}
+
 const struct address_space_operations exofs_aops = {
 	.readpage	= exofs_readpage,
 	.readpages	= exofs_readpages,
 	.writepage	= exofs_writepage,
 	.writepages	= exofs_writepages,
 	.write_begin	= exofs_write_begin_export,
-	.write_end	= simple_write_end,
+	.write_end	= exofs_write_end,
 };
 
 /******************************************************************************
@@ -771,19 +793,28 @@ static int exofs_get_block(struct inode *inode, sector_t iblock,
 const struct osd_attr g_attr_logical_length = ATTR_DEF(
 	OSD_APAGE_OBJECT_INFORMATION, OSD_ATTR_OI_LOGICAL_LENGTH, 8);
 
+static int _do_truncate(struct inode *inode)
+{
+	struct exofs_i_info *oi = exofs_i(inode);
+	loff_t isize = i_size_read(inode);
+	int ret;
+
+	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+
+	nobh_truncate_page(inode->i_mapping, isize, exofs_get_block);
+
+	ret = exofs_oi_truncate(oi, (u64)isize);
+	EXOFS_DBGMSG("(0x%lx) size=0x%llx\n", inode->i_ino, isize);
+	return ret;
+}
+
 /*
  * Truncate a file to the specified size - all we have to do is set the size
  * attribute.  We make sure the object exists first.
  */
 void exofs_truncate(struct inode *inode)
 {
-	struct exofs_sb_info *sbi = inode->i_sb->s_fs_info;
 	struct exofs_i_info *oi = exofs_i(inode);
-	struct osd_obj_id obj = {sbi->s_pid, inode->i_ino + EXOFS_OBJ_OFF};
-	struct osd_request *or;
-	struct osd_attr attr;
-	loff_t isize = i_size_read(inode);
-	__be64 newsize;
 	int ret;
 
 	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode)
@@ -793,22 +824,6 @@ void exofs_truncate(struct inode *inode)
 		return;
 	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
 		return;
-	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-
-	nobh_truncate_page(inode->i_mapping, isize, exofs_get_block);
-
-	or = osd_start_request(sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		EXOFS_ERR("ERROR: exofs_truncate: osd_start_request failed\n");
-		goto fail;
-	}
-
-	osd_req_set_attributes(or, &obj);
-
-	newsize = cpu_to_be64((u64)isize);
-	attr = g_attr_logical_length;
-	attr.val_ptr = &newsize;
-	osd_req_add_set_attr_list(or, &attr, 1);
 
 	/* if we are about to truncate an object, and it hasn't been
 	 * created yet, wait
@@ -816,8 +831,7 @@ void exofs_truncate(struct inode *inode)
 	if (unlikely(wait_obj_created(oi)))
 		goto fail;
 
-	ret = exofs_sync_op(or, sbi->s_timeout, oi->i_cred);
-	osd_end_request(or);
+	ret = _do_truncate(inode);
 	if (ret)
 		goto fail;
 
@@ -847,65 +861,62 @@ int exofs_setattr(struct dentry *dentry, struct iattr *iattr)
 
 /*
  * Read an inode from the OSD, and return it as is.  We also return the size
- * attribute in the 'sanity' argument if we got compiled with debugging turned
- * on.
+ * attribute in the 'obj_size' argument.
  */
 static int exofs_get_inode(struct super_block *sb, struct exofs_i_info *oi,
-		    struct exofs_fcb *inode, uint64_t *sanity)
+		    struct exofs_fcb *inode, uint64_t *obj_size)
 {
 	struct exofs_sb_info *sbi = sb->s_fs_info;
-	struct osd_request *or;
-	struct osd_attr attr;
-	struct osd_obj_id obj = {sbi->s_pid,
-				 oi->vfs_inode.i_ino + EXOFS_OBJ_OFF};
+	struct osd_attr attrs[2];
+	struct exofs_io_state *ios;
 	int ret;
 
-	exofs_make_credential(oi->i_cred, &obj);
-
-	or = osd_start_request(sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		EXOFS_ERR("exofs_get_inode: osd_start_request failed.\n");
-		return -ENOMEM;
+	*obj_size = ~0;
+	ret = exofs_get_io_state(sbi, &ios);
+	if (unlikely(ret)) {
+		EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__);
+		return ret;
 	}
-	osd_req_get_attributes(or, &obj);
 
-	/* we need the inode attribute */
-	osd_req_add_get_attr_list(or, &g_attr_inode_data, 1);
+	ios->obj.id = exofs_oi_objno(oi);
+	exofs_make_credential(oi->i_cred, &ios->obj);
+	ios->cred = oi->i_cred;
 
-#ifdef EXOFS_DEBUG_OBJ_ISIZE
-	/* we get the size attributes to do a sanity check */
-	osd_req_add_get_attr_list(or, &g_attr_logical_length, 1);
-#endif
+	attrs[0] = g_attr_inode_data;
+	attrs[1] = g_attr_logical_length;
+	ios->in_attr = attrs;
+	ios->in_attr_len = ARRAY_SIZE(attrs);
 
-	ret = exofs_sync_op(or, sbi->s_timeout, oi->i_cred);
+	ret = exofs_sbi_read(ios);
 	if (ret)
 		goto out;
 
-	attr = g_attr_inode_data;
-	ret = extract_attr_from_req(or, &attr);
+	ret = extract_attr_from_ios(ios, &attrs[0]);
 	if (ret) {
-		EXOFS_ERR("exofs_get_inode: extract_attr_from_req failed\n");
+		EXOFS_ERR("%s: extract_attr of inode_data failed\n", __func__);
 		goto out;
 	}
+	WARN_ON(attrs[0].len != EXOFS_INO_ATTR_SIZE);
+	memcpy(inode, attrs[0].val_ptr, EXOFS_INO_ATTR_SIZE);
 
-	WARN_ON(attr.len != EXOFS_INO_ATTR_SIZE);
-	memcpy(inode, attr.val_ptr, EXOFS_INO_ATTR_SIZE);
-
-#ifdef EXOFS_DEBUG_OBJ_ISIZE
-	attr = g_attr_logical_length;
-	ret = extract_attr_from_req(or, &attr);
+	ret = extract_attr_from_ios(ios, &attrs[1]);
 	if (ret) {
-		EXOFS_ERR("ERROR: extract attr from or failed\n");
+		EXOFS_ERR("%s: extract_attr of logical_length failed\n",
+			  __func__);
 		goto out;
 	}
-	*sanity = get_unaligned_be64(attr.val_ptr);
-#endif
+	*obj_size = get_unaligned_be64(attrs[1].val_ptr);
 
 out:
-	osd_end_request(or);
+	exofs_put_io_state(ios);
 	return ret;
 }
 
+static void __oi_init(struct exofs_i_info *oi)
+{
+	init_waitqueue_head(&oi->i_wq);
+	oi->i_flags = 0;
+}
 /*
  * Fill in an inode read from the OSD and set it up for use
  */
@@ -914,7 +925,7 @@ struct inode *exofs_iget(struct super_block *sb, unsigned long ino)
 	struct exofs_i_info *oi;
 	struct exofs_fcb fcb;
 	struct inode *inode;
-	uint64_t uninitialized_var(sanity);
+	uint64_t obj_size;
 	int ret;
 
 	inode = iget_locked(sb, ino);
@@ -923,13 +934,13 @@ struct inode *exofs_iget(struct super_block *sb, unsigned long ino)
 	if (!(inode->i_state & I_NEW))
 		return inode;
 	oi = exofs_i(inode);
+	__oi_init(oi);
 
 	/* read the inode from the osd */
-	ret = exofs_get_inode(sb, oi, &fcb, &sanity);
+	ret = exofs_get_inode(sb, oi, &fcb, &obj_size);
 	if (ret)
 		goto bad_inode;
 
-	init_waitqueue_head(&oi->i_wq);
 	set_obj_created(oi);
 
 	/* copy stuff from on-disk struct to in-memory struct */
@@ -947,14 +958,12 @@ struct inode *exofs_iget(struct super_block *sb, unsigned long ino)
 	inode->i_blkbits = EXOFS_BLKSHIFT;
 	inode->i_generation = le32_to_cpu(fcb.i_generation);
 
-#ifdef EXOFS_DEBUG_OBJ_ISIZE
-	if ((inode->i_size != sanity) &&
+	if ((inode->i_size != obj_size) &&
 		(!exofs_inode_is_fast_symlink(inode))) {
-		EXOFS_ERR("WARNING: Size of object from inode and "
-			  "attributes differ (%lld != %llu)\n",
-			  inode->i_size, _LLU(sanity));
+		EXOFS_ERR("WARNING: Size of inode=%llu != object=%llu\n",
+			  inode->i_size, _LLU(obj_size));
+		/* FIXME: call exofs_inode_recovery() */
 	}
-#endif
 
 	oi->i_dir_start_lookup = 0;
 
@@ -1020,23 +1029,30 @@ int __exofs_wait_obj_created(struct exofs_i_info *oi)
  * set the obj_created flag so that other methods know that the object exists on
  * the OSD.
  */
-static void create_done(struct osd_request *or, void *p)
+static void create_done(struct exofs_io_state *ios, void *p)
 {
 	struct inode *inode = p;
 	struct exofs_i_info *oi = exofs_i(inode);
 	struct exofs_sb_info *sbi = inode->i_sb->s_fs_info;
 	int ret;
 
-	ret = exofs_check_ok(or);
-	osd_end_request(or);
+	ret = exofs_check_io(ios, NULL);
+	exofs_put_io_state(ios);
+
 	atomic_dec(&sbi->s_curr_pending);
 
 	if (unlikely(ret)) {
 		EXOFS_ERR("object=0x%llx creation faild in pid=0x%llx",
-			  _LLU(sbi->s_pid), _LLU(inode->i_ino + EXOFS_OBJ_OFF));
-		make_bad_inode(inode);
-	} else
-		set_obj_created(oi);
+			  _LLU(exofs_oi_objno(oi)), _LLU(sbi->s_pid));
+		/*TODO: When FS is corrupted creation can fail, object already
+		 * exist. Get rid of this asynchronous creation, if exist
+		 * increment the obj counter and try the next object. Until we
+		 * succeed. All these dangling objects will be made into lost
+		 * files by chkfs.exofs
+		 */
+	}
+
+	set_obj_created(oi);
 
 	atomic_dec(&inode->i_count);
 	wake_up(&oi->i_wq);
@@ -1051,8 +1067,7 @@ struct inode *exofs_new_inode(struct inode *dir, int mode)
 	struct inode *inode;
 	struct exofs_i_info *oi;
 	struct exofs_sb_info *sbi;
-	struct osd_request *or;
-	struct osd_obj_id obj;
+	struct exofs_io_state *ios;
 	int ret;
 
 	sb = dir->i_sb;
@@ -1061,8 +1076,8 @@ struct inode *exofs_new_inode(struct inode *dir, int mode)
 		return ERR_PTR(-ENOMEM);
 
 	oi = exofs_i(inode);
+	__oi_init(oi);
 
-	init_waitqueue_head(&oi->i_wq);
 	set_obj_2bcreated(oi);
 
 	sbi = sb->s_fs_info;
@@ -1089,28 +1104,28 @@ struct inode *exofs_new_inode(struct inode *dir, int mode)
 
 	mark_inode_dirty(inode);
 
-	obj.partition = sbi->s_pid;
-	obj.id = inode->i_ino + EXOFS_OBJ_OFF;
-	exofs_make_credential(oi->i_cred, &obj);
-
-	or = osd_start_request(sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		EXOFS_ERR("exofs_new_inode: osd_start_request failed\n");
-		return ERR_PTR(-ENOMEM);
+	ret = exofs_get_io_state(sbi, &ios);
+	if (unlikely(ret)) {
+		EXOFS_ERR("exofs_new_inode: exofs_get_io_state failed\n");
+		return ERR_PTR(ret);
 	}
 
-	osd_req_create_object(or, &obj);
+	ios->obj.id = exofs_oi_objno(oi);
+	exofs_make_credential(oi->i_cred, &ios->obj);
 
 	/* increment the refcount so that the inode will still be around when we
 	 * reach the callback
 	 */
 	atomic_inc(&inode->i_count);
 
-	ret = exofs_async_op(or, create_done, inode, oi->i_cred);
+	ios->done = create_done;
+	ios->private = inode;
+	ios->cred = oi->i_cred;
+	ret = exofs_sbi_create(ios);
 	if (ret) {
 		atomic_dec(&inode->i_count);
-		osd_end_request(or);
-		return ERR_PTR(-EIO);
+		exofs_put_io_state(ios);
+		return ERR_PTR(ret);
 	}
 	atomic_inc(&sbi->s_curr_pending);
 
@@ -1128,11 +1143,11 @@ struct updatei_args {
 /*
  * Callback function from exofs_update_inode().
  */
-static void updatei_done(struct osd_request *or, void *p)
+static void updatei_done(struct exofs_io_state *ios, void *p)
 {
 	struct updatei_args *args = p;
 
-	osd_end_request(or);
+	exofs_put_io_state(ios);
 
 	atomic_dec(&args->sbi->s_curr_pending);
 
@@ -1148,8 +1163,7 @@ static int exofs_update_inode(struct inode *inode, int do_sync)
 	struct exofs_i_info *oi = exofs_i(inode);
 	struct super_block *sb = inode->i_sb;
 	struct exofs_sb_info *sbi = sb->s_fs_info;
-	struct osd_obj_id obj = {sbi->s_pid, inode->i_ino + EXOFS_OBJ_OFF};
-	struct osd_request *or;
+	struct exofs_io_state *ios;
 	struct osd_attr attr;
 	struct exofs_fcb *fcb;
 	struct updatei_args *args;
@@ -1186,18 +1200,16 @@ static int exofs_update_inode(struct inode *inode, int do_sync)
 	} else
 		memcpy(fcb->i_data, oi->i_data, sizeof(fcb->i_data));
 
-	or = osd_start_request(sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		EXOFS_ERR("exofs_update_inode: osd_start_request failed.\n");
-		ret = -ENOMEM;
+	ret = exofs_get_io_state(sbi, &ios);
+	if (unlikely(ret)) {
+		EXOFS_ERR("%s: exofs_get_io_state failed.\n", __func__);
 		goto free_args;
 	}
 
-	osd_req_set_attributes(or, &obj);
-
 	attr = g_attr_inode_data;
 	attr.val_ptr = fcb;
-	osd_req_add_set_attr_list(or, &attr, 1);
+	ios->out_attr_len = 1;
+	ios->out_attr = &attr;
 
 	if (!obj_created(oi)) {
 		EXOFS_DBGMSG("!obj_created\n");
@@ -1206,22 +1218,19 @@ static int exofs_update_inode(struct inode *inode, int do_sync)
 		EXOFS_DBGMSG("wait_event done\n");
 	}
 
-	if (do_sync) {
-		ret = exofs_sync_op(or, sbi->s_timeout, oi->i_cred);
-		osd_end_request(or);
-		goto free_args;
-	} else {
+	if (!do_sync) {
 		args->sbi = sbi;
+		ios->done = updatei_done;
+		ios->private = args;
+	}
 
-		ret = exofs_async_op(or, updatei_done, args, oi->i_cred);
-		if (ret) {
-			osd_end_request(or);
-			goto free_args;
-		}
+	ret = exofs_oi_write(oi, ios);
+	if (!do_sync && !ret) {
 		atomic_inc(&sbi->s_curr_pending);
 		goto out; /* deallocation in updatei_done */
 	}
 
+	exofs_put_io_state(ios);
 free_args:
 	kfree(args);
 out:
@@ -1238,11 +1247,12 @@ int exofs_write_inode(struct inode *inode, int wait)
  * Callback function from exofs_delete_inode() - don't have much cleaning up to
  * do.
  */
-static void delete_done(struct osd_request *or, void *p)
+static void delete_done(struct exofs_io_state *ios, void *p)
 {
-	struct exofs_sb_info *sbi;
-	osd_end_request(or);
-	sbi = p;
+	struct exofs_sb_info *sbi = p;
+
+	exofs_put_io_state(ios);
+
 	atomic_dec(&sbi->s_curr_pending);
 }
 
@@ -1256,8 +1266,7 @@ void exofs_delete_inode(struct inode *inode)
 	struct exofs_i_info *oi = exofs_i(inode);
 	struct super_block *sb = inode->i_sb;
 	struct exofs_sb_info *sbi = sb->s_fs_info;
-	struct osd_obj_id obj = {sbi->s_pid, inode->i_ino + EXOFS_OBJ_OFF};
-	struct osd_request *or;
+	struct exofs_io_state *ios;
 	int ret;
 
 	truncate_inode_pages(&inode->i_data, 0);
@@ -1274,25 +1283,26 @@ void exofs_delete_inode(struct inode *inode)
 
 	clear_inode(inode);
 
-	or = osd_start_request(sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		EXOFS_ERR("exofs_delete_inode: osd_start_request failed\n");
+	ret = exofs_get_io_state(sbi, &ios);
+	if (unlikely(ret)) {
+		EXOFS_ERR("%s: exofs_get_io_state failed\n", __func__);
 		return;
 	}
 
-	osd_req_remove_object(or, &obj);
-
 	/* if we are deleting an obj that hasn't been created yet, wait */
 	if (!obj_created(oi)) {
 		BUG_ON(!obj_2bcreated(oi));
 		wait_event(oi->i_wq, obj_created(oi));
 	}
 
-	ret = exofs_async_op(or, delete_done, sbi, oi->i_cred);
+	ios->obj.id = exofs_oi_objno(oi);
+	ios->done = delete_done;
+	ios->private = sbi;
+	ios->cred = oi->i_cred;
+	ret = exofs_sbi_remove(ios);
 	if (ret) {
-		EXOFS_ERR(
-		       "ERROR: @exofs_delete_inode exofs_async_op failed\n");
-		osd_end_request(or);
+		EXOFS_ERR("%s: exofs_sbi_remove failed\n", __func__);
+		exofs_put_io_state(ios);
 		return;
 	}
 	atomic_inc(&sbi->s_curr_pending);
diff --git a/fs/exofs/ios.c b/fs/exofs/ios.c
new file mode 100644
index 000000000000..5bad01fa1f9f
--- /dev/null
+++ b/fs/exofs/ios.c
@@ -0,0 +1,421 @@
+/*
+ * Copyright (C) 2005, 2006
+ * Avishay Traeger (avishay@gmail.com)
+ * Copyright (C) 2008, 2009
+ * Boaz Harrosh <bharrosh@panasas.com>
+ *
+ * This file is part of exofs.
+ *
+ * exofs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.  Since it is based on ext2, and the only
+ * valid version of GPL for the Linux kernel is version 2, the only valid
+ * version of GPL for exofs is version 2.
+ *
+ * exofs is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with exofs; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#include <scsi/scsi_device.h>
+
+#include "exofs.h"
+
+void exofs_make_credential(u8 cred_a[OSD_CAP_LEN], const struct osd_obj_id *obj)
+{
+	osd_sec_init_nosec_doall_caps(cred_a, obj, false, true);
+}
+
+int exofs_read_kern(struct osd_dev *od, u8 *cred, struct osd_obj_id *obj,
+		    u64 offset, void *p, unsigned length)
+{
+	struct osd_request *or = osd_start_request(od, GFP_KERNEL);
+/*	struct osd_sense_info osi = {.key = 0};*/
+	int ret;
+
+	if (unlikely(!or)) {
+		EXOFS_DBGMSG("%s: osd_start_request failed.\n", __func__);
+		return -ENOMEM;
+	}
+	ret = osd_req_read_kern(or, obj, offset, p, length);
+	if (unlikely(ret)) {
+		EXOFS_DBGMSG("%s: osd_req_read_kern failed.\n", __func__);
+		goto out;
+	}
+
+	ret = osd_finalize_request(or, 0, cred, NULL);
+	if (unlikely(ret)) {
+		EXOFS_DBGMSG("Faild to osd_finalize_request() => %d\n", ret);
+		goto out;
+	}
+
+	ret = osd_execute_request(or);
+	if (unlikely(ret))
+		EXOFS_DBGMSG("osd_execute_request() => %d\n", ret);
+	/* osd_req_decode_sense(or, ret); */
+
+out:
+	osd_end_request(or);
+	return ret;
+}
+
+int exofs_get_io_state(struct exofs_sb_info *sbi, struct exofs_io_state** pios)
+{
+	struct exofs_io_state *ios;
+
+	/*TODO: Maybe use kmem_cach per sbi of size
+	 * exofs_io_state_size(sbi->s_numdevs)
+	 */
+	ios = kzalloc(exofs_io_state_size(sbi->s_numdevs), GFP_KERNEL);
+	if (unlikely(!ios)) {
+		*pios = NULL;
+		return -ENOMEM;
+	}
+
+	ios->sbi = sbi;
+	ios->obj.partition = sbi->s_pid;
+	*pios = ios;
+	return 0;
+}
+
+void exofs_put_io_state(struct exofs_io_state *ios)
+{
+	if (ios) {
+		unsigned i;
+
+		for (i = 0; i < ios->numdevs; i++) {
+			struct exofs_per_dev_state *per_dev = &ios->per_dev[i];
+
+			if (per_dev->or)
+				osd_end_request(per_dev->or);
+			if (per_dev->bio)
+				bio_put(per_dev->bio);
+		}
+
+		kfree(ios);
+	}
+}
+
+static void _sync_done(struct exofs_io_state *ios, void *p)
+{
+	struct completion *waiting = p;
+
+	complete(waiting);
+}
+
+static void _last_io(struct kref *kref)
+{
+	struct exofs_io_state *ios = container_of(
+					kref, struct exofs_io_state, kref);
+
+	ios->done(ios, ios->private);
+}
+
+static void _done_io(struct osd_request *or, void *p)
+{
+	struct exofs_io_state *ios = p;
+
+	kref_put(&ios->kref, _last_io);
+}
+
+static int exofs_io_execute(struct exofs_io_state *ios)
+{
+	DECLARE_COMPLETION_ONSTACK(wait);
+	bool sync = (ios->done == NULL);
+	int i, ret;
+
+	if (sync) {
+		ios->done = _sync_done;
+		ios->private = &wait;
+	}
+
+	for (i = 0; i < ios->numdevs; i++) {
+		struct osd_request *or = ios->per_dev[i].or;
+		if (unlikely(!or))
+			continue;
+
+		ret = osd_finalize_request(or, 0, ios->cred, NULL);
+		if (unlikely(ret)) {
+			EXOFS_DBGMSG("Faild to osd_finalize_request() => %d\n",
+				     ret);
+			return ret;
+		}
+	}
+
+	kref_init(&ios->kref);
+
+	for (i = 0; i < ios->numdevs; i++) {
+		struct osd_request *or = ios->per_dev[i].or;
+		if (unlikely(!or))
+			continue;
+
+		kref_get(&ios->kref);
+		osd_execute_request_async(or, _done_io, ios);
+	}
+
+	kref_put(&ios->kref, _last_io);
+	ret = 0;
+
+	if (sync) {
+		wait_for_completion(&wait);
+		ret = exofs_check_io(ios, NULL);
+	}
+	return ret;
+}
+
+int exofs_check_io(struct exofs_io_state *ios, u64 *resid)
+{
+	enum osd_err_priority acumulated_osd_err = 0;
+	int acumulated_lin_err = 0;
+	int i;
+
+	for (i = 0; i < ios->numdevs; i++) {
+		struct osd_sense_info osi;
+		int ret = osd_req_decode_sense(ios->per_dev[i].or, &osi);
+
+		if (likely(!ret))
+			continue;
+
+		if (unlikely(ret == -EFAULT)) {
+			EXOFS_DBGMSG("%s: EFAULT Need page clear\n", __func__);
+			/*FIXME: All the pages in this device range should:
+			 *	clear_highpage(page);
+			 */
+		}
+
+		if (osi.osd_err_pri >= acumulated_osd_err) {
+			acumulated_osd_err = osi.osd_err_pri;
+			acumulated_lin_err = ret;
+		}
+	}
+
+	/* TODO: raid specific residual calculations */
+	if (resid) {
+		if (likely(!acumulated_lin_err))
+			*resid = 0;
+		else
+			*resid = ios->length;
+	}
+
+	return acumulated_lin_err;
+}
+
+int exofs_sbi_create(struct exofs_io_state *ios)
+{
+	int i, ret;
+
+	for (i = 0; i < ios->sbi->s_numdevs; i++) {
+		struct osd_request *or;
+
+		or = osd_start_request(ios->sbi->s_ods[i], GFP_KERNEL);
+		if (unlikely(!or)) {
+			EXOFS_ERR("%s: osd_start_request failed\n", __func__);
+			ret = -ENOMEM;
+			goto out;
+		}
+		ios->per_dev[i].or = or;
+		ios->numdevs++;
+
+		osd_req_create_object(or, &ios->obj);
+	}
+	ret = exofs_io_execute(ios);
+
+out:
+	return ret;
+}
+
+int exofs_sbi_remove(struct exofs_io_state *ios)
+{
+	int i, ret;
+
+	for (i = 0; i < ios->sbi->s_numdevs; i++) {
+		struct osd_request *or;
+
+		or = osd_start_request(ios->sbi->s_ods[i], GFP_KERNEL);
+		if (unlikely(!or)) {
+			EXOFS_ERR("%s: osd_start_request failed\n", __func__);
+			ret = -ENOMEM;
+			goto out;
+		}
+		ios->per_dev[i].or = or;
+		ios->numdevs++;
+
+		osd_req_remove_object(or, &ios->obj);
+	}
+	ret = exofs_io_execute(ios);
+
+out:
+	return ret;
+}
+
+int exofs_sbi_write(struct exofs_io_state *ios)
+{
+	int i, ret;
+
+	for (i = 0; i < ios->sbi->s_numdevs; i++) {
+		struct osd_request *or;
+
+		or = osd_start_request(ios->sbi->s_ods[i], GFP_KERNEL);
+		if (unlikely(!or)) {
+			EXOFS_ERR("%s: osd_start_request failed\n", __func__);
+			ret = -ENOMEM;
+			goto out;
+		}
+		ios->per_dev[i].or = or;
+		ios->numdevs++;
+
+		if (ios->bio) {
+			struct bio *bio;
+
+			if (i != 0) {
+				bio = bio_kmalloc(GFP_KERNEL,
+						  ios->bio->bi_max_vecs);
+				if (unlikely(!bio)) {
+					ret = -ENOMEM;
+					goto out;
+				}
+
+				__bio_clone(bio, ios->bio);
+				bio->bi_bdev = NULL;
+				bio->bi_next = NULL;
+				ios->per_dev[i].bio =  bio;
+			} else {
+				bio = ios->bio;
+			}
+
+			osd_req_write(or, &ios->obj, ios->offset, bio,
+				      ios->length);
+/*			EXOFS_DBGMSG("write sync=%d\n", sync);*/
+		} else if (ios->kern_buff) {
+			osd_req_write_kern(or, &ios->obj, ios->offset,
+					   ios->kern_buff, ios->length);
+/*			EXOFS_DBGMSG("write_kern sync=%d\n", sync);*/
+		} else {
+			osd_req_set_attributes(or, &ios->obj);
+/*			EXOFS_DBGMSG("set_attributes sync=%d\n", sync);*/
+		}
+
+		if (ios->out_attr)
+			osd_req_add_set_attr_list(or, ios->out_attr,
+						  ios->out_attr_len);
+
+		if (ios->in_attr)
+			osd_req_add_get_attr_list(or, ios->in_attr,
+						  ios->in_attr_len);
+	}
+	ret = exofs_io_execute(ios);
+
+out:
+	return ret;
+}
+
+int exofs_sbi_read(struct exofs_io_state *ios)
+{
+	int i, ret;
+
+	for (i = 0; i < 1; i++) {
+		struct osd_request *or;
+		unsigned first_dev = (unsigned)ios->obj.id;
+
+		first_dev %= ios->sbi->s_numdevs;
+		or = osd_start_request(ios->sbi->s_ods[first_dev], GFP_KERNEL);
+		if (unlikely(!or)) {
+			EXOFS_ERR("%s: osd_start_request failed\n", __func__);
+			ret = -ENOMEM;
+			goto out;
+		}
+		ios->per_dev[i].or = or;
+		ios->numdevs++;
+
+		if (ios->bio) {
+			osd_req_read(or, &ios->obj, ios->offset, ios->bio,
+				     ios->length);
+/*			EXOFS_DBGMSG("read sync=%d\n", sync);*/
+		} else if (ios->kern_buff) {
+			osd_req_read_kern(or, &ios->obj, ios->offset,
+					   ios->kern_buff, ios->length);
+/*			EXOFS_DBGMSG("read_kern sync=%d\n", sync);*/
+		} else {
+			osd_req_get_attributes(or, &ios->obj);
+/*			EXOFS_DBGMSG("get_attributes sync=%d\n", sync);*/
+		}
+
+		if (ios->out_attr)
+			osd_req_add_set_attr_list(or, ios->out_attr,
+						  ios->out_attr_len);
+
+		if (ios->in_attr)
+			osd_req_add_get_attr_list(or, ios->in_attr,
+						  ios->in_attr_len);
+	}
+	ret = exofs_io_execute(ios);
+
+out:
+	return ret;
+}
+
+int extract_attr_from_ios(struct exofs_io_state *ios, struct osd_attr *attr)
+{
+	struct osd_attr cur_attr = {.attr_page = 0}; /* start with zeros */
+	void *iter = NULL;
+	int nelem;
+
+	do {
+		nelem = 1;
+		osd_req_decode_get_attr_list(ios->per_dev[0].or,
+					     &cur_attr, &nelem, &iter);
+		if ((cur_attr.attr_page == attr->attr_page) &&
+		    (cur_attr.attr_id == attr->attr_id)) {
+			attr->len = cur_attr.len;
+			attr->val_ptr = cur_attr.val_ptr;
+			return 0;
+		}
+	} while (iter);
+
+	return -EIO;
+}
+
+int exofs_oi_truncate(struct exofs_i_info *oi, u64 size)
+{
+	struct exofs_sb_info *sbi = oi->vfs_inode.i_sb->s_fs_info;
+	struct exofs_io_state *ios;
+	struct osd_attr attr;
+	__be64 newsize;
+	int i, ret;
+
+	if (exofs_get_io_state(sbi, &ios))
+		return -ENOMEM;
+
+	ios->obj.id = exofs_oi_objno(oi);
+	ios->cred = oi->i_cred;
+
+	newsize = cpu_to_be64(size);
+	attr = g_attr_logical_length;
+	attr.val_ptr = &newsize;
+
+	for (i = 0; i < sbi->s_numdevs; i++) {
+		struct osd_request *or;
+
+		or = osd_start_request(sbi->s_ods[i], GFP_KERNEL);
+		if (unlikely(!or)) {
+			EXOFS_ERR("%s: osd_start_request failed\n", __func__);
+			ret = -ENOMEM;
+			goto out;
+		}
+		ios->per_dev[i].or = or;
+		ios->numdevs++;
+
+		osd_req_set_attributes(or, &ios->obj);
+		osd_req_add_set_attr_list(or, &attr, 1);
+	}
+	ret = exofs_io_execute(ios);
+
+out:
+	exofs_put_io_state(ios);
+	return ret;
+}
diff --git a/fs/exofs/osd.c b/fs/exofs/osd.c
deleted file mode 100644
index 4372542df284..000000000000
--- a/fs/exofs/osd.c
+++ /dev/null
@@ -1,125 +0,0 @@
-/*
- * Copyright (C) 2005, 2006
- * Avishay Traeger (avishay@gmail.com)
- * Copyright (C) 2008, 2009
- * Boaz Harrosh <bharrosh@panasas.com>
- *
- * This file is part of exofs.
- *
- * exofs is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation.  Since it is based on ext2, and the only
- * valid version of GPL for the Linux kernel is version 2, the only valid
- * version of GPL for exofs is version 2.
- *
- * exofs is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with exofs; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-
-#include <scsi/scsi_device.h>
-#include <scsi/osd_sense.h>
-
-#include "exofs.h"
-
-int exofs_check_ok_resid(struct osd_request *or, u64 *in_resid, u64 *out_resid)
-{
-	struct osd_sense_info osi;
-	int ret = osd_req_decode_sense(or, &osi);
-
-	if (ret) { /* translate to Linux codes */
-		if (osi.additional_code == scsi_invalid_field_in_cdb) {
-			if (osi.cdb_field_offset == OSD_CFO_STARTING_BYTE)
-				ret = -EFAULT;
-			if (osi.cdb_field_offset == OSD_CFO_OBJECT_ID)
-				ret = -ENOENT;
-			else
-				ret = -EINVAL;
-		} else if (osi.additional_code == osd_quota_error)
-			ret = -ENOSPC;
-		else
-			ret = -EIO;
-	}
-
-	/* FIXME: should be include in osd_sense_info */
-	if (in_resid)
-		*in_resid = or->in.req ? or->in.req->resid_len : 0;
-
-	if (out_resid)
-		*out_resid = or->out.req ? or->out.req->resid_len : 0;
-
-	return ret;
-}
-
-void exofs_make_credential(u8 cred_a[OSD_CAP_LEN], const struct osd_obj_id *obj)
-{
-	osd_sec_init_nosec_doall_caps(cred_a, obj, false, true);
-}
-
-/*
- * Perform a synchronous OSD operation.
- */
-int exofs_sync_op(struct osd_request *or, int timeout, uint8_t *credential)
-{
-	int ret;
-
-	or->timeout = timeout;
-	ret = osd_finalize_request(or, 0, credential, NULL);
-	if (ret) {
-		EXOFS_DBGMSG("Faild to osd_finalize_request() => %d\n", ret);
-		return ret;
-	}
-
-	ret = osd_execute_request(or);
-
-	if (ret)
-		EXOFS_DBGMSG("osd_execute_request() => %d\n", ret);
-	/* osd_req_decode_sense(or, ret); */
-	return ret;
-}
-
-/*
- * Perform an asynchronous OSD operation.
- */
-int exofs_async_op(struct osd_request *or, osd_req_done_fn *async_done,
-		   void *caller_context, u8 *cred)
-{
-	int ret;
-
-	ret = osd_finalize_request(or, 0, cred, NULL);
-	if (ret) {
-		EXOFS_DBGMSG("Faild to osd_finalize_request() => %d\n", ret);
-		return ret;
-	}
-
-	ret = osd_execute_request_async(or, async_done, caller_context);
-
-	if (ret)
-		EXOFS_DBGMSG("osd_execute_request_async() => %d\n", ret);
-	return ret;
-}
-
-int extract_attr_from_req(struct osd_request *or, struct osd_attr *attr)
-{
-	struct osd_attr cur_attr = {.attr_page = 0}; /* start with zeros */
-	void *iter = NULL;
-	int nelem;
-
-	do {
-		nelem = 1;
-		osd_req_decode_get_attr_list(or, &cur_attr, &nelem, &iter);
-		if ((cur_attr.attr_page == attr->attr_page) &&
-		    (cur_attr.attr_id == attr->attr_id)) {
-			attr->len = cur_attr.len;
-			attr->val_ptr = cur_attr.val_ptr;
-			return 0;
-		}
-	} while (iter);
-
-	return -EIO;
-}
diff --git a/fs/exofs/pnfs.h b/fs/exofs/pnfs.h
new file mode 100644
index 000000000000..c52e9888b8ab
--- /dev/null
+++ b/fs/exofs/pnfs.h
@@ -0,0 +1,45 @@
+/*
+ * Copyright (C) 2008, 2009
+ * Boaz Harrosh <bharrosh@panasas.com>
+ *
+ * This file is part of exofs.
+ *
+ * exofs is free software; you can redistribute it and/or modify it under the
+ * terms of the GNU General Public License  version 2 as published by the Free
+ * Software Foundation.
+ *
+ */
+
+/* FIXME: Remove this file once pnfs hits mainline */
+
+#ifndef __EXOFS_PNFS_H__
+#define __EXOFS_PNFS_H__
+
+#if ! defined(__PNFS_OSD_XDR_H__)
+
+enum pnfs_iomode {
+	IOMODE_READ = 1,
+	IOMODE_RW = 2,
+	IOMODE_ANY = 3,
+};
+
+/* Layout Structure */
+enum pnfs_osd_raid_algorithm4 {
+	PNFS_OSD_RAID_0		= 1,
+	PNFS_OSD_RAID_4		= 2,
+	PNFS_OSD_RAID_5		= 3,
+	PNFS_OSD_RAID_PQ	= 4     /* Reed-Solomon P+Q */
+};
+
+struct pnfs_osd_data_map {
+	u32	odm_num_comps;
+	u64	odm_stripe_unit;
+	u32	odm_group_width;
+	u32	odm_group_depth;
+	u32	odm_mirror_cnt;
+	u32	odm_raid_algorithm;
+};
+
+#endif /* ! defined(__PNFS_OSD_XDR_H__) */
+
+#endif /* __EXOFS_PNFS_H__ */
diff --git a/fs/exofs/super.c b/fs/exofs/super.c
index 9f500dec3b59..a1d1e77b12eb 100644
--- a/fs/exofs/super.c
+++ b/fs/exofs/super.c
@@ -203,49 +203,45 @@ int exofs_sync_fs(struct super_block *sb, int wait)
 {
 	struct exofs_sb_info *sbi;
 	struct exofs_fscb *fscb;
-	struct osd_request *or;
-	struct osd_obj_id obj;
+	struct exofs_io_state *ios;
 	int ret = -ENOMEM;
 
-	fscb = kzalloc(sizeof(struct exofs_fscb), GFP_KERNEL);
-	if (!fscb) {
-		EXOFS_ERR("exofs_write_super: memory allocation failed.\n");
-		return -ENOMEM;
-	}
-
 	lock_super(sb);
 	sbi = sb->s_fs_info;
+	fscb = &sbi->s_fscb;
+
+	ret = exofs_get_io_state(sbi, &ios);
+	if (ret)
+		goto out;
+
+	/* Note: We only write the changing part of the fscb. .i.e upto the
+	 *       the fscb->s_dev_table_oid member. There is no read-modify-write
+	 *       here.
+	 */
+	ios->length = offsetof(struct exofs_fscb, s_dev_table_oid);
+	memset(fscb, 0, ios->length);
 	fscb->s_nextid = cpu_to_le64(sbi->s_nextid);
 	fscb->s_numfiles = cpu_to_le32(sbi->s_numfiles);
 	fscb->s_magic = cpu_to_le16(sb->s_magic);
 	fscb->s_newfs = 0;
+	fscb->s_version = EXOFS_FSCB_VER;
 
-	or = osd_start_request(sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		EXOFS_ERR("exofs_write_super: osd_start_request failed.\n");
-		goto out;
-	}
+	ios->obj.id = EXOFS_SUPER_ID;
+	ios->offset = 0;
+	ios->kern_buff = fscb;
+	ios->cred = sbi->s_cred;
 
-	obj.partition = sbi->s_pid;
-	obj.id = EXOFS_SUPER_ID;
-	ret = osd_req_write_kern(or, &obj, 0, fscb, sizeof(*fscb));
+	ret = exofs_sbi_write(ios);
 	if (unlikely(ret)) {
-		EXOFS_ERR("exofs_write_super: osd_req_write_kern failed.\n");
-		goto out;
-	}
-
-	ret = exofs_sync_op(or, sbi->s_timeout, sbi->s_cred);
-	if (unlikely(ret)) {
-		EXOFS_ERR("exofs_write_super: exofs_sync_op failed.\n");
+		EXOFS_ERR("%s: exofs_sbi_write failed.\n", __func__);
 		goto out;
 	}
 	sb->s_dirt = 0;
 
 out:
-	if (or)
-		osd_end_request(or);
+	EXOFS_DBGMSG("s_nextid=0x%llx ret=%d\n", _LLU(sbi->s_nextid), ret);
+	exofs_put_io_state(ios);
 	unlock_super(sb);
-	kfree(fscb);
 	return ret;
 }
 
@@ -257,6 +253,29 @@ static void exofs_write_super(struct super_block *sb)
 		sb->s_dirt = 0;
 }
 
+static void _exofs_print_device(const char *msg, const char *dev_path,
+				struct osd_dev *od, u64 pid)
+{
+	const struct osd_dev_info *odi = osduld_device_info(od);
+
+	printk(KERN_NOTICE "exofs: %s %s osd_name-%s pid-0x%llx\n",
+		msg, dev_path ?: "", odi->osdname, _LLU(pid));
+}
+
+void exofs_free_sbi(struct exofs_sb_info *sbi)
+{
+	while (sbi->s_numdevs) {
+		int i = --sbi->s_numdevs;
+		struct osd_dev *od = sbi->s_ods[i];
+
+		if (od) {
+			sbi->s_ods[i] = NULL;
+			osduld_put_device(od);
+		}
+	}
+	kfree(sbi);
+}
+
 /*
  * This function is called when the vfs is freeing the superblock.  We just
  * need to free our own part.
@@ -279,11 +298,182 @@ static void exofs_put_super(struct super_block *sb)
 				  msecs_to_jiffies(100));
 	}
 
-	osduld_put_device(sbi->s_dev);
-	kfree(sb->s_fs_info);
+	_exofs_print_device("Unmounting", NULL, sbi->s_ods[0], sbi->s_pid);
+
+	exofs_free_sbi(sbi);
 	sb->s_fs_info = NULL;
 }
 
+static int _read_and_match_data_map(struct exofs_sb_info *sbi, unsigned numdevs,
+				    struct exofs_device_table *dt)
+{
+	sbi->data_map.odm_num_comps   =
+				le32_to_cpu(dt->dt_data_map.cb_num_comps);
+	sbi->data_map.odm_stripe_unit =
+				le64_to_cpu(dt->dt_data_map.cb_stripe_unit);
+	sbi->data_map.odm_group_width =
+				le32_to_cpu(dt->dt_data_map.cb_group_width);
+	sbi->data_map.odm_group_depth =
+				le32_to_cpu(dt->dt_data_map.cb_group_depth);
+	sbi->data_map.odm_mirror_cnt  =
+				le32_to_cpu(dt->dt_data_map.cb_mirror_cnt);
+	sbi->data_map.odm_raid_algorithm  =
+				le32_to_cpu(dt->dt_data_map.cb_raid_algorithm);
+
+/* FIXME: Hard coded mirror only for now. if not so do not mount */
+	if ((sbi->data_map.odm_num_comps != numdevs) ||
+	    (sbi->data_map.odm_stripe_unit != EXOFS_BLKSIZE) ||
+	    (sbi->data_map.odm_raid_algorithm != PNFS_OSD_RAID_0) ||
+	    (sbi->data_map.odm_mirror_cnt != (numdevs - 1)))
+		return -EINVAL;
+	else
+		return 0;
+}
+
+/* @odi is valid only as long as @fscb_dev is valid */
+static int exofs_devs_2_odi(struct exofs_dt_device_info *dt_dev,
+			     struct osd_dev_info *odi)
+{
+	odi->systemid_len = le32_to_cpu(dt_dev->systemid_len);
+	memcpy(odi->systemid, dt_dev->systemid, odi->systemid_len);
+
+	odi->osdname_len = le32_to_cpu(dt_dev->osdname_len);
+	odi->osdname = dt_dev->osdname;
+
+	/* FIXME support long names. Will need a _put function */
+	if (dt_dev->long_name_offset)
+		return -EINVAL;
+
+	/* Make sure osdname is printable!
+	 * mkexofs should give us space for a null-terminator else the
+	 * device-table is invalid.
+	 */
+	if (unlikely(odi->osdname_len >= sizeof(dt_dev->osdname)))
+		odi->osdname_len = sizeof(dt_dev->osdname) - 1;
+	dt_dev->osdname[odi->osdname_len] = 0;
+
+	/* If it's all zeros something is bad we read past end-of-obj */
+	return !(odi->systemid_len || odi->osdname_len);
+}
+
+static int exofs_read_lookup_dev_table(struct exofs_sb_info **psbi,
+				       unsigned table_count)
+{
+	struct exofs_sb_info *sbi = *psbi;
+	struct osd_dev *fscb_od;
+	struct osd_obj_id obj = {.partition = sbi->s_pid,
+				 .id = EXOFS_DEVTABLE_ID};
+	struct exofs_device_table *dt;
+	unsigned table_bytes = table_count * sizeof(dt->dt_dev_table[0]) +
+					     sizeof(*dt);
+	unsigned numdevs, i;
+	int ret;
+
+	dt = kmalloc(table_bytes, GFP_KERNEL);
+	if (unlikely(!dt)) {
+		EXOFS_ERR("ERROR: allocating %x bytes for device table\n",
+			  table_bytes);
+		return -ENOMEM;
+	}
+
+	fscb_od = sbi->s_ods[0];
+	sbi->s_ods[0] = NULL;
+	sbi->s_numdevs = 0;
+	ret = exofs_read_kern(fscb_od, sbi->s_cred, &obj, 0, dt, table_bytes);
+	if (unlikely(ret)) {
+		EXOFS_ERR("ERROR: reading device table\n");
+		goto out;
+	}
+
+	numdevs = le64_to_cpu(dt->dt_num_devices);
+	if (unlikely(!numdevs)) {
+		ret = -EINVAL;
+		goto out;
+	}
+	WARN_ON(table_count != numdevs);
+
+	ret = _read_and_match_data_map(sbi, numdevs, dt);
+	if (unlikely(ret))
+		goto out;
+
+	if (likely(numdevs > 1)) {
+		unsigned size = numdevs * sizeof(sbi->s_ods[0]);
+
+		sbi = krealloc(sbi, sizeof(*sbi) + size, GFP_KERNEL);
+		if (unlikely(!sbi)) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		memset(&sbi->s_ods[1], 0, size - sizeof(sbi->s_ods[0]));
+		*psbi = sbi;
+	}
+
+	for (i = 0; i < numdevs; i++) {
+		struct exofs_fscb fscb;
+		struct osd_dev_info odi;
+		struct osd_dev *od;
+
+		if (exofs_devs_2_odi(&dt->dt_dev_table[i], &odi)) {
+			EXOFS_ERR("ERROR: Read all-zeros device entry\n");
+			ret = -EINVAL;
+			goto out;
+		}
+
+		printk(KERN_NOTICE "Add device[%d]: osd_name-%s\n",
+		       i, odi.osdname);
+
+		/* On all devices the device table is identical. The user can
+		 * specify any one of the participating devices on the command
+		 * line. We always keep them in device-table order.
+		 */
+		if (fscb_od && osduld_device_same(fscb_od, &odi)) {
+			sbi->s_ods[i] = fscb_od;
+			++sbi->s_numdevs;
+			fscb_od = NULL;
+			continue;
+		}
+
+		od = osduld_info_lookup(&odi);
+		if (unlikely(IS_ERR(od))) {
+			ret = PTR_ERR(od);
+			EXOFS_ERR("ERROR: device requested is not found "
+				  "osd_name-%s =>%d\n", odi.osdname, ret);
+			goto out;
+		}
+
+		sbi->s_ods[i] = od;
+		++sbi->s_numdevs;
+
+		/* Read the fscb of the other devices to make sure the FS
+		 * partition is there.
+		 */
+		ret = exofs_read_kern(od, sbi->s_cred, &obj, 0, &fscb,
+				      sizeof(fscb));
+		if (unlikely(ret)) {
+			EXOFS_ERR("ERROR: Malformed participating device "
+				  "error reading fscb osd_name-%s\n",
+				  odi.osdname);
+			goto out;
+		}
+
+		/* TODO: verify other information is correct and FS-uuid
+		 *	 matches. Benny what did you say about device table
+		 *	 generation and old devices?
+		 */
+	}
+
+out:
+	kfree(dt);
+	if (unlikely(!ret && fscb_od)) {
+		EXOFS_ERR(
+		      "ERROR: Bad device-table container device not present\n");
+		osduld_put_device(fscb_od);
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
 /*
  * Read the superblock from the OSD and fill in the fields
  */
@@ -292,24 +482,25 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent)
 	struct inode *root;
 	struct exofs_mountopt *opts = data;
 	struct exofs_sb_info *sbi;	/*extended info                  */
+	struct osd_dev *od;		/* Master device                 */
 	struct exofs_fscb fscb;		/*on-disk superblock info        */
-	struct osd_request *or = NULL;
 	struct osd_obj_id obj;
+	unsigned table_count;
 	int ret;
 
 	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
-	sb->s_fs_info = sbi;
 
 	/* use mount options to fill superblock */
-	sbi->s_dev = osduld_path_lookup(opts->dev_name);
-	if (IS_ERR(sbi->s_dev)) {
-		ret = PTR_ERR(sbi->s_dev);
-		sbi->s_dev = NULL;
+	od = osduld_path_lookup(opts->dev_name);
+	if (IS_ERR(od)) {
+		ret = PTR_ERR(od);
 		goto free_sbi;
 	}
 
+	sbi->s_ods[0] = od;
+	sbi->s_numdevs = 1;
 	sbi->s_pid = opts->pid;
 	sbi->s_timeout = opts->timeout;
 
@@ -323,35 +514,13 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_bdev = NULL;
 	sb->s_dev = 0;
 
-	/* read data from on-disk superblock object */
 	obj.partition = sbi->s_pid;
 	obj.id = EXOFS_SUPER_ID;
 	exofs_make_credential(sbi->s_cred, &obj);
 
-	or = osd_start_request(sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		if (!silent)
-			EXOFS_ERR(
-			       "exofs_fill_super: osd_start_request failed.\n");
-		ret = -ENOMEM;
-		goto free_sbi;
-	}
-	ret = osd_req_read_kern(or, &obj, 0, &fscb, sizeof(fscb));
-	if (unlikely(ret)) {
-		if (!silent)
-			EXOFS_ERR(
-			       "exofs_fill_super: osd_req_read_kern failed.\n");
-		ret = -ENOMEM;
-		goto free_sbi;
-	}
-
-	ret = exofs_sync_op(or, sbi->s_timeout, sbi->s_cred);
-	if (unlikely(ret)) {
-		if (!silent)
-			EXOFS_ERR("exofs_fill_super: exofs_sync_op failed.\n");
-		ret = -EIO;
+	ret = exofs_read_kern(od, sbi->s_cred, &obj, 0, &fscb, sizeof(fscb));
+	if (unlikely(ret))
 		goto free_sbi;
-	}
 
 	sb->s_magic = le16_to_cpu(fscb.s_magic);
 	sbi->s_nextid = le64_to_cpu(fscb.s_nextid);
@@ -364,12 +533,26 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent)
 		ret = -EINVAL;
 		goto free_sbi;
 	}
+	if (le32_to_cpu(fscb.s_version) != EXOFS_FSCB_VER) {
+		EXOFS_ERR("ERROR: Bad FSCB version expected-%d got-%d\n",
+			  EXOFS_FSCB_VER, le32_to_cpu(fscb.s_version));
+		ret = -EINVAL;
+		goto free_sbi;
+	}
 
 	/* start generation numbers from a random point */
 	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
 	spin_lock_init(&sbi->s_next_gen_lock);
 
+	table_count = le64_to_cpu(fscb.s_dev_table_count);
+	if (table_count) {
+		ret = exofs_read_lookup_dev_table(&sbi, table_count);
+		if (unlikely(ret))
+			goto free_sbi;
+	}
+
 	/* set up operation vectors */
+	sb->s_fs_info = sbi;
 	sb->s_op = &exofs_sops;
 	sb->s_export_op = &exofs_export_ops;
 	root = exofs_iget(sb, EXOFS_ROOT_ID - EXOFS_OBJ_OFF);
@@ -395,16 +578,15 @@ static int exofs_fill_super(struct super_block *sb, void *data, int silent)
 		goto free_sbi;
 	}
 
-	ret = 0;
-out:
-	if (or)
-		osd_end_request(or);
-	return ret;
+	_exofs_print_device("Mounting", opts->dev_name, sbi->s_ods[0],
+			    sbi->s_pid);
+	return 0;
 
 free_sbi:
-	osduld_put_device(sbi->s_dev); /* NULL safe */
-	kfree(sbi);
-	goto out;
+	EXOFS_ERR("Unable to mount exofs on %s pid=0x%llx err=%d\n",
+		  opts->dev_name, sbi->s_pid, ret);
+	exofs_free_sbi(sbi);
+	return ret;
 }
 
 /*
@@ -433,7 +615,7 @@ static int exofs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct super_block *sb = dentry->d_sb;
 	struct exofs_sb_info *sbi = sb->s_fs_info;
-	struct osd_obj_id obj = {sbi->s_pid, 0};
+	struct exofs_io_state *ios;
 	struct osd_attr attrs[] = {
 		ATTR_DEF(OSD_APAGE_PARTITION_QUOTAS,
 			OSD_ATTR_PQ_CAPACITY_QUOTA, sizeof(__be64)),
@@ -442,32 +624,33 @@ static int exofs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	};
 	uint64_t capacity = ULLONG_MAX;
 	uint64_t used = ULLONG_MAX;
-	struct osd_request *or;
 	uint8_t cred_a[OSD_CAP_LEN];
 	int ret;
 
-	/* get used/capacity attributes */
-	exofs_make_credential(cred_a, &obj);
-
-	or = osd_start_request(sbi->s_dev, GFP_KERNEL);
-	if (unlikely(!or)) {
-		EXOFS_DBGMSG("exofs_statfs: osd_start_request failed.\n");
-		return -ENOMEM;
+	ret = exofs_get_io_state(sbi, &ios);
+	if (ret) {
+		EXOFS_DBGMSG("exofs_get_io_state failed.\n");
+		return ret;
 	}
 
-	osd_req_get_attributes(or, &obj);
-	osd_req_add_get_attr_list(or, attrs, ARRAY_SIZE(attrs));
-	ret = exofs_sync_op(or, sbi->s_timeout, cred_a);
+	exofs_make_credential(cred_a, &ios->obj);
+	ios->cred = sbi->s_cred;
+	ios->in_attr = attrs;
+	ios->in_attr_len = ARRAY_SIZE(attrs);
+
+	ret = exofs_sbi_read(ios);
 	if (unlikely(ret))
 		goto out;
 
-	ret = extract_attr_from_req(or, &attrs[0]);
-	if (likely(!ret))
+	ret = extract_attr_from_ios(ios, &attrs[0]);
+	if (likely(!ret)) {
 		capacity = get_unaligned_be64(attrs[0].val_ptr);
-	else
+		if (unlikely(!capacity))
+			capacity = ULLONG_MAX;
+	} else
 		EXOFS_DBGMSG("exofs_statfs: get capacity failed.\n");
 
-	ret = extract_attr_from_req(or, &attrs[1]);
+	ret = extract_attr_from_ios(ios, &attrs[1]);
 	if (likely(!ret))
 		used = get_unaligned_be64(attrs[1].val_ptr);
 	else
@@ -476,15 +659,15 @@ static int exofs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	/* fill in the stats buffer */
 	buf->f_type = EXOFS_SUPER_MAGIC;
 	buf->f_bsize = EXOFS_BLKSIZE;
-	buf->f_blocks = (capacity >> EXOFS_BLKSHIFT);
-	buf->f_bfree = ((capacity - used) >> EXOFS_BLKSHIFT);
+	buf->f_blocks = capacity >> 9;
+	buf->f_bfree = (capacity - used) >> 9;
 	buf->f_bavail = buf->f_bfree;
 	buf->f_files = sbi->s_numfiles;
 	buf->f_ffree = EXOFS_MAX_ID - sbi->s_numfiles;
 	buf->f_namelen = EXOFS_NAME_LEN;
 
 out:
-	osd_end_request(or);
+	exofs_put_io_state(ios);
 	return ret;
 }
 
diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c
index 197c7db583c7..e9e175949a63 100644
--- a/fs/exportfs/expfs.c
+++ b/fs/exportfs/expfs.c
@@ -6,7 +6,7 @@
  * and for mapping back from file handles to dentries.
  *
  * For details on why we do all the strange and hairy things in here
- * take a look at Documentation/filesystems/Exporting.
+ * take a look at Documentation/filesystems/nfs/Exporting.
  */
 #include <linux/exportfs.h>
 #include <linux/fs.h>
diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index a63d44256a70..a99e54318c3d 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -339,12 +339,12 @@ ext2_acl_chmod(struct inode *inode)
  * Extended attribut handlers
  */
 static size_t
-ext2_xattr_list_acl_access(struct inode *inode, char *list, size_t list_size,
-			   const char *name, size_t name_len)
+ext2_xattr_list_acl_access(struct dentry *dentry, char *list, size_t list_size,
+			   const char *name, size_t name_len, int type)
 {
 	const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS);
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_size)
 		memcpy(list, POSIX_ACL_XATTR_ACCESS, size);
@@ -352,12 +352,12 @@ ext2_xattr_list_acl_access(struct inode *inode, char *list, size_t list_size,
 }
 
 static size_t
-ext2_xattr_list_acl_default(struct inode *inode, char *list, size_t list_size,
-			    const char *name, size_t name_len)
+ext2_xattr_list_acl_default(struct dentry *dentry, char *list, size_t list_size,
+			    const char *name, size_t name_len, int type)
 {
 	const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT);
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_size)
 		memcpy(list, POSIX_ACL_XATTR_DEFAULT, size);
@@ -365,15 +365,18 @@ ext2_xattr_list_acl_default(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext2_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size)
+ext2_xattr_get_acl(struct dentry *dentry, const char *name, void *buffer,
+		   size_t size, int type)
 {
 	struct posix_acl *acl;
 	int error;
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return -EOPNOTSUPP;
 
-	acl = ext2_get_acl(inode, type);
+	acl = ext2_get_acl(dentry->d_inode, type);
 	if (IS_ERR(acl))
 		return PTR_ERR(acl);
 	if (acl == NULL)
@@ -385,33 +388,17 @@ ext2_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size)
 }
 
 static int
-ext2_xattr_get_acl_access(struct inode *inode, const char *name,
-			  void *buffer, size_t size)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext2_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size);
-}
-
-static int
-ext2_xattr_get_acl_default(struct inode *inode, const char *name,
-			   void *buffer, size_t size)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext2_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size);
-}
-
-static int
-ext2_xattr_set_acl(struct inode *inode, int type, const void *value,
-		   size_t size)
+ext2_xattr_set_acl(struct dentry *dentry, const char *name, const void *value,
+		   size_t size, int flags, int type)
 {
 	struct posix_acl *acl;
 	int error;
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return -EOPNOTSUPP;
-	if (!is_owner_or_cap(inode))
+	if (!is_owner_or_cap(dentry->d_inode))
 		return -EPERM;
 
 	if (value) {
@@ -426,41 +413,25 @@ ext2_xattr_set_acl(struct inode *inode, int type, const void *value,
 	} else
 		acl = NULL;
 
-	error = ext2_set_acl(inode, type, acl);
+	error = ext2_set_acl(dentry->d_inode, type, acl);
 
 release_and_out:
 	posix_acl_release(acl);
 	return error;
 }
 
-static int
-ext2_xattr_set_acl_access(struct inode *inode, const char *name,
-			  const void *value, size_t size, int flags)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext2_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size);
-}
-
-static int
-ext2_xattr_set_acl_default(struct inode *inode, const char *name,
-			   const void *value, size_t size, int flags)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext2_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size);
-}
-
 struct xattr_handler ext2_xattr_acl_access_handler = {
 	.prefix	= POSIX_ACL_XATTR_ACCESS,
+	.flags	= ACL_TYPE_ACCESS,
 	.list	= ext2_xattr_list_acl_access,
-	.get	= ext2_xattr_get_acl_access,
-	.set	= ext2_xattr_set_acl_access,
+	.get	= ext2_xattr_get_acl,
+	.set	= ext2_xattr_set_acl,
 };
 
 struct xattr_handler ext2_xattr_acl_default_handler = {
 	.prefix	= POSIX_ACL_XATTR_DEFAULT,
+	.flags	= ACL_TYPE_DEFAULT,
 	.list	= ext2_xattr_list_acl_default,
-	.get	= ext2_xattr_get_acl_default,
-	.set	= ext2_xattr_set_acl_default,
+	.get	= ext2_xattr_get_acl,
+	.set	= ext2_xattr_set_acl,
 };
diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c
index 6cde970b0a1a..7516957273ed 100644
--- a/fs/ext2/dir.c
+++ b/fs/ext2/dir.c
@@ -353,8 +353,8 @@ ext2_readdir (struct file * filp, void * dirent, filldir_t filldir)
  *	ext2_find_entry()
  *
  * finds an entry in the specified directory with the wanted name. It
- * returns the page in which the entry was found, and the entry itself
- * (as a parameter - res_dir). Page is returned mapped and unlocked.
+ * returns the page in which the entry was found (as a parameter - res_page),
+ * and the entry itself. Page is returned mapped and unlocked.
  * Entry is guaranteed to be valid.
  */
 struct ext2_dir_entry_2 *ext2_find_entry (struct inode * dir,
@@ -721,5 +721,5 @@ const struct file_operations ext2_dir_operations = {
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= ext2_compat_ioctl,
 #endif
-	.fsync		= simple_fsync,
+	.fsync		= ext2_fsync,
 };
diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index 9a8a8e27a063..061914add3cf 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -142,7 +142,7 @@ struct dentry *ext2_get_parent(struct dentry *child);
 /* super.c */
 extern void ext2_error (struct super_block *, const char *, const char *, ...)
 	__attribute__ ((format (printf, 3, 4)));
-extern void ext2_warning (struct super_block *, const char *, const char *, ...)
+extern void ext2_msg(struct super_block *, const char *, const char *, ...)
 	__attribute__ ((format (printf, 3, 4)));
 extern void ext2_update_dynamic_rev (struct super_block *sb);
 extern void ext2_write_super (struct super_block *);
@@ -155,6 +155,7 @@ extern void ext2_write_super (struct super_block *);
 extern const struct file_operations ext2_dir_operations;
 
 /* file.c */
+extern int ext2_fsync(struct file *file, struct dentry *dentry, int datasync);
 extern const struct inode_operations ext2_file_inode_operations;
 extern const struct file_operations ext2_file_operations;
 extern const struct file_operations ext2_xip_file_operations;
diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index a2f3afd1a1c1..586e3589d4c2 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -19,6 +19,7 @@
  */
 
 #include <linux/time.h>
+#include <linux/pagemap.h>
 #include "ext2.h"
 #include "xattr.h"
 #include "acl.h"
@@ -38,6 +39,22 @@ static int ext2_release_file (struct inode * inode, struct file * filp)
 	return 0;
 }
 
+int ext2_fsync(struct file *file, struct dentry *dentry, int datasync)
+{
+	int ret;
+	struct super_block *sb = dentry->d_inode->i_sb;
+	struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping;
+
+	ret = simple_fsync(file, dentry, datasync);
+	if (ret == -EIO || test_and_clear_bit(AS_EIO, &mapping->flags)) {
+		/* We don't really know where the IO error happened... */
+		ext2_error(sb, __func__,
+			   "detected IO error when writing metadata buffers");
+		ret = -EIO;
+	}
+	return ret;
+}
+
 /*
  * We have mostly NULL's here: the current defaults are ok for
  * the ext2 filesystem.
@@ -55,7 +72,7 @@ const struct file_operations ext2_file_operations = {
 	.mmap		= generic_file_mmap,
 	.open		= generic_file_open,
 	.release	= ext2_release_file,
-	.fsync		= simple_fsync,
+	.fsync		= ext2_fsync,
 	.splice_read	= generic_file_splice_read,
 	.splice_write	= generic_file_splice_write,
 };
@@ -72,7 +89,7 @@ const struct file_operations ext2_xip_file_operations = {
 	.mmap		= xip_file_mmap,
 	.open		= generic_file_open,
 	.release	= ext2_release_file,
-	.fsync		= simple_fsync,
+	.fsync		= ext2_fsync,
 };
 #endif
 
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index ade634076d0a..71b032c65a02 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -137,7 +137,8 @@ static int ext2_block_to_path(struct inode *inode,
 	int final = 0;
 
 	if (i_block < 0) {
-		ext2_warning (inode->i_sb, "ext2_block_to_path", "block < 0");
+		ext2_msg(inode->i_sb, KERN_WARNING,
+			"warning: %s: block < 0", __func__);
 	} else if (i_block < direct_blocks) {
 		offsets[n++] = i_block;
 		final = direct_blocks;
@@ -157,7 +158,8 @@ static int ext2_block_to_path(struct inode *inode,
 		offsets[n++] = i_block & (ptrs - 1);
 		final = ptrs;
 	} else {
-		ext2_warning (inode->i_sb, "ext2_block_to_path", "block > big");
+		ext2_msg(inode->i_sb, KERN_WARNING,
+			"warning: %s: block is too big", __func__);
 	}
 	if (boundary)
 		*boundary = final - 1 - (i_block & (ptrs - 1));
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 1a9ffee47d56..f9cb54a585ce 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -58,27 +58,27 @@ void ext2_error (struct super_block * sb, const char * function,
 	}
 
 	va_start(args, fmt);
-	printk(KERN_CRIT "EXT2-fs error (device %s): %s: ",sb->s_id, function);
+	printk(KERN_CRIT "EXT2-fs (%s): error: %s: ", sb->s_id, function);
 	vprintk(fmt, args);
 	printk("\n");
 	va_end(args);
 
 	if (test_opt(sb, ERRORS_PANIC))
-		panic("EXT2-fs panic from previous error\n");
+		panic("EXT2-fs: panic from previous error\n");
 	if (test_opt(sb, ERRORS_RO)) {
-		printk("Remounting filesystem read-only\n");
+		ext2_msg(sb, KERN_CRIT,
+			     "error: remounting filesystem read-only");
 		sb->s_flags |= MS_RDONLY;
 	}
 }
 
-void ext2_warning (struct super_block * sb, const char * function,
-		   const char * fmt, ...)
+void ext2_msg(struct super_block *sb, const char *prefix,
+		const char *fmt, ...)
 {
 	va_list args;
 
 	va_start(args, fmt);
-	printk(KERN_WARNING "EXT2-fs warning (device %s): %s: ",
-	       sb->s_id, function);
+	printk("%sEXT2-fs (%s): ", prefix, sb->s_id);
 	vprintk(fmt, args);
 	printk("\n");
 	va_end(args);
@@ -91,9 +91,9 @@ void ext2_update_dynamic_rev(struct super_block *sb)
 	if (le32_to_cpu(es->s_rev_level) > EXT2_GOOD_OLD_REV)
 		return;
 
-	ext2_warning(sb, __func__,
-		     "updating to rev %d because of new feature flag, "
-		     "running e2fsck is recommended",
+	ext2_msg(sb, KERN_WARNING,
+		     "warning: updating to rev %d because of "
+		     "new feature flag, running e2fsck is recommended",
 		     EXT2_DYNAMIC_REV);
 
 	es->s_first_ino = cpu_to_le32(EXT2_GOOD_OLD_FIRST_INO);
@@ -419,10 +419,10 @@ static const match_table_t tokens = {
 	{Opt_err, NULL}
 };
 
-static int parse_options (char * options,
-			  struct ext2_sb_info *sbi)
+static int parse_options(char *options, struct super_block *sb)
 {
-	char * p;
+	char *p;
+	struct ext2_sb_info *sbi = EXT2_SB(sb);
 	substring_t args[MAX_OPT_ARGS];
 	int option;
 
@@ -505,7 +505,8 @@ static int parse_options (char * options,
 #else
 		case Opt_user_xattr:
 		case Opt_nouser_xattr:
-			printk("EXT2 (no)user_xattr options not supported\n");
+			ext2_msg(sb, KERN_INFO, "(no)user_xattr options"
+				"not supported");
 			break;
 #endif
 #ifdef CONFIG_EXT2_FS_POSIX_ACL
@@ -518,14 +519,15 @@ static int parse_options (char * options,
 #else
 		case Opt_acl:
 		case Opt_noacl:
-			printk("EXT2 (no)acl options not supported\n");
+			ext2_msg(sb, KERN_INFO,
+				"(no)acl options not supported");
 			break;
 #endif
 		case Opt_xip:
 #ifdef CONFIG_EXT2_FS_XIP
 			set_opt (sbi->s_mount_opt, XIP);
 #else
-			printk("EXT2 xip option not supported\n");
+			ext2_msg(sb, KERN_INFO, "xip option not supported");
 #endif
 			break;
 
@@ -542,19 +544,18 @@ static int parse_options (char * options,
 		case Opt_quota:
 		case Opt_usrquota:
 		case Opt_grpquota:
-			printk(KERN_ERR
-				"EXT2-fs: quota operations not supported.\n");
-
+			ext2_msg(sb, KERN_INFO,
+				"quota operations not supported");
 			break;
 #endif
 
 		case Opt_reservation:
 			set_opt(sbi->s_mount_opt, RESERVATION);
-			printk("reservations ON\n");
+			ext2_msg(sb, KERN_INFO, "reservations ON");
 			break;
 		case Opt_noreservation:
 			clear_opt(sbi->s_mount_opt, RESERVATION);
-			printk("reservations OFF\n");
+			ext2_msg(sb, KERN_INFO, "reservations OFF");
 			break;
 		case Opt_ignore:
 			break;
@@ -573,34 +574,40 @@ static int ext2_setup_super (struct super_block * sb,
 	struct ext2_sb_info *sbi = EXT2_SB(sb);
 
 	if (le32_to_cpu(es->s_rev_level) > EXT2_MAX_SUPP_REV) {
-		printk ("EXT2-fs warning: revision level too high, "
-			"forcing read-only mode\n");
+		ext2_msg(sb, KERN_ERR,
+			"error: revision level too high, "
+			"forcing read-only mode");
 		res = MS_RDONLY;
 	}
 	if (read_only)
 		return res;
 	if (!(sbi->s_mount_state & EXT2_VALID_FS))
-		printk ("EXT2-fs warning: mounting unchecked fs, "
-			"running e2fsck is recommended\n");
+		ext2_msg(sb, KERN_WARNING,
+			"warning: mounting unchecked fs, "
+			"running e2fsck is recommended");
 	else if ((sbi->s_mount_state & EXT2_ERROR_FS))
-		printk ("EXT2-fs warning: mounting fs with errors, "
-			"running e2fsck is recommended\n");
+		ext2_msg(sb, KERN_WARNING,
+			"warning: mounting fs with errors, "
+			"running e2fsck is recommended");
 	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 &&
 		 le16_to_cpu(es->s_mnt_count) >=
 		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
-		printk ("EXT2-fs warning: maximal mount count reached, "
-			"running e2fsck is recommended\n");
+		ext2_msg(sb, KERN_WARNING,
+			"warning: maximal mount count reached, "
+			"running e2fsck is recommended");
 	else if (le32_to_cpu(es->s_checkinterval) &&
-		(le32_to_cpu(es->s_lastcheck) + le32_to_cpu(es->s_checkinterval) <= get_seconds()))
-		printk ("EXT2-fs warning: checktime reached, "
-			"running e2fsck is recommended\n");
+		(le32_to_cpu(es->s_lastcheck) +
+			le32_to_cpu(es->s_checkinterval) <= get_seconds()))
+		ext2_msg(sb, KERN_WARNING,
+			"warning: checktime reached, "
+			"running e2fsck is recommended");
 	if (!le16_to_cpu(es->s_max_mnt_count))
 		es->s_max_mnt_count = cpu_to_le16(EXT2_DFL_MAX_MNT_COUNT);
 	le16_add_cpu(&es->s_mnt_count, 1);
 	ext2_write_super(sb);
 	if (test_opt (sb, DEBUG))
-		printk ("[EXT II FS %s, %s, bs=%lu, fs=%lu, gc=%lu, "
-			"bpg=%lu, ipg=%lu, mo=%04lx]\n",
+		ext2_msg(sb, KERN_INFO, "%s, %s, bs=%lu, fs=%lu, gc=%lu, "
+			"bpg=%lu, ipg=%lu, mo=%04lx]",
 			EXT2FS_VERSION, EXT2FS_DATE, sb->s_blocksize,
 			sbi->s_frag_size,
 			sbi->s_groups_count,
@@ -767,7 +774,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	 */
 	blocksize = sb_min_blocksize(sb, BLOCK_SIZE);
 	if (!blocksize) {
-		printk ("EXT2-fs: unable to set blocksize\n");
+		ext2_msg(sb, KERN_ERR, "error: unable to set blocksize");
 		goto failed_sbi;
 	}
 
@@ -783,7 +790,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	}
 
 	if (!(bh = sb_bread(sb, logic_sb_block))) {
-		printk ("EXT2-fs: unable to read superblock\n");
+		ext2_msg(sb, KERN_ERR, "error: unable to read superblock");
 		goto failed_sbi;
 	}
 	/*
@@ -826,7 +833,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	
 	set_opt(sbi->s_mount_opt, RESERVATION);
 
-	if (!parse_options ((char *) data, sbi))
+	if (!parse_options((char *) data, sb))
 		goto failed_mount;
 
 	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
@@ -840,8 +847,9 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	    (EXT2_HAS_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT2_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT2_HAS_INCOMPAT_FEATURE(sb, ~0U)))
-		printk("EXT2-fs warning: feature flags set on rev 0 fs, "
-		       "running e2fsck is recommended\n");
+		ext2_msg(sb, KERN_WARNING,
+			"warning: feature flags set on rev 0 fs, "
+			"running e2fsck is recommended");
 	/*
 	 * Check feature flags regardless of the revision level, since we
 	 * previously didn't change the revision level when setting the flags,
@@ -849,16 +857,16 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	 */
 	features = EXT2_HAS_INCOMPAT_FEATURE(sb, ~EXT2_FEATURE_INCOMPAT_SUPP);
 	if (features) {
-		printk("EXT2-fs: %s: couldn't mount because of "
-		       "unsupported optional features (%x).\n",
-		       sb->s_id, le32_to_cpu(features));
+		ext2_msg(sb, KERN_ERR,	"error: couldn't mount because of "
+		       "unsupported optional features (%x)",
+			le32_to_cpu(features));
 		goto failed_mount;
 	}
 	if (!(sb->s_flags & MS_RDONLY) &&
 	    (features = EXT2_HAS_RO_COMPAT_FEATURE(sb, ~EXT2_FEATURE_RO_COMPAT_SUPP))){
-		printk("EXT2-fs: %s: couldn't mount RDWR because of "
-		       "unsupported optional features (%x).\n",
-		       sb->s_id, le32_to_cpu(features));
+		ext2_msg(sb, KERN_ERR, "error: couldn't mount RDWR because of "
+		       "unsupported optional features (%x)",
+		       le32_to_cpu(features));
 		goto failed_mount;
 	}
 
@@ -866,7 +874,8 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 
 	if (ext2_use_xip(sb) && blocksize != PAGE_SIZE) {
 		if (!silent)
-			printk("XIP: Unsupported blocksize\n");
+			ext2_msg(sb, KERN_ERR,
+				"error: unsupported blocksize for xip");
 		goto failed_mount;
 	}
 
@@ -875,7 +884,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 		brelse(bh);
 
 		if (!sb_set_blocksize(sb, blocksize)) {
-			printk(KERN_ERR "EXT2-fs: blocksize too small for device.\n");
+			ext2_msg(sb, KERN_ERR, "error: blocksize is too small");
 			goto failed_sbi;
 		}
 
@@ -883,14 +892,14 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 		offset = (sb_block*BLOCK_SIZE) % blocksize;
 		bh = sb_bread(sb, logic_sb_block);
 		if(!bh) {
-			printk("EXT2-fs: Couldn't read superblock on "
-			       "2nd try.\n");
+			ext2_msg(sb, KERN_ERR, "error: couldn't read"
+				"superblock on 2nd try");
 			goto failed_sbi;
 		}
 		es = (struct ext2_super_block *) (((char *)bh->b_data) + offset);
 		sbi->s_es = es;
 		if (es->s_magic != cpu_to_le16(EXT2_SUPER_MAGIC)) {
-			printk ("EXT2-fs: Magic mismatch, very weird !\n");
+			ext2_msg(sb, KERN_ERR, "error: magic mismatch");
 			goto failed_mount;
 		}
 	}
@@ -906,7 +915,8 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 		if ((sbi->s_inode_size < EXT2_GOOD_OLD_INODE_SIZE) ||
 		    !is_power_of_2(sbi->s_inode_size) ||
 		    (sbi->s_inode_size > blocksize)) {
-			printk ("EXT2-fs: unsupported inode size: %d\n",
+			ext2_msg(sb, KERN_ERR,
+				"error: unsupported inode size: %d",
 				sbi->s_inode_size);
 			goto failed_mount;
 		}
@@ -943,29 +953,33 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 
 	if (sb->s_blocksize != bh->b_size) {
 		if (!silent)
-			printk ("VFS: Unsupported blocksize on dev "
-				"%s.\n", sb->s_id);
+			ext2_msg(sb, KERN_ERR, "error: unsupported blocksize");
 		goto failed_mount;
 	}
 
 	if (sb->s_blocksize != sbi->s_frag_size) {
-		printk ("EXT2-fs: fragsize %lu != blocksize %lu (not supported yet)\n",
+		ext2_msg(sb, KERN_ERR,
+			"error: fragsize %lu != blocksize %lu"
+			"(not supported yet)",
 			sbi->s_frag_size, sb->s_blocksize);
 		goto failed_mount;
 	}
 
 	if (sbi->s_blocks_per_group > sb->s_blocksize * 8) {
-		printk ("EXT2-fs: #blocks per group too big: %lu\n",
+		ext2_msg(sb, KERN_ERR,
+			"error: #blocks per group too big: %lu",
 			sbi->s_blocks_per_group);
 		goto failed_mount;
 	}
 	if (sbi->s_frags_per_group > sb->s_blocksize * 8) {
-		printk ("EXT2-fs: #fragments per group too big: %lu\n",
+		ext2_msg(sb, KERN_ERR,
+			"error: #fragments per group too big: %lu",
 			sbi->s_frags_per_group);
 		goto failed_mount;
 	}
 	if (sbi->s_inodes_per_group > sb->s_blocksize * 8) {
-		printk ("EXT2-fs: #inodes per group too big: %lu\n",
+		ext2_msg(sb, KERN_ERR,
+			"error: #inodes per group too big: %lu",
 			sbi->s_inodes_per_group);
 		goto failed_mount;
 	}
@@ -979,13 +993,13 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 		   EXT2_DESC_PER_BLOCK(sb);
 	sbi->s_group_desc = kmalloc (db_count * sizeof (struct buffer_head *), GFP_KERNEL);
 	if (sbi->s_group_desc == NULL) {
-		printk ("EXT2-fs: not enough memory\n");
+		ext2_msg(sb, KERN_ERR, "error: not enough memory");
 		goto failed_mount;
 	}
 	bgl_lock_init(sbi->s_blockgroup_lock);
 	sbi->s_debts = kcalloc(sbi->s_groups_count, sizeof(*sbi->s_debts), GFP_KERNEL);
 	if (!sbi->s_debts) {
-		printk ("EXT2-fs: not enough memory\n");
+		ext2_msg(sb, KERN_ERR, "error: not enough memory");
 		goto failed_mount_group_desc;
 	}
 	for (i = 0; i < db_count; i++) {
@@ -994,12 +1008,13 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 		if (!sbi->s_group_desc[i]) {
 			for (j = 0; j < i; j++)
 				brelse (sbi->s_group_desc[j]);
-			printk ("EXT2-fs: unable to read group descriptors\n");
+			ext2_msg(sb, KERN_ERR,
+				"error: unable to read group descriptors");
 			goto failed_mount_group_desc;
 		}
 	}
 	if (!ext2_check_descriptors (sb)) {
-		printk ("EXT2-fs: group descriptors corrupted!\n");
+		ext2_msg(sb, KERN_ERR, "group descriptors corrupted");
 		goto failed_mount2;
 	}
 	sbi->s_gdb_count = db_count;
@@ -1032,7 +1047,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 				ext2_count_dirs(sb));
 	}
 	if (err) {
-		printk(KERN_ERR "EXT2-fs: insufficient memory\n");
+		ext2_msg(sb, KERN_ERR, "error: insufficient memory");
 		goto failed_mount3;
 	}
 	/*
@@ -1048,27 +1063,28 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	}
 	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
 		iput(root);
-		printk(KERN_ERR "EXT2-fs: corrupt root inode, run e2fsck\n");
+		ext2_msg(sb, KERN_ERR, "error: corrupt root inode, run e2fsck");
 		goto failed_mount3;
 	}
 
 	sb->s_root = d_alloc_root(root);
 	if (!sb->s_root) {
 		iput(root);
-		printk(KERN_ERR "EXT2-fs: get root inode failed\n");
+		ext2_msg(sb, KERN_ERR, "error: get root inode failed");
 		ret = -ENOMEM;
 		goto failed_mount3;
 	}
 	if (EXT2_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL))
-		ext2_warning(sb, __func__,
-			"mounting ext3 filesystem as ext2");
+		ext2_msg(sb, KERN_WARNING,
+			"warning: mounting ext3 filesystem as ext2");
 	ext2_setup_super (sb, es, sb->s_flags & MS_RDONLY);
 	return 0;
 
 cantfind_ext2:
 	if (!silent)
-		printk("VFS: Can't find an ext2 filesystem on dev %s.\n",
-		       sb->s_id);
+		ext2_msg(sb, KERN_ERR,
+			"error: can't find an ext2 filesystem on dev %s.",
+			sb->s_id);
 	goto failed_mount;
 failed_mount3:
 	percpu_counter_destroy(&sbi->s_freeblocks_counter);
@@ -1089,9 +1105,30 @@ failed_sbi:
 	return ret;
 }
 
+static void ext2_clear_super_error(struct super_block *sb)
+{
+	struct buffer_head *sbh = EXT2_SB(sb)->s_sbh;
+
+	if (buffer_write_io_error(sbh)) {
+		/*
+		 * Oh, dear.  A previous attempt to write the
+		 * superblock failed.  This could happen because the
+		 * USB device was yanked out.  Or it could happen to
+		 * be a transient write error and maybe the block will
+		 * be remapped.  Nothing we can do but to retry the
+		 * write and hope for the best.
+		 */
+		printk(KERN_ERR "EXT2-fs: %s previous I/O error to "
+		       "superblock detected", sb->s_id);
+		clear_buffer_write_io_error(sbh);
+		set_buffer_uptodate(sbh);
+	}
+}
+
 static void ext2_commit_super (struct super_block * sb,
 			       struct ext2_super_block * es)
 {
+	ext2_clear_super_error(sb);
 	es->s_wtime = cpu_to_le32(get_seconds());
 	mark_buffer_dirty(EXT2_SB(sb)->s_sbh);
 	sb->s_dirt = 0;
@@ -1099,6 +1136,7 @@ static void ext2_commit_super (struct super_block * sb,
 
 static void ext2_sync_super(struct super_block *sb, struct ext2_super_block *es)
 {
+	ext2_clear_super_error(sb);
 	es->s_free_blocks_count = cpu_to_le32(ext2_count_free_blocks(sb));
 	es->s_free_inodes_count = cpu_to_le32(ext2_count_free_inodes(sb));
 	es->s_wtime = cpu_to_le32(get_seconds());
@@ -1121,8 +1159,24 @@ static void ext2_sync_super(struct super_block *sb, struct ext2_super_block *es)
 static int ext2_sync_fs(struct super_block *sb, int wait)
 {
 	struct ext2_super_block *es = EXT2_SB(sb)->s_es;
+	struct buffer_head *sbh = EXT2_SB(sb)->s_sbh;
 
 	lock_kernel();
+	if (buffer_write_io_error(sbh)) {
+		/*
+		 * Oh, dear.  A previous attempt to write the
+		 * superblock failed.  This could happen because the
+		 * USB device was yanked out.  Or it could happen to
+		 * be a transient write error and maybe the block will
+		 * be remapped.  Nothing we can do but to retry the
+		 * write and hope for the best.
+		 */
+		ext2_msg(sb, KERN_ERR,
+		       "previous I/O error to superblock detected\n");
+		clear_buffer_write_io_error(sbh);
+		set_buffer_uptodate(sbh);
+	}
+
 	if (es->s_state & cpu_to_le16(EXT2_VALID_FS)) {
 		ext2_debug("setting valid to 0\n");
 		es->s_state &= cpu_to_le16(~EXT2_VALID_FS);
@@ -1170,7 +1224,7 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 	/*
 	 * Allow the "check" option to be passed as a remount option.
 	 */
-	if (!parse_options (data, sbi)) {
+	if (!parse_options(data, sb)) {
 		err = -EINVAL;
 		goto restore_opts;
 	}
@@ -1182,7 +1236,8 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 				    EXT2_MOUNT_XIP if not */
 
 	if ((ext2_use_xip(sb)) && (sb->s_blocksize != PAGE_SIZE)) {
-		printk("XIP: Unsupported blocksize\n");
+		ext2_msg(sb, KERN_WARNING,
+			"warning: unsupported blocksize for xip");
 		err = -EINVAL;
 		goto restore_opts;
 	}
@@ -1191,8 +1246,8 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 	if (((sbi->s_mount_opt & EXT2_MOUNT_XIP) !=
 	    (old_mount_opt & EXT2_MOUNT_XIP)) &&
 	    invalidate_inodes(sb)) {
-		ext2_warning(sb, __func__, "refusing change of xip flag "
-			     "with busy inodes while remounting");
+		ext2_msg(sb, KERN_WARNING, "warning: refusing change of "
+			 "xip flag with busy inodes while remounting");
 		sbi->s_mount_opt &= ~EXT2_MOUNT_XIP;
 		sbi->s_mount_opt |= old_mount_opt & EXT2_MOUNT_XIP;
 	}
@@ -1216,9 +1271,10 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data)
 		__le32 ret = EXT2_HAS_RO_COMPAT_FEATURE(sb,
 					       ~EXT2_FEATURE_RO_COMPAT_SUPP);
 		if (ret) {
-			printk("EXT2-fs: %s: couldn't remount RDWR because of "
-			       "unsupported optional features (%x).\n",
-			       sb->s_id, le32_to_cpu(ret));
+			ext2_msg(sb, KERN_WARNING,
+				"warning: couldn't remount RDWR because of "
+				"unsupported optional features (%x).",
+				le32_to_cpu(ret));
 			err = -EROFS;
 			goto restore_opts;
 		}
diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index 7913531ec6d5..904f00642f84 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -60,6 +60,7 @@
 #include <linux/mbcache.h>
 #include <linux/quotaops.h>
 #include <linux/rwsem.h>
+#include <linux/security.h>
 #include "ext2.h"
 #include "xattr.h"
 #include "acl.h"
@@ -249,8 +250,9 @@ cleanup:
  * used / required on success.
  */
 static int
-ext2_xattr_list(struct inode *inode, char *buffer, size_t buffer_size)
+ext2_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
+	struct inode *inode = dentry->d_inode;
 	struct buffer_head *bh = NULL;
 	struct ext2_xattr_entry *entry;
 	char *end;
@@ -300,9 +302,10 @@ bad_block:	ext2_error(inode->i_sb, "ext2_xattr_list",
 			ext2_xattr_handler(entry->e_name_index);
 
 		if (handler) {
-			size_t size = handler->list(inode, buffer, rest,
+			size_t size = handler->list(dentry, buffer, rest,
 						    entry->e_name,
-						    entry->e_name_len);
+						    entry->e_name_len,
+						    handler->flags);
 			if (buffer) {
 				if (size > rest) {
 					error = -ERANGE;
@@ -330,7 +333,7 @@ cleanup:
 ssize_t
 ext2_listxattr(struct dentry *dentry, char *buffer, size_t size)
 {
-	return ext2_xattr_list(dentry->d_inode, buffer, size);
+	return ext2_xattr_list(dentry, buffer, size);
 }
 
 /*
diff --git a/fs/ext2/xattr_security.c b/fs/ext2/xattr_security.c
index 70c0dbdcdcb7..c8155845ac05 100644
--- a/fs/ext2/xattr_security.c
+++ b/fs/ext2/xattr_security.c
@@ -11,8 +11,8 @@
 #include "xattr.h"
 
 static size_t
-ext2_xattr_security_list(struct inode *inode, char *list, size_t list_size,
-			 const char *name, size_t name_len)
+ext2_xattr_security_list(struct dentry *dentry, char *list, size_t list_size,
+			 const char *name, size_t name_len, int type)
 {
 	const int prefix_len = XATTR_SECURITY_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
@@ -26,22 +26,22 @@ ext2_xattr_security_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext2_xattr_security_get(struct inode *inode, const char *name,
-		       void *buffer, size_t size)
+ext2_xattr_security_get(struct dentry *dentry, const char *name,
+		       void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext2_xattr_get(inode, EXT2_XATTR_INDEX_SECURITY, name,
+	return ext2_xattr_get(dentry->d_inode, EXT2_XATTR_INDEX_SECURITY, name,
 			      buffer, size);
 }
 
 static int
-ext2_xattr_security_set(struct inode *inode, const char *name,
-		       const void *value, size_t size, int flags)
+ext2_xattr_security_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext2_xattr_set(inode, EXT2_XATTR_INDEX_SECURITY, name,
+	return ext2_xattr_set(dentry->d_inode, EXT2_XATTR_INDEX_SECURITY, name,
 			      value, size, flags);
 }
 
diff --git a/fs/ext2/xattr_trusted.c b/fs/ext2/xattr_trusted.c
index e8219f8eae9f..2a26d71f4771 100644
--- a/fs/ext2/xattr_trusted.c
+++ b/fs/ext2/xattr_trusted.c
@@ -13,8 +13,8 @@
 #include "xattr.h"
 
 static size_t
-ext2_xattr_trusted_list(struct inode *inode, char *list, size_t list_size,
-			const char *name, size_t name_len)
+ext2_xattr_trusted_list(struct dentry *dentry, char *list, size_t list_size,
+		const char *name, size_t name_len, int type)
 {
 	const int prefix_len = XATTR_TRUSTED_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
@@ -31,22 +31,22 @@ ext2_xattr_trusted_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext2_xattr_trusted_get(struct inode *inode, const char *name,
-		       void *buffer, size_t size)
+ext2_xattr_trusted_get(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext2_xattr_get(inode, EXT2_XATTR_INDEX_TRUSTED, name,
+	return ext2_xattr_get(dentry->d_inode, EXT2_XATTR_INDEX_TRUSTED, name,
 			      buffer, size);
 }
 
 static int
-ext2_xattr_trusted_set(struct inode *inode, const char *name,
-		       const void *value, size_t size, int flags)
+ext2_xattr_trusted_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext2_xattr_set(inode, EXT2_XATTR_INDEX_TRUSTED, name,
+	return ext2_xattr_set(dentry->d_inode, EXT2_XATTR_INDEX_TRUSTED, name,
 			      value, size, flags);
 }
 
diff --git a/fs/ext2/xattr_user.c b/fs/ext2/xattr_user.c
index 92495d28c62f..3f6caf3684b4 100644
--- a/fs/ext2/xattr_user.c
+++ b/fs/ext2/xattr_user.c
@@ -12,13 +12,13 @@
 #include "xattr.h"
 
 static size_t
-ext2_xattr_user_list(struct inode *inode, char *list, size_t list_size,
-		     const char *name, size_t name_len)
+ext2_xattr_user_list(struct dentry *dentry, char *list, size_t list_size,
+		const char *name, size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_USER_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return 0;
 
 	if (list && total_len <= list_size) {
@@ -30,27 +30,28 @@ ext2_xattr_user_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext2_xattr_user_get(struct inode *inode, const char *name,
-		    void *buffer, size_t size)
+ext2_xattr_user_get(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return -EOPNOTSUPP;
-	return ext2_xattr_get(inode, EXT2_XATTR_INDEX_USER, name, buffer, size);
+	return ext2_xattr_get(dentry->d_inode, EXT2_XATTR_INDEX_USER,
+			      name, buffer, size);
 }
 
 static int
-ext2_xattr_user_set(struct inode *inode, const char *name,
-		    const void *value, size_t size, int flags)
+ext2_xattr_user_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return -EOPNOTSUPP;
 
-	return ext2_xattr_set(inode, EXT2_XATTR_INDEX_USER, name,
-			      value, size, flags);
+	return ext2_xattr_set(dentry->d_inode, EXT2_XATTR_INDEX_USER,
+			      name, value, size, flags);
 }
 
 struct xattr_handler ext2_xattr_user_handler = {
diff --git a/fs/ext2/xip.c b/fs/ext2/xip.c
index c18fbf3e4068..322a56b2dfb1 100644
--- a/fs/ext2/xip.c
+++ b/fs/ext2/xip.c
@@ -69,8 +69,9 @@ void ext2_xip_verify_sb(struct super_block *sb)
 	if ((sbi->s_mount_opt & EXT2_MOUNT_XIP) &&
 	    !sb->s_bdev->bd_disk->fops->direct_access) {
 		sbi->s_mount_opt &= (~EXT2_MOUNT_XIP);
-		ext2_warning(sb, __func__,
-			     "ignoring xip option - not supported by bdev");
+		ext2_msg(sb, KERN_WARNING,
+			     "warning: ignoring xip option - "
+			     "not supported by bdev");
 	}
 }
 
diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index c9b0df376b5f..82ba34158661 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -366,12 +366,12 @@ out:
  * Extended attribute handlers
  */
 static size_t
-ext3_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len,
-			   const char *name, size_t name_len)
+ext3_xattr_list_acl_access(struct dentry *dentry, char *list, size_t list_len,
+			   const char *name, size_t name_len, int type)
 {
 	const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS);
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_len)
 		memcpy(list, POSIX_ACL_XATTR_ACCESS, size);
@@ -379,12 +379,12 @@ ext3_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len,
 }
 
 static size_t
-ext3_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len,
-			    const char *name, size_t name_len)
+ext3_xattr_list_acl_default(struct dentry *dentry, char *list, size_t list_len,
+			    const char *name, size_t name_len, int type)
 {
 	const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT);
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_len)
 		memcpy(list, POSIX_ACL_XATTR_DEFAULT, size);
@@ -392,15 +392,18 @@ ext3_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len,
 }
 
 static int
-ext3_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size)
+ext3_xattr_get_acl(struct dentry *dentry, const char *name, void *buffer,
+		   size_t size, int type)
 {
 	struct posix_acl *acl;
 	int error;
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return -EOPNOTSUPP;
 
-	acl = ext3_get_acl(inode, type);
+	acl = ext3_get_acl(dentry->d_inode, type);
 	if (IS_ERR(acl))
 		return PTR_ERR(acl);
 	if (acl == NULL)
@@ -412,31 +415,16 @@ ext3_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size)
 }
 
 static int
-ext3_xattr_get_acl_access(struct inode *inode, const char *name,
-			  void *buffer, size_t size)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext3_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size);
-}
-
-static int
-ext3_xattr_get_acl_default(struct inode *inode, const char *name,
-			   void *buffer, size_t size)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext3_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size);
-}
-
-static int
-ext3_xattr_set_acl(struct inode *inode, int type, const void *value,
-		   size_t size)
+ext3_xattr_set_acl(struct dentry *dentry, const char *name, const void *value,
+		   size_t size, int flags, int type)
 {
+	struct inode *inode = dentry->d_inode;
 	handle_t *handle;
 	struct posix_acl *acl;
 	int error, retries = 0;
 
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return -EOPNOTSUPP;
 	if (!is_owner_or_cap(inode))
@@ -468,34 +456,18 @@ release_and_out:
 	return error;
 }
 
-static int
-ext3_xattr_set_acl_access(struct inode *inode, const char *name,
-			  const void *value, size_t size, int flags)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext3_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size);
-}
-
-static int
-ext3_xattr_set_acl_default(struct inode *inode, const char *name,
-			   const void *value, size_t size, int flags)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext3_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size);
-}
-
 struct xattr_handler ext3_xattr_acl_access_handler = {
 	.prefix	= POSIX_ACL_XATTR_ACCESS,
+	.flags	= ACL_TYPE_ACCESS,
 	.list	= ext3_xattr_list_acl_access,
-	.get	= ext3_xattr_get_acl_access,
-	.set	= ext3_xattr_set_acl_access,
+	.get	= ext3_xattr_get_acl,
+	.set	= ext3_xattr_set_acl,
 };
 
 struct xattr_handler ext3_xattr_acl_default_handler = {
 	.prefix	= POSIX_ACL_XATTR_DEFAULT,
+	.flags	= ACL_TYPE_DEFAULT,
 	.list	= ext3_xattr_list_acl_default,
-	.get	= ext3_xattr_get_acl_default,
-	.set	= ext3_xattr_set_acl_default,
+	.get	= ext3_xattr_get_acl,
+	.set	= ext3_xattr_set_acl,
 };
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 2db957778903..455e6e6e5cb9 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -970,7 +970,7 @@ static int ext3_get_block(struct inode *inode, sector_t iblock,
 		if (max_blocks > DIO_MAX_BLOCKS)
 			max_blocks = DIO_MAX_BLOCKS;
 		handle = ext3_journal_start(inode, DIO_CREDITS +
-				2 * EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb));
+				EXT3_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb));
 		if (IS_ERR(handle)) {
 			ret = PTR_ERR(handle);
 			goto out;
@@ -1151,6 +1151,16 @@ static int do_journal_get_write_access(handle_t *handle,
 	return ext3_journal_get_write_access(handle, bh);
 }
 
+/*
+ * Truncate blocks that were not used by write. We have to truncate the
+ * pagecache as well so that corresponding buffers get properly unmapped.
+ */
+static void ext3_truncate_failed_write(struct inode *inode)
+{
+	truncate_inode_pages(inode->i_mapping, inode->i_size);
+	ext3_truncate(inode);
+}
+
 static int ext3_write_begin(struct file *file, struct address_space *mapping,
 				loff_t pos, unsigned len, unsigned flags,
 				struct page **pagep, void **fsdata)
@@ -1209,7 +1219,7 @@ write_begin_failed:
 		unlock_page(page);
 		page_cache_release(page);
 		if (pos + len > inode->i_size)
-			ext3_truncate(inode);
+			ext3_truncate_failed_write(inode);
 	}
 	if (ret == -ENOSPC && ext3_should_retry_alloc(inode->i_sb, &retries))
 		goto retry;
@@ -1304,7 +1314,7 @@ static int ext3_ordered_write_end(struct file *file,
 	page_cache_release(page);
 
 	if (pos + len > inode->i_size)
-		ext3_truncate(inode);
+		ext3_truncate_failed_write(inode);
 	return ret ? ret : copied;
 }
 
@@ -1330,7 +1340,7 @@ static int ext3_writeback_write_end(struct file *file,
 	page_cache_release(page);
 
 	if (pos + len > inode->i_size)
-		ext3_truncate(inode);
+		ext3_truncate_failed_write(inode);
 	return ret ? ret : copied;
 }
 
@@ -1383,7 +1393,7 @@ static int ext3_journalled_write_end(struct file *file,
 	page_cache_release(page);
 
 	if (pos + len > inode->i_size)
-		ext3_truncate(inode);
+		ext3_truncate_failed_write(inode);
 	return ret ? ret : copied;
 }
 
@@ -3136,8 +3146,8 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr)
 
 		/* (user+group)*(old+new) structure, inode write (sb,
 		 * inode block, ? - but truncate inode update has it) */
-		handle = ext3_journal_start(inode, 2*(EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)+
-					EXT3_QUOTA_DEL_BLOCKS(inode->i_sb))+3);
+		handle = ext3_journal_start(inode, EXT3_MAXQUOTAS_INIT_BLOCKS(inode->i_sb)+
+					EXT3_MAXQUOTAS_DEL_BLOCKS(inode->i_sb)+3);
 		if (IS_ERR(handle)) {
 			error = PTR_ERR(handle);
 			goto err_out;
@@ -3229,7 +3239,7 @@ static int ext3_writepage_trans_blocks(struct inode *inode)
 #ifdef CONFIG_QUOTA
 	/* We know that structure was already allocated during vfs_dq_init so
 	 * we will be updating only the data blocks + inodes */
-	ret += 2*EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb);
+	ret += EXT3_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb);
 #endif
 
 	return ret;
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index aad6400c9b77..7b0e44f7d66f 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -1699,7 +1699,7 @@ static int ext3_create (struct inode * dir, struct dentry * dentry, int mode,
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
+					EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -1733,7 +1733,7 @@ static int ext3_mknod (struct inode * dir, struct dentry *dentry,
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
+					EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -1769,7 +1769,7 @@ static int ext3_mkdir(struct inode * dir, struct dentry * dentry, int mode)
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
+					EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -1920,7 +1920,7 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode)
 	struct ext3_iloc iloc;
 	int err = 0, rc;
 
-	lock_super(sb);
+	mutex_lock(&EXT3_SB(sb)->s_orphan_lock);
 	if (!list_empty(&EXT3_I(inode)->i_orphan))
 		goto out_unlock;
 
@@ -1929,9 +1929,13 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode)
 
 	/* @@@ FIXME: Observation from aviro:
 	 * I think I can trigger J_ASSERT in ext3_orphan_add().  We block
-	 * here (on lock_super()), so race with ext3_link() which might bump
+	 * here (on s_orphan_lock), so race with ext3_link() which might bump
 	 * ->i_nlink. For, say it, character device. Not a regular file,
 	 * not a directory, not a symlink and ->i_nlink > 0.
+	 *
+	 * tytso, 4/25/2009: I'm not sure how that could happen;
+	 * shouldn't the fs core protect us from these sort of
+	 * unlink()/link() races?
 	 */
 	J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
 		S_ISLNK(inode->i_mode)) || inode->i_nlink == 0);
@@ -1968,7 +1972,7 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode)
 	jbd_debug(4, "orphan inode %lu will point to %d\n",
 			inode->i_ino, NEXT_ORPHAN(inode));
 out_unlock:
-	unlock_super(sb);
+	mutex_unlock(&EXT3_SB(sb)->s_orphan_lock);
 	ext3_std_error(inode->i_sb, err);
 	return err;
 }
@@ -1986,11 +1990,9 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode)
 	struct ext3_iloc iloc;
 	int err = 0;
 
-	lock_super(inode->i_sb);
-	if (list_empty(&ei->i_orphan)) {
-		unlock_super(inode->i_sb);
-		return 0;
-	}
+	mutex_lock(&EXT3_SB(inode->i_sb)->s_orphan_lock);
+	if (list_empty(&ei->i_orphan))
+		goto out;
 
 	ino_next = NEXT_ORPHAN(inode);
 	prev = ei->i_orphan.prev;
@@ -2040,7 +2042,7 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode)
 out_err:
 	ext3_std_error(inode->i_sb, err);
 out:
-	unlock_super(inode->i_sb);
+	mutex_unlock(&EXT3_SB(inode->i_sb)->s_orphan_lock);
 	return err;
 
 out_brelse:
@@ -2175,7 +2177,7 @@ static int ext3_symlink (struct inode * dir,
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 +
-					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
+					EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c
index 8359e7b3dc89..54351ac7cef9 100644
--- a/fs/ext3/resize.c
+++ b/fs/ext3/resize.c
@@ -209,7 +209,7 @@ static int setup_new_group_blocks(struct super_block *sb,
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
-	lock_super(sb);
+	mutex_lock(&sbi->s_resize_lock);
 	if (input->group != sbi->s_groups_count) {
 		err = -EBUSY;
 		goto exit_journal;
@@ -266,7 +266,7 @@ static int setup_new_group_blocks(struct super_block *sb,
 			goto exit_bh;
 
 		if (IS_ERR(gdb = bclean(handle, sb, block))) {
-			err = PTR_ERR(bh);
+			err = PTR_ERR(gdb);
 			goto exit_bh;
 		}
 		ext3_journal_dirty_metadata(handle, gdb);
@@ -324,7 +324,7 @@ exit_bh:
 	brelse(bh);
 
 exit_journal:
-	unlock_super(sb);
+	mutex_unlock(&sbi->s_resize_lock);
 	if ((err2 = ext3_journal_stop(handle)) && !err)
 		err = err2;
 
@@ -662,11 +662,12 @@ exit_free:
  * important part is that the new block and inode counts are in the backup
  * superblocks, and the location of the new group metadata in the GDT backups.
  *
- * We do not need lock_super() for this, because these blocks are not
- * otherwise touched by the filesystem code when it is mounted.  We don't
- * need to worry about last changing from sbi->s_groups_count, because the
- * worst that can happen is that we do not copy the full number of backups
- * at this time.  The resize which changed s_groups_count will backup again.
+ * We do not need take the s_resize_lock for this, because these
+ * blocks are not otherwise touched by the filesystem code when it is
+ * mounted.  We don't need to worry about last changing from
+ * sbi->s_groups_count, because the worst that can happen is that we
+ * do not copy the full number of backups at this time.  The resize
+ * which changed s_groups_count will backup again.
  */
 static void update_backups(struct super_block *sb,
 			   int blk_off, char *data, int size)
@@ -825,7 +826,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input)
 		goto exit_put;
 	}
 
-	lock_super(sb);
+	mutex_lock(&sbi->s_resize_lock);
 	if (input->group != sbi->s_groups_count) {
 		ext3_warning(sb, __func__,
 			     "multiple resizers run on filesystem!");
@@ -856,7 +857,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input)
 	/*
 	 * OK, now we've set up the new group.  Time to make it active.
 	 *
-	 * Current kernels don't lock all allocations via lock_super(),
+	 * We do not lock all allocations via s_resize_lock
 	 * so we have to be safe wrt. concurrent accesses the group
 	 * data.  So we need to be careful to set all of the relevant
 	 * group descriptor data etc. *before* we enable the group.
@@ -900,12 +901,12 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input)
 	 *
 	 * The precise rules we use are:
 	 *
-	 * * Writers of s_groups_count *must* hold lock_super
+	 * * Writers of s_groups_count *must* hold s_resize_lock
 	 * AND
 	 * * Writers must perform a smp_wmb() after updating all dependent
 	 *   data and before modifying the groups count
 	 *
-	 * * Readers must hold lock_super() over the access
+	 * * Readers must hold s_resize_lock over the access
 	 * OR
 	 * * Readers must perform an smp_rmb() after reading the groups count
 	 *   and before reading any dependent data.
@@ -936,7 +937,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input)
 	ext3_journal_dirty_metadata(handle, sbi->s_sbh);
 
 exit_journal:
-	unlock_super(sb);
+	mutex_unlock(&sbi->s_resize_lock);
 	if ((err2 = ext3_journal_stop(handle)) && !err)
 		err = err2;
 	if (!err) {
@@ -973,7 +974,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es,
 
 	/* We don't need to worry about locking wrt other resizers just
 	 * yet: we're going to revalidate es->s_blocks_count after
-	 * taking lock_super() below. */
+	 * taking the s_resize_lock below. */
 	o_blocks_count = le32_to_cpu(es->s_blocks_count);
 	o_groups_count = EXT3_SB(sb)->s_groups_count;
 
@@ -1045,11 +1046,11 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es,
 		goto exit_put;
 	}
 
-	lock_super(sb);
+	mutex_lock(&EXT3_SB(sb)->s_resize_lock);
 	if (o_blocks_count != le32_to_cpu(es->s_blocks_count)) {
 		ext3_warning(sb, __func__,
 			     "multiple resizers run on filesystem!");
-		unlock_super(sb);
+		mutex_unlock(&EXT3_SB(sb)->s_resize_lock);
 		ext3_journal_stop(handle);
 		err = -EBUSY;
 		goto exit_put;
@@ -1059,13 +1060,13 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es,
 						 EXT3_SB(sb)->s_sbh))) {
 		ext3_warning(sb, __func__,
 			     "error %d on journal write access", err);
-		unlock_super(sb);
+		mutex_unlock(&EXT3_SB(sb)->s_resize_lock);
 		ext3_journal_stop(handle);
 		goto exit_put;
 	}
 	es->s_blocks_count = cpu_to_le32(o_blocks_count + add);
 	ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh);
-	unlock_super(sb);
+	mutex_unlock(&EXT3_SB(sb)->s_resize_lock);
 	ext3_debug("freeing blocks %lu through "E3FSBLK"\n", o_blocks_count,
 		   o_blocks_count + add);
 	ext3_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 427496c4767c..afa2b569da10 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -135,12 +135,24 @@ void ext3_journal_abort_handle(const char *caller, const char *err_fn,
 	if (is_handle_aborted(handle))
 		return;
 
-	printk(KERN_ERR "%s: aborting transaction: %s in %s\n",
-	       caller, errstr, err_fn);
+	printk(KERN_ERR "EXT3-fs: %s: aborting transaction: %s in %s\n",
+		caller, errstr, err_fn);
 
 	journal_abort_handle(handle);
 }
 
+void ext3_msg(struct super_block *sb, const char *prefix,
+		const char *fmt, ...)
+{
+	va_list args;
+
+	va_start(args, fmt);
+	printk("%sEXT3-fs (%s): ", prefix, sb->s_id);
+	vprintk(fmt, args);
+	printk("\n");
+	va_end(args);
+}
+
 /* Deal with the reporting of failure conditions on a filesystem such as
  * inconsistencies detected or read IO failures.
  *
@@ -174,12 +186,13 @@ static void ext3_handle_error(struct super_block *sb)
 			journal_abort(journal, -EIO);
 	}
 	if (test_opt (sb, ERRORS_RO)) {
-		printk (KERN_CRIT "Remounting filesystem read-only\n");
+		ext3_msg(sb, KERN_CRIT,
+			"error: remounting filesystem read-only");
 		sb->s_flags |= MS_RDONLY;
 	}
 	ext3_commit_super(sb, es, 1);
 	if (test_opt(sb, ERRORS_PANIC))
-		panic("EXT3-fs (device %s): panic forced after error\n",
+		panic("EXT3-fs (%s): panic forced after error\n",
 			sb->s_id);
 }
 
@@ -247,8 +260,7 @@ void __ext3_std_error (struct super_block * sb, const char * function,
 		return;
 
 	errstr = ext3_decode_error(sb, errno, nbuf);
-	printk (KERN_CRIT "EXT3-fs error (device %s) in %s: %s\n",
-		sb->s_id, function, errstr);
+	ext3_msg(sb, KERN_CRIT, "error in %s: %s", function, errstr);
 
 	ext3_handle_error(sb);
 }
@@ -268,21 +280,20 @@ void ext3_abort (struct super_block * sb, const char * function,
 {
 	va_list args;
 
-	printk (KERN_CRIT "ext3_abort called.\n");
-
 	va_start(args, fmt);
-	printk(KERN_CRIT "EXT3-fs error (device %s): %s: ",sb->s_id, function);
+	printk(KERN_CRIT "EXT3-fs (%s): error: %s: ", sb->s_id, function);
 	vprintk(fmt, args);
 	printk("\n");
 	va_end(args);
 
 	if (test_opt(sb, ERRORS_PANIC))
-		panic("EXT3-fs panic from previous error\n");
+		panic("EXT3-fs: panic from previous error\n");
 
 	if (sb->s_flags & MS_RDONLY)
 		return;
 
-	printk(KERN_CRIT "Remounting filesystem read-only\n");
+	ext3_msg(sb, KERN_CRIT,
+		"error: remounting filesystem read-only");
 	EXT3_SB(sb)->s_mount_state |= EXT3_ERROR_FS;
 	sb->s_flags |= MS_RDONLY;
 	EXT3_SB(sb)->s_mount_opt |= EXT3_MOUNT_ABORT;
@@ -296,7 +307,7 @@ void ext3_warning (struct super_block * sb, const char * function,
 	va_list args;
 
 	va_start(args, fmt);
-	printk(KERN_WARNING "EXT3-fs warning (device %s): %s: ",
+	printk(KERN_WARNING "EXT3-fs (%s): warning: %s: ",
 	       sb->s_id, function);
 	vprintk(fmt, args);
 	printk("\n");
@@ -310,10 +321,10 @@ void ext3_update_dynamic_rev(struct super_block *sb)
 	if (le32_to_cpu(es->s_rev_level) > EXT3_GOOD_OLD_REV)
 		return;
 
-	ext3_warning(sb, __func__,
-		     "updating to rev %d because of new feature flag, "
-		     "running e2fsck is recommended",
-		     EXT3_DYNAMIC_REV);
+	ext3_msg(sb, KERN_WARNING,
+		"warning: updating to rev %d because of "
+		"new feature flag, running e2fsck is recommended",
+		EXT3_DYNAMIC_REV);
 
 	es->s_first_ino = cpu_to_le32(EXT3_GOOD_OLD_FIRST_INO);
 	es->s_inode_size = cpu_to_le16(EXT3_GOOD_OLD_INODE_SIZE);
@@ -331,7 +342,7 @@ void ext3_update_dynamic_rev(struct super_block *sb)
 /*
  * Open the external journal device
  */
-static struct block_device *ext3_blkdev_get(dev_t dev)
+static struct block_device *ext3_blkdev_get(dev_t dev, struct super_block *sb)
 {
 	struct block_device *bdev;
 	char b[BDEVNAME_SIZE];
@@ -342,8 +353,9 @@ static struct block_device *ext3_blkdev_get(dev_t dev)
 	return bdev;
 
 fail:
-	printk(KERN_ERR "EXT3: failed to open journal device %s: %ld\n",
-			__bdevname(dev, b), PTR_ERR(bdev));
+	ext3_msg(sb, "error: failed to open journal device %s: %ld",
+		__bdevname(dev, b), PTR_ERR(bdev));
+
 	return NULL;
 }
 
@@ -378,13 +390,13 @@ static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi)
 {
 	struct list_head *l;
 
-	printk(KERN_ERR "sb orphan head is %d\n",
+	ext3_msg(sb, KERN_ERR, "error: sb orphan head is %d",
 	       le32_to_cpu(sbi->s_es->s_last_orphan));
 
-	printk(KERN_ERR "sb_info orphan list:\n");
+	ext3_msg(sb, KERN_ERR, "sb_info orphan list:");
 	list_for_each(l, &sbi->s_orphan) {
 		struct inode *inode = orphan_list_entry(l);
-		printk(KERN_ERR "  "
+		ext3_msg(sb, KERN_ERR, "  "
 		       "inode %s:%lu at %p: mode %o, nlink %d, next %d\n",
 		       inode->i_sb->s_id, inode->i_ino, inode,
 		       inode->i_mode, inode->i_nlink,
@@ -527,9 +539,22 @@ static inline void ext3_show_quota_options(struct seq_file *seq, struct super_bl
 #if defined(CONFIG_QUOTA)
 	struct ext3_sb_info *sbi = EXT3_SB(sb);
 
-	if (sbi->s_jquota_fmt)
-		seq_printf(seq, ",jqfmt=%s",
-		(sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0");
+	if (sbi->s_jquota_fmt) {
+		char *fmtname = "";
+
+		switch (sbi->s_jquota_fmt) {
+		case QFMT_VFS_OLD:
+			fmtname = "vfsold";
+			break;
+		case QFMT_VFS_V0:
+			fmtname = "vfsv0";
+			break;
+		case QFMT_VFS_V1:
+			fmtname = "vfsv1";
+			break;
+		}
+		seq_printf(seq, ",jqfmt=%s", fmtname);
+	}
 
 	if (sbi->s_qf_names[USRQUOTA])
 		seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]);
@@ -636,6 +661,9 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs)
 	if (test_opt(sb, DATA_ERR_ABORT))
 		seq_puts(seq, ",data_err=abort");
 
+	if (test_opt(sb, NOLOAD))
+		seq_puts(seq, ",norecovery");
+
 	ext3_show_quota_options(seq, sb);
 
 	return 0;
@@ -787,9 +815,9 @@ enum {
 	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
 	Opt_data_err_abort, Opt_data_err_ignore,
 	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
-	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
-	Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota,
-	Opt_grpquota
+	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota,
+	Opt_noquota, Opt_ignore, Opt_barrier, Opt_err, Opt_resize,
+	Opt_usrquota, Opt_grpquota
 };
 
 static const match_table_t tokens = {
@@ -818,6 +846,7 @@ static const match_table_t tokens = {
 	{Opt_reservation, "reservation"},
 	{Opt_noreservation, "noreservation"},
 	{Opt_noload, "noload"},
+	{Opt_noload, "norecovery"},
 	{Opt_nobh, "nobh"},
 	{Opt_bh, "bh"},
 	{Opt_commit, "commit=%u"},
@@ -836,6 +865,7 @@ static const match_table_t tokens = {
 	{Opt_grpjquota, "grpjquota=%s"},
 	{Opt_jqfmt_vfsold, "jqfmt=vfsold"},
 	{Opt_jqfmt_vfsv0, "jqfmt=vfsv0"},
+	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
 	{Opt_grpquota, "grpquota"},
 	{Opt_noquota, "noquota"},
 	{Opt_quota, "quota"},
@@ -845,7 +875,7 @@ static const match_table_t tokens = {
 	{Opt_err, NULL},
 };
 
-static ext3_fsblk_t get_sb_block(void **data)
+static ext3_fsblk_t get_sb_block(void **data, struct super_block *sb)
 {
 	ext3_fsblk_t	sb_block;
 	char		*options = (char *) *data;
@@ -856,7 +886,7 @@ static ext3_fsblk_t get_sb_block(void **data)
 	/*todo: use simple_strtoll with >32bit ext3 */
 	sb_block = simple_strtoul(options, &options, 0);
 	if (*options && *options != ',') {
-		printk("EXT3-fs: Invalid sb specification: %s\n",
+		ext3_msg(sb, "error: invalid sb specification: %s",
 		       (char *) *data);
 		return 1;
 	}
@@ -956,7 +986,8 @@ static int parse_options (char *options, struct super_block *sb,
 #else
 		case Opt_user_xattr:
 		case Opt_nouser_xattr:
-			printk("EXT3 (no)user_xattr options not supported\n");
+			ext3_msg(sb, KERN_INFO,
+				"(no)user_xattr options not supported");
 			break;
 #endif
 #ifdef CONFIG_EXT3_FS_POSIX_ACL
@@ -969,7 +1000,8 @@ static int parse_options (char *options, struct super_block *sb,
 #else
 		case Opt_acl:
 		case Opt_noacl:
-			printk("EXT3 (no)acl options not supported\n");
+			ext3_msg(sb, KERN_INFO,
+				"(no)acl options not supported");
 			break;
 #endif
 		case Opt_reservation:
@@ -985,16 +1017,16 @@ static int parse_options (char *options, struct super_block *sb,
 			   user to specify an existing inode to be the
 			   journal file. */
 			if (is_remount) {
-				printk(KERN_ERR "EXT3-fs: cannot specify "
-				       "journal on remount\n");
+				ext3_msg(sb, KERN_ERR, "error: cannot specify "
+					"journal on remount");
 				return 0;
 			}
 			set_opt (sbi->s_mount_opt, UPDATE_JOURNAL);
 			break;
 		case Opt_journal_inum:
 			if (is_remount) {
-				printk(KERN_ERR "EXT3-fs: cannot specify "
-				       "journal on remount\n");
+				ext3_msg(sb, KERN_ERR, "error: cannot specify "
+				       "journal on remount");
 				return 0;
 			}
 			if (match_int(&args[0], &option))
@@ -1003,8 +1035,8 @@ static int parse_options (char *options, struct super_block *sb,
 			break;
 		case Opt_journal_dev:
 			if (is_remount) {
-				printk(KERN_ERR "EXT3-fs: cannot specify "
-				       "journal on remount\n");
+				ext3_msg(sb, KERN_ERR, "error: cannot specify "
+				       "journal on remount");
 				return 0;
 			}
 			if (match_int(&args[0], &option))
@@ -1036,12 +1068,11 @@ static int parse_options (char *options, struct super_block *sb,
 				if ((sbi->s_mount_opt & EXT3_MOUNT_DATA_FLAGS)
 						== data_opt)
 					break;
-				printk(KERN_ERR
-					"EXT3-fs (device %s): Cannot change "
+				ext3_msg(sb, KERN_ERR,
+					"error: cannot change "
 					"data mode on remount. The filesystem "
 					"is mounted in data=%s mode and you "
-					"try to remount it in data=%s mode.\n",
-					sb->s_id,
+					"try to remount it in data=%s mode.",
 					data_mode_string(sbi->s_mount_opt &
 							EXT3_MOUNT_DATA_FLAGS),
 					data_mode_string(data_opt));
@@ -1066,31 +1097,31 @@ static int parse_options (char *options, struct super_block *sb,
 set_qf_name:
 			if (sb_any_quota_loaded(sb) &&
 			    !sbi->s_qf_names[qtype]) {
-				printk(KERN_ERR
-					"EXT3-fs: Cannot change journaled "
-					"quota options when quota turned on.\n");
+				ext3_msg(sb, KERN_ERR,
+					"error: cannot change journaled "
+					"quota options when quota turned on.");
 				return 0;
 			}
 			qname = match_strdup(&args[0]);
 			if (!qname) {
-				printk(KERN_ERR
-					"EXT3-fs: not enough memory for "
-					"storing quotafile name.\n");
+				ext3_msg(sb, KERN_ERR,
+					"error: not enough memory for "
+					"storing quotafile name.");
 				return 0;
 			}
 			if (sbi->s_qf_names[qtype] &&
 			    strcmp(sbi->s_qf_names[qtype], qname)) {
-				printk(KERN_ERR
-					"EXT3-fs: %s quota file already "
-					"specified.\n", QTYPE2NAME(qtype));
+				ext3_msg(sb, KERN_ERR,
+					"error: %s quota file already "
+					"specified.", QTYPE2NAME(qtype));
 				kfree(qname);
 				return 0;
 			}
 			sbi->s_qf_names[qtype] = qname;
 			if (strchr(sbi->s_qf_names[qtype], '/')) {
-				printk(KERN_ERR
-					"EXT3-fs: quotafile must be on "
-					"filesystem root.\n");
+				ext3_msg(sb, KERN_ERR,
+					"error: quotafile must be on "
+					"filesystem root.");
 				kfree(sbi->s_qf_names[qtype]);
 				sbi->s_qf_names[qtype] = NULL;
 				return 0;
@@ -1105,9 +1136,9 @@ set_qf_name:
 clear_qf_name:
 			if (sb_any_quota_loaded(sb) &&
 			    sbi->s_qf_names[qtype]) {
-				printk(KERN_ERR "EXT3-fs: Cannot change "
+				ext3_msg(sb, KERN_ERR, "error: cannot change "
 					"journaled quota options when "
-					"quota turned on.\n");
+					"quota turned on.");
 				return 0;
 			}
 			/*
@@ -1121,12 +1152,15 @@ clear_qf_name:
 			goto set_qf_format;
 		case Opt_jqfmt_vfsv0:
 			qfmt = QFMT_VFS_V0;
+			goto set_qf_format;
+		case Opt_jqfmt_vfsv1:
+			qfmt = QFMT_VFS_V1;
 set_qf_format:
 			if (sb_any_quota_loaded(sb) &&
 			    sbi->s_jquota_fmt != qfmt) {
-				printk(KERN_ERR "EXT3-fs: Cannot change "
+				ext3_msg(sb, KERN_ERR, "error: cannot change "
 					"journaled quota options when "
-					"quota turned on.\n");
+					"quota turned on.");
 				return 0;
 			}
 			sbi->s_jquota_fmt = qfmt;
@@ -1142,8 +1176,8 @@ set_qf_format:
 			break;
 		case Opt_noquota:
 			if (sb_any_quota_loaded(sb)) {
-				printk(KERN_ERR "EXT3-fs: Cannot change quota "
-					"options when quota turned on.\n");
+				ext3_msg(sb, KERN_ERR, "error: cannot change "
+					"quota options when quota turned on.");
 				return 0;
 			}
 			clear_opt(sbi->s_mount_opt, QUOTA);
@@ -1154,8 +1188,8 @@ set_qf_format:
 		case Opt_quota:
 		case Opt_usrquota:
 		case Opt_grpquota:
-			printk(KERN_ERR
-				"EXT3-fs: quota options not supported.\n");
+			ext3_msg(sb, KERN_ERR,
+				"error: quota options not supported.");
 			break;
 		case Opt_usrjquota:
 		case Opt_grpjquota:
@@ -1163,9 +1197,10 @@ set_qf_format:
 		case Opt_offgrpjquota:
 		case Opt_jqfmt_vfsold:
 		case Opt_jqfmt_vfsv0:
-			printk(KERN_ERR
-				"EXT3-fs: journaled quota options not "
-				"supported.\n");
+		case Opt_jqfmt_vfsv1:
+			ext3_msg(sb, KERN_ERR,
+				"error: journaled quota options not "
+				"supported.");
 			break;
 		case Opt_noquota:
 			break;
@@ -1185,8 +1220,9 @@ set_qf_format:
 			break;
 		case Opt_resize:
 			if (!is_remount) {
-				printk("EXT3-fs: resize option only available "
-					"for remount\n");
+				ext3_msg(sb, KERN_ERR,
+					"error: resize option only available "
+					"for remount");
 				return 0;
 			}
 			if (match_int(&args[0], &option) != 0)
@@ -1200,9 +1236,9 @@ set_qf_format:
 			clear_opt(sbi->s_mount_opt, NOBH);
 			break;
 		default:
-			printk (KERN_ERR
-				"EXT3-fs: Unrecognized mount option \"%s\" "
-				"or missing value\n", p);
+			ext3_msg(sb, KERN_ERR,
+				"error: unrecognized mount option \"%s\" "
+				"or missing value", p);
 			return 0;
 		}
 	}
@@ -1220,21 +1256,21 @@ set_qf_format:
 				(sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA)) ||
 		    (sbi->s_qf_names[GRPQUOTA] &&
 				(sbi->s_mount_opt & EXT3_MOUNT_USRQUOTA))) {
-			printk(KERN_ERR "EXT3-fs: old and new quota "
-					"format mixing.\n");
+			ext3_msg(sb, KERN_ERR, "error: old and new quota "
+					"format mixing.");
 			return 0;
 		}
 
 		if (!sbi->s_jquota_fmt) {
-			printk(KERN_ERR "EXT3-fs: journaled quota format "
-					"not specified.\n");
+			ext3_msg(sb, KERN_ERR, "error: journaled quota format "
+					"not specified.");
 			return 0;
 		}
 	} else {
 		if (sbi->s_jquota_fmt) {
-			printk(KERN_ERR "EXT3-fs: journaled quota format "
+			ext3_msg(sb, KERN_ERR, "error: journaled quota format "
 					"specified with no journaling "
-					"enabled.\n");
+					"enabled.");
 			return 0;
 		}
 	}
@@ -1249,31 +1285,33 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es,
 	int res = 0;
 
 	if (le32_to_cpu(es->s_rev_level) > EXT3_MAX_SUPP_REV) {
-		printk (KERN_ERR "EXT3-fs warning: revision level too high, "
-			"forcing read-only mode\n");
+		ext3_msg(sb, KERN_ERR,
+			"error: revision level too high, "
+			"forcing read-only mode");
 		res = MS_RDONLY;
 	}
 	if (read_only)
 		return res;
 	if (!(sbi->s_mount_state & EXT3_VALID_FS))
-		printk (KERN_WARNING "EXT3-fs warning: mounting unchecked fs, "
-			"running e2fsck is recommended\n");
+		ext3_msg(sb, KERN_WARNING,
+			"warning: mounting unchecked fs, "
+			"running e2fsck is recommended");
 	else if ((sbi->s_mount_state & EXT3_ERROR_FS))
-		printk (KERN_WARNING
-			"EXT3-fs warning: mounting fs with errors, "
-			"running e2fsck is recommended\n");
+		ext3_msg(sb, KERN_WARNING,
+			"warning: mounting fs with errors, "
+			"running e2fsck is recommended");
 	else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 &&
 		 le16_to_cpu(es->s_mnt_count) >=
 		 (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count))
-		printk (KERN_WARNING
-			"EXT3-fs warning: maximal mount count reached, "
-			"running e2fsck is recommended\n");
+		ext3_msg(sb, KERN_WARNING,
+			"warning: maximal mount count reached, "
+			"running e2fsck is recommended");
 	else if (le32_to_cpu(es->s_checkinterval) &&
 		(le32_to_cpu(es->s_lastcheck) +
 			le32_to_cpu(es->s_checkinterval) <= get_seconds()))
-		printk (KERN_WARNING
-			"EXT3-fs warning: checktime reached, "
-			"running e2fsck is recommended\n");
+		ext3_msg(sb, KERN_WARNING,
+			"warning: checktime reached, "
+			"running e2fsck is recommended");
 #if 0
 		/* @@@ We _will_ want to clear the valid bit if we find
                    inconsistencies, to force a fsck at reboot.  But for
@@ -1290,22 +1328,20 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es,
 
 	ext3_commit_super(sb, es, 1);
 	if (test_opt(sb, DEBUG))
-		printk(KERN_INFO "[EXT3 FS bs=%lu, gc=%lu, "
-				"bpg=%lu, ipg=%lu, mo=%04lx]\n",
+		ext3_msg(sb, KERN_INFO, "[bs=%lu, gc=%lu, "
+				"bpg=%lu, ipg=%lu, mo=%04lx]",
 			sb->s_blocksize,
 			sbi->s_groups_count,
 			EXT3_BLOCKS_PER_GROUP(sb),
 			EXT3_INODES_PER_GROUP(sb),
 			sbi->s_mount_opt);
 
-	printk(KERN_INFO "EXT3 FS on %s, ", sb->s_id);
 	if (EXT3_SB(sb)->s_journal->j_inode == NULL) {
 		char b[BDEVNAME_SIZE];
-
-		printk("external journal on %s\n",
+		ext3_msg(sb, KERN_INFO, "using external journal on %s",
 			bdevname(EXT3_SB(sb)->s_journal->j_dev, b));
 	} else {
-		printk("internal journal\n");
+		ext3_msg(sb, KERN_INFO, "using internal journal");
 	}
 	return res;
 }
@@ -1399,8 +1435,8 @@ static void ext3_orphan_cleanup (struct super_block * sb,
 	}
 
 	if (bdev_read_only(sb->s_bdev)) {
-		printk(KERN_ERR "EXT3-fs: write access "
-			"unavailable, skipping orphan cleanup.\n");
+		ext3_msg(sb, KERN_ERR, "error: write access "
+			"unavailable, skipping orphan cleanup.");
 		return;
 	}
 
@@ -1414,8 +1450,7 @@ static void ext3_orphan_cleanup (struct super_block * sb,
 	}
 
 	if (s_flags & MS_RDONLY) {
-		printk(KERN_INFO "EXT3-fs: %s: orphan cleanup on readonly fs\n",
-		       sb->s_id);
+		ext3_msg(sb, KERN_INFO, "orphan cleanup on readonly fs");
 		sb->s_flags &= ~MS_RDONLY;
 	}
 #ifdef CONFIG_QUOTA
@@ -1426,9 +1461,9 @@ static void ext3_orphan_cleanup (struct super_block * sb,
 		if (EXT3_SB(sb)->s_qf_names[i]) {
 			int ret = ext3_quota_on_mount(sb, i);
 			if (ret < 0)
-				printk(KERN_ERR
-					"EXT3-fs: Cannot turn on journaled "
-					"quota: error %d\n", ret);
+				ext3_msg(sb, KERN_ERR,
+					"error: cannot turn on journaled "
+					"quota: %d", ret);
 		}
 	}
 #endif
@@ -1466,11 +1501,11 @@ static void ext3_orphan_cleanup (struct super_block * sb,
 #define PLURAL(x) (x), ((x)==1) ? "" : "s"
 
 	if (nr_orphans)
-		printk(KERN_INFO "EXT3-fs: %s: %d orphan inode%s deleted\n",
-		       sb->s_id, PLURAL(nr_orphans));
+		ext3_msg(sb, KERN_INFO, "%d orphan inode%s deleted",
+		       PLURAL(nr_orphans));
 	if (nr_truncates)
-		printk(KERN_INFO "EXT3-fs: %s: %d truncate%s cleaned up\n",
-		       sb->s_id, PLURAL(nr_truncates));
+		ext3_msg(sb, KERN_INFO, "%d truncate%s cleaned up",
+		       PLURAL(nr_truncates));
 #ifdef CONFIG_QUOTA
 	/* Turn quotas off */
 	for (i = 0; i < MAXQUOTAS; i++) {
@@ -1554,7 +1589,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	struct ext3_super_block *es = NULL;
 	struct ext3_sb_info *sbi;
 	ext3_fsblk_t block;
-	ext3_fsblk_t sb_block = get_sb_block(&data);
+	ext3_fsblk_t sb_block = get_sb_block(&data, sb);
 	ext3_fsblk_t logic_sb_block;
 	unsigned long offset = 0;
 	unsigned int journal_inum = 0;
@@ -1590,7 +1625,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
 	blocksize = sb_min_blocksize(sb, EXT3_MIN_BLOCK_SIZE);
 	if (!blocksize) {
-		printk(KERN_ERR "EXT3-fs: unable to set blocksize\n");
+		ext3_msg(sb, KERN_ERR, "error: unable to set blocksize");
 		goto out_fail;
 	}
 
@@ -1606,7 +1641,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	}
 
 	if (!(bh = sb_bread(sb, logic_sb_block))) {
-		printk (KERN_ERR "EXT3-fs: unable to read superblock\n");
+		ext3_msg(sb, KERN_ERR, "error: unable to read superblock");
 		goto out_fail;
 	}
 	/*
@@ -1665,9 +1700,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	    (EXT3_HAS_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT3_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT3_HAS_INCOMPAT_FEATURE(sb, ~0U)))
-		printk(KERN_WARNING
-		       "EXT3-fs warning: feature flags set on rev 0 fs, "
-		       "running e2fsck is recommended\n");
+		ext3_msg(sb, KERN_WARNING,
+			"warning: feature flags set on rev 0 fs, "
+			"running e2fsck is recommended");
 	/*
 	 * Check feature flags regardless of the revision level, since we
 	 * previously didn't change the revision level when setting the flags,
@@ -1675,25 +1710,25 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	 */
 	features = EXT3_HAS_INCOMPAT_FEATURE(sb, ~EXT3_FEATURE_INCOMPAT_SUPP);
 	if (features) {
-		printk(KERN_ERR "EXT3-fs: %s: couldn't mount because of "
-		       "unsupported optional features (%x).\n",
-		       sb->s_id, le32_to_cpu(features));
+		ext3_msg(sb, KERN_ERR,
+			"error: couldn't mount because of unsupported "
+			"optional features (%x)", le32_to_cpu(features));
 		goto failed_mount;
 	}
 	features = EXT3_HAS_RO_COMPAT_FEATURE(sb, ~EXT3_FEATURE_RO_COMPAT_SUPP);
 	if (!(sb->s_flags & MS_RDONLY) && features) {
-		printk(KERN_ERR "EXT3-fs: %s: couldn't mount RDWR because of "
-		       "unsupported optional features (%x).\n",
-		       sb->s_id, le32_to_cpu(features));
+		ext3_msg(sb, KERN_ERR,
+			"error: couldn't mount RDWR because of unsupported "
+			"optional features (%x)", le32_to_cpu(features));
 		goto failed_mount;
 	}
 	blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size);
 
 	if (blocksize < EXT3_MIN_BLOCK_SIZE ||
 	    blocksize > EXT3_MAX_BLOCK_SIZE) {
-		printk(KERN_ERR
-		       "EXT3-fs: Unsupported filesystem blocksize %d on %s.\n",
-		       blocksize, sb->s_id);
+		ext3_msg(sb, KERN_ERR,
+			"error: couldn't mount because of unsupported "
+			"filesystem blocksize %d", blocksize);
 		goto failed_mount;
 	}
 
@@ -1704,30 +1739,31 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 		 * than the hardware sectorsize for the machine.
 		 */
 		if (blocksize < hblock) {
-			printk(KERN_ERR "EXT3-fs: blocksize %d too small for "
-			       "device blocksize %d.\n", blocksize, hblock);
+			ext3_msg(sb, KERN_ERR,
+				"error: fsblocksize %d too small for "
+				"hardware sectorsize %d", blocksize, hblock);
 			goto failed_mount;
 		}
 
 		brelse (bh);
 		if (!sb_set_blocksize(sb, blocksize)) {
-			printk(KERN_ERR "EXT3-fs: bad blocksize %d.\n",
-				blocksize);
+			ext3_msg(sb, KERN_ERR,
+				"error: bad blocksize %d", blocksize);
 			goto out_fail;
 		}
 		logic_sb_block = (sb_block * EXT3_MIN_BLOCK_SIZE) / blocksize;
 		offset = (sb_block * EXT3_MIN_BLOCK_SIZE) % blocksize;
 		bh = sb_bread(sb, logic_sb_block);
 		if (!bh) {
-			printk(KERN_ERR
-			       "EXT3-fs: Can't read superblock on 2nd try.\n");
+			ext3_msg(sb, KERN_ERR,
+			       "error: can't read superblock on 2nd try");
 			goto failed_mount;
 		}
 		es = (struct ext3_super_block *)(((char *)bh->b_data) + offset);
 		sbi->s_es = es;
 		if (es->s_magic != cpu_to_le16(EXT3_SUPER_MAGIC)) {
-			printk (KERN_ERR
-				"EXT3-fs: Magic mismatch, very weird !\n");
+			ext3_msg(sb, KERN_ERR,
+				"error: magic mismatch");
 			goto failed_mount;
 		}
 	}
@@ -1743,8 +1779,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 		if ((sbi->s_inode_size < EXT3_GOOD_OLD_INODE_SIZE) ||
 		    (!is_power_of_2(sbi->s_inode_size)) ||
 		    (sbi->s_inode_size > blocksize)) {
-			printk (KERN_ERR
-				"EXT3-fs: unsupported inode size: %d\n",
+			ext3_msg(sb, KERN_ERR,
+				"error: unsupported inode size: %d",
 				sbi->s_inode_size);
 			goto failed_mount;
 		}
@@ -1752,8 +1788,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	sbi->s_frag_size = EXT3_MIN_FRAG_SIZE <<
 				   le32_to_cpu(es->s_log_frag_size);
 	if (blocksize != sbi->s_frag_size) {
-		printk(KERN_ERR
-		       "EXT3-fs: fragsize %lu != blocksize %u (unsupported)\n",
+		ext3_msg(sb, KERN_ERR,
+		       "error: fragsize %lu != blocksize %u (unsupported)",
 		       sbi->s_frag_size, blocksize);
 		goto failed_mount;
 	}
@@ -1789,31 +1825,31 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	}
 
 	if (sbi->s_blocks_per_group > blocksize * 8) {
-		printk (KERN_ERR
-			"EXT3-fs: #blocks per group too big: %lu\n",
+		ext3_msg(sb, KERN_ERR,
+			"#blocks per group too big: %lu",
 			sbi->s_blocks_per_group);
 		goto failed_mount;
 	}
 	if (sbi->s_frags_per_group > blocksize * 8) {
-		printk (KERN_ERR
-			"EXT3-fs: #fragments per group too big: %lu\n",
+		ext3_msg(sb, KERN_ERR,
+			"error: #fragments per group too big: %lu",
 			sbi->s_frags_per_group);
 		goto failed_mount;
 	}
 	if (sbi->s_inodes_per_group > blocksize * 8) {
-		printk (KERN_ERR
-			"EXT3-fs: #inodes per group too big: %lu\n",
+		ext3_msg(sb, KERN_ERR,
+			"error: #inodes per group too big: %lu",
 			sbi->s_inodes_per_group);
 		goto failed_mount;
 	}
 
 	if (le32_to_cpu(es->s_blocks_count) >
 		    (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) {
-		printk(KERN_ERR "EXT3-fs: filesystem on %s:"
-			" too large to mount safely\n", sb->s_id);
+		ext3_msg(sb, KERN_ERR,
+			"error: filesystem is too large to mount safely");
 		if (sizeof(sector_t) < 8)
-			printk(KERN_WARNING "EXT3-fs: CONFIG_LBDAF not "
-					"enabled\n");
+			ext3_msg(sb, KERN_ERR,
+				"error: CONFIG_LBDAF not enabled");
 		goto failed_mount;
 	}
 
@@ -1827,7 +1863,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *),
 				    GFP_KERNEL);
 	if (sbi->s_group_desc == NULL) {
-		printk (KERN_ERR "EXT3-fs: not enough memory\n");
+		ext3_msg(sb, KERN_ERR,
+			"error: not enough memory");
 		goto failed_mount;
 	}
 
@@ -1837,14 +1874,15 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 		block = descriptor_loc(sb, logic_sb_block, i);
 		sbi->s_group_desc[i] = sb_bread(sb, block);
 		if (!sbi->s_group_desc[i]) {
-			printk (KERN_ERR "EXT3-fs: "
-				"can't read group descriptor %d\n", i);
+			ext3_msg(sb, KERN_ERR,
+				"error: can't read group descriptor %d", i);
 			db_count = i;
 			goto failed_mount2;
 		}
 	}
 	if (!ext3_check_descriptors (sb)) {
-		printk(KERN_ERR "EXT3-fs: group descriptors corrupted!\n");
+		ext3_msg(sb, KERN_ERR,
+			"error: group descriptors corrupted");
 		goto failed_mount2;
 	}
 	sbi->s_gdb_count = db_count;
@@ -1862,7 +1900,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 				ext3_count_dirs(sb));
 	}
 	if (err) {
-		printk(KERN_ERR "EXT3-fs: insufficient memory\n");
+		ext3_msg(sb, KERN_ERR, "error: insufficient memory");
 		goto failed_mount3;
 	}
 
@@ -1890,6 +1928,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	sb->dq_op = &ext3_quota_operations;
 #endif
 	INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */
+	mutex_init(&sbi->s_orphan_lock);
+	mutex_init(&sbi->s_resize_lock);
 
 	sb->s_root = NULL;
 
@@ -1910,9 +1950,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 			goto failed_mount3;
 	} else {
 		if (!silent)
-			printk (KERN_ERR
-				"ext3: No journal on filesystem on %s\n",
-				sb->s_id);
+			ext3_msg(sb, KERN_ERR,
+				"error: no journal found. "
+				"mounting ext3 over ext2?");
 		goto failed_mount3;
 	}
 
@@ -1934,8 +1974,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	case EXT3_MOUNT_WRITEBACK_DATA:
 		if (!journal_check_available_features
 		    (sbi->s_journal, 0, 0, JFS_FEATURE_INCOMPAT_REVOKE)) {
-			printk(KERN_ERR "EXT3-fs: Journal does not support "
-			       "requested data journaling mode\n");
+			ext3_msg(sb, KERN_ERR,
+				"error: journal does not support "
+				"requested data journaling mode");
 			goto failed_mount4;
 		}
 	default:
@@ -1944,8 +1985,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
 	if (test_opt(sb, NOBH)) {
 		if (!(test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA)) {
-			printk(KERN_WARNING "EXT3-fs: Ignoring nobh option - "
-				"its supported only with writeback mode\n");
+			ext3_msg(sb, KERN_WARNING,
+				"warning: ignoring nobh option - "
+				"it is supported only with writeback mode");
 			clear_opt(sbi->s_mount_opt, NOBH);
 		}
 	}
@@ -1956,39 +1998,32 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
 	root = ext3_iget(sb, EXT3_ROOT_INO);
 	if (IS_ERR(root)) {
-		printk(KERN_ERR "EXT3-fs: get root inode failed\n");
+		ext3_msg(sb, KERN_ERR, "error: get root inode failed");
 		ret = PTR_ERR(root);
 		goto failed_mount4;
 	}
 	if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) {
 		iput(root);
-		printk(KERN_ERR "EXT3-fs: corrupt root inode, run e2fsck\n");
+		ext3_msg(sb, KERN_ERR, "error: corrupt root inode, run e2fsck");
 		goto failed_mount4;
 	}
 	sb->s_root = d_alloc_root(root);
 	if (!sb->s_root) {
-		printk(KERN_ERR "EXT3-fs: get root dentry failed\n");
+		ext3_msg(sb, KERN_ERR, "error: get root dentry failed");
 		iput(root);
 		ret = -ENOMEM;
 		goto failed_mount4;
 	}
 
 	ext3_setup_super (sb, es, sb->s_flags & MS_RDONLY);
-	/*
-	 * akpm: core read_super() calls in here with the superblock locked.
-	 * That deadlocks, because orphan cleanup needs to lock the superblock
-	 * in numerous places.  Here we just pop the lock - it's relatively
-	 * harmless, because we are now ready to accept write_super() requests,
-	 * and aviro says that's the only reason for hanging onto the
-	 * superblock lock.
-	 */
+
 	EXT3_SB(sb)->s_mount_state |= EXT3_ORPHAN_FS;
 	ext3_orphan_cleanup(sb, es);
 	EXT3_SB(sb)->s_mount_state &= ~EXT3_ORPHAN_FS;
 	if (needs_recovery)
-		printk (KERN_INFO "EXT3-fs: recovery complete.\n");
+		ext3_msg(sb, KERN_INFO, "recovery complete");
 	ext3_mark_recovery_complete(sb, es);
-	printk (KERN_INFO "EXT3-fs: mounted filesystem with %s data mode.\n",
+	ext3_msg(sb, KERN_INFO, "mounted filesystem with %s data mode",
 		test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA ? "journal":
 		test_opt(sb,DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA ? "ordered":
 		"writeback");
@@ -1998,7 +2033,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
 cantfind_ext3:
 	if (!silent)
-		printk(KERN_ERR "VFS: Can't find ext3 filesystem on dev %s.\n",
+		ext3_msg(sb, KERN_INFO,
+			"error: can't find ext3 filesystem on dev %s.",
 		       sb->s_id);
 	goto failed_mount;
 
@@ -2066,27 +2102,27 @@ static journal_t *ext3_get_journal(struct super_block *sb,
 
 	journal_inode = ext3_iget(sb, journal_inum);
 	if (IS_ERR(journal_inode)) {
-		printk(KERN_ERR "EXT3-fs: no journal found.\n");
+		ext3_msg(sb, KERN_ERR, "error: no journal found");
 		return NULL;
 	}
 	if (!journal_inode->i_nlink) {
 		make_bad_inode(journal_inode);
 		iput(journal_inode);
-		printk(KERN_ERR "EXT3-fs: journal inode is deleted.\n");
+		ext3_msg(sb, KERN_ERR, "error: journal inode is deleted");
 		return NULL;
 	}
 
 	jbd_debug(2, "Journal inode found at %p: %Ld bytes\n",
 		  journal_inode, journal_inode->i_size);
 	if (!S_ISREG(journal_inode->i_mode)) {
-		printk(KERN_ERR "EXT3-fs: invalid journal inode.\n");
+		ext3_msg(sb, KERN_ERR, "error: invalid journal inode");
 		iput(journal_inode);
 		return NULL;
 	}
 
 	journal = journal_init_inode(journal_inode);
 	if (!journal) {
-		printk(KERN_ERR "EXT3-fs: Could not load journal inode\n");
+		ext3_msg(sb, KERN_ERR, "error: could not load journal inode");
 		iput(journal_inode);
 		return NULL;
 	}
@@ -2108,13 +2144,13 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
 	struct ext3_super_block * es;
 	struct block_device *bdev;
 
-	bdev = ext3_blkdev_get(j_dev);
+	bdev = ext3_blkdev_get(j_dev, sb);
 	if (bdev == NULL)
 		return NULL;
 
 	if (bd_claim(bdev, sb)) {
-		printk(KERN_ERR
-		        "EXT3: failed to claim external journal device.\n");
+		ext3_msg(sb, KERN_ERR,
+			"error: failed to claim external journal device");
 		blkdev_put(bdev, FMODE_READ|FMODE_WRITE);
 		return NULL;
 	}
@@ -2122,8 +2158,8 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
 	blocksize = sb->s_blocksize;
 	hblock = bdev_logical_block_size(bdev);
 	if (blocksize < hblock) {
-		printk(KERN_ERR
-			"EXT3-fs: blocksize too small for journal device.\n");
+		ext3_msg(sb, KERN_ERR,
+			"error: blocksize too small for journal device");
 		goto out_bdev;
 	}
 
@@ -2131,8 +2167,8 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
 	offset = EXT3_MIN_BLOCK_SIZE % blocksize;
 	set_blocksize(bdev, blocksize);
 	if (!(bh = __bread(bdev, sb_block, blocksize))) {
-		printk(KERN_ERR "EXT3-fs: couldn't read superblock of "
-		       "external journal\n");
+		ext3_msg(sb, KERN_ERR, "error: couldn't read superblock of "
+			"external journal");
 		goto out_bdev;
 	}
 
@@ -2140,14 +2176,14 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
 	if ((le16_to_cpu(es->s_magic) != EXT3_SUPER_MAGIC) ||
 	    !(le32_to_cpu(es->s_feature_incompat) &
 	      EXT3_FEATURE_INCOMPAT_JOURNAL_DEV)) {
-		printk(KERN_ERR "EXT3-fs: external journal has "
-					"bad superblock\n");
+		ext3_msg(sb, KERN_ERR, "error: external journal has "
+			"bad superblock");
 		brelse(bh);
 		goto out_bdev;
 	}
 
 	if (memcmp(EXT3_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) {
-		printk(KERN_ERR "EXT3-fs: journal UUID does not match\n");
+		ext3_msg(sb, KERN_ERR, "error: journal UUID does not match");
 		brelse(bh);
 		goto out_bdev;
 	}
@@ -2159,19 +2195,21 @@ static journal_t *ext3_get_dev_journal(struct super_block *sb,
 	journal = journal_init_dev(bdev, sb->s_bdev,
 					start, len, blocksize);
 	if (!journal) {
-		printk(KERN_ERR "EXT3-fs: failed to create device journal\n");
+		ext3_msg(sb, KERN_ERR,
+			"error: failed to create device journal");
 		goto out_bdev;
 	}
 	journal->j_private = sb;
 	ll_rw_block(READ, 1, &journal->j_sb_buffer);
 	wait_on_buffer(journal->j_sb_buffer);
 	if (!buffer_uptodate(journal->j_sb_buffer)) {
-		printk(KERN_ERR "EXT3-fs: I/O error on journal device\n");
+		ext3_msg(sb, KERN_ERR, "I/O error on journal device");
 		goto out_journal;
 	}
 	if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) {
-		printk(KERN_ERR "EXT3-fs: External journal has more than one "
-					"user (unsupported) - %d\n",
+		ext3_msg(sb, KERN_ERR,
+			"error: external journal has more than one "
+			"user (unsupported) - %d",
 			be32_to_cpu(journal->j_superblock->s_nr_users));
 		goto out_journal;
 	}
@@ -2197,8 +2235,8 @@ static int ext3_load_journal(struct super_block *sb,
 
 	if (journal_devnum &&
 	    journal_devnum != le32_to_cpu(es->s_journal_dev)) {
-		printk(KERN_INFO "EXT3-fs: external journal device major/minor "
-			"numbers have changed\n");
+		ext3_msg(sb, KERN_INFO, "external journal device major/minor "
+			"numbers have changed");
 		journal_dev = new_decode_dev(journal_devnum);
 	} else
 		journal_dev = new_decode_dev(le32_to_cpu(es->s_journal_dev));
@@ -2213,21 +2251,21 @@ static int ext3_load_journal(struct super_block *sb,
 
 	if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER)) {
 		if (sb->s_flags & MS_RDONLY) {
-			printk(KERN_INFO "EXT3-fs: INFO: recovery "
-					"required on readonly filesystem.\n");
+			ext3_msg(sb, KERN_INFO,
+				"recovery required on readonly filesystem");
 			if (really_read_only) {
-				printk(KERN_ERR "EXT3-fs: write access "
-					"unavailable, cannot proceed.\n");
+				ext3_msg(sb, KERN_ERR, "error: write access "
+					"unavailable, cannot proceed");
 				return -EROFS;
 			}
-			printk (KERN_INFO "EXT3-fs: write access will "
-					"be enabled during recovery.\n");
+			ext3_msg(sb, KERN_INFO,
+				"write access will be enabled during recovery");
 		}
 	}
 
 	if (journal_inum && journal_dev) {
-		printk(KERN_ERR "EXT3-fs: filesystem has both journal "
-		       "and inode journals!\n");
+		ext3_msg(sb, KERN_ERR, "error: filesystem has both journal "
+		       "and inode journals");
 		return -EINVAL;
 	}
 
@@ -2242,7 +2280,7 @@ static int ext3_load_journal(struct super_block *sb,
 	if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) {
 		err = journal_update_format(journal);
 		if (err)  {
-			printk(KERN_ERR "EXT3-fs: error updating journal.\n");
+			ext3_msg(sb, KERN_ERR, "error updating journal");
 			journal_destroy(journal);
 			return err;
 		}
@@ -2254,7 +2292,7 @@ static int ext3_load_journal(struct super_block *sb,
 		err = journal_load(journal);
 
 	if (err) {
-		printk(KERN_ERR "EXT3-fs: error loading journal.\n");
+		ext3_msg(sb, KERN_ERR, "error loading journal");
 		journal_destroy(journal);
 		return err;
 	}
@@ -2273,16 +2311,17 @@ static int ext3_load_journal(struct super_block *sb,
 	return 0;
 }
 
-static int ext3_create_journal(struct super_block * sb,
-			       struct ext3_super_block * es,
+static int ext3_create_journal(struct super_block *sb,
+			       struct ext3_super_block *es,
 			       unsigned int journal_inum)
 {
 	journal_t *journal;
 	int err;
 
 	if (sb->s_flags & MS_RDONLY) {
-		printk(KERN_ERR "EXT3-fs: readonly filesystem when trying to "
-				"create journal.\n");
+		ext3_msg(sb, KERN_ERR,
+			"error: readonly filesystem when trying to "
+			"create journal");
 		return -EROFS;
 	}
 
@@ -2290,12 +2329,12 @@ static int ext3_create_journal(struct super_block * sb,
 	if (!journal)
 		return -EINVAL;
 
-	printk(KERN_INFO "EXT3-fs: creating new journal on inode %u\n",
+	ext3_msg(sb, KERN_INFO, "creating new journal on inode %u",
 	       journal_inum);
 
 	err = journal_create(journal);
 	if (err) {
-		printk(KERN_ERR "EXT3-fs: error creating journal.\n");
+		ext3_msg(sb, KERN_ERR, "error creating journal");
 		journal_destroy(journal);
 		return -EIO;
 	}
@@ -2359,13 +2398,11 @@ static void ext3_mark_recovery_complete(struct super_block * sb,
 	if (journal_flush(journal) < 0)
 		goto out;
 
-	lock_super(sb);
 	if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER) &&
 	    sb->s_flags & MS_RDONLY) {
 		EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER);
 		ext3_commit_super(sb, es, 1);
 	}
-	unlock_super(sb);
 
 out:
 	journal_unlock_updates(journal);
@@ -2376,8 +2413,8 @@ out:
  * has recorded an error from a previous lifetime, move that error to the
  * main filesystem now.
  */
-static void ext3_clear_journal_err(struct super_block * sb,
-				   struct ext3_super_block * es)
+static void ext3_clear_journal_err(struct super_block *sb,
+				   struct ext3_super_block *es)
 {
 	journal_t *journal;
 	int j_errno;
@@ -2557,21 +2594,15 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data)
 			    (sbi->s_mount_state & EXT3_VALID_FS))
 				es->s_state = cpu_to_le16(sbi->s_mount_state);
 
-			/*
-			 * We have to unlock super so that we can wait for
-			 * transactions.
-			 */
-			unlock_super(sb);
 			ext3_mark_recovery_complete(sb, es);
-			lock_super(sb);
 		} else {
 			__le32 ret;
 			if ((ret = EXT3_HAS_RO_COMPAT_FEATURE(sb,
 					~EXT3_FEATURE_RO_COMPAT_SUPP))) {
-				printk(KERN_WARNING "EXT3-fs: %s: couldn't "
-				       "remount RDWR because of unsupported "
-				       "optional features (%x).\n",
-				       sb->s_id, le32_to_cpu(ret));
+				ext3_msg(sb, KERN_WARNING,
+					"warning: couldn't remount RDWR "
+					"because of unsupported optional "
+					"features (%x)", le32_to_cpu(ret));
 				err = -EROFS;
 				goto restore_opts;
 			}
@@ -2582,11 +2613,10 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data)
 			 * require a full umount/remount for now.
 			 */
 			if (es->s_last_orphan) {
-				printk(KERN_WARNING "EXT3-fs: %s: couldn't "
+				ext3_msg(sb, KERN_WARNING, "warning: couldn't "
 				       "remount RDWR because of unprocessed "
 				       "orphan inode list.  Please "
-				       "umount/remount instead.\n",
-				       sb->s_id);
+				       "umount/remount instead.");
 				err = -EINVAL;
 				goto restore_opts;
 			}
@@ -2686,13 +2716,11 @@ static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf)
 	buf->f_bsize = sb->s_blocksize;
 	buf->f_blocks = le32_to_cpu(es->s_blocks_count) - sbi->s_overhead_last;
 	buf->f_bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter);
-	es->s_free_blocks_count = cpu_to_le32(buf->f_bfree);
 	buf->f_bavail = buf->f_bfree - le32_to_cpu(es->s_r_blocks_count);
 	if (buf->f_bfree < le32_to_cpu(es->s_r_blocks_count))
 		buf->f_bavail = 0;
 	buf->f_files = le32_to_cpu(es->s_inodes_count);
 	buf->f_ffree = percpu_counter_sum_positive(&sbi->s_freeinodes_counter);
-	es->s_free_inodes_count = cpu_to_le32(buf->f_ffree);
 	buf->f_namelen = EXT3_NAME_LEN;
 	fsid = le64_to_cpup((void *)es->s_uuid) ^
 	       le64_to_cpup((void *)es->s_uuid + sizeof(u64));
@@ -2837,9 +2865,9 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id,
 	if (EXT3_SB(sb)->s_qf_names[type]) {
 		/* Quotafile not of fs root? */
 		if (path.dentry->d_parent != sb->s_root)
-			printk(KERN_WARNING
-				"EXT3-fs: Quota file not on filesystem root. "
-				"Journaled quota will not work.\n");
+			ext3_msg(sb, KERN_WARNING,
+				"warning: Quota file not on filesystem root. "
+				"Journaled quota will not work.");
 	}
 
 	/*
@@ -2921,8 +2949,9 @@ static ssize_t ext3_quota_write(struct super_block *sb, int type,
 	handle_t *handle = journal_current_handle();
 
 	if (!handle) {
-		printk(KERN_WARNING "EXT3-fs: Quota write (off=%Lu, len=%Lu)"
-			" cancelled because transaction is not started.\n",
+		ext3_msg(sb, KERN_WARNING,
+			"warning: quota write (off=%llu, len=%llu)"
+			" cancelled because transaction is not started.",
 			(unsigned long long)off, (unsigned long long)len);
 		return -EIO;
 	}
diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c
index 545e37c4b91e..66895ccf76c7 100644
--- a/fs/ext3/xattr.c
+++ b/fs/ext3/xattr.c
@@ -99,7 +99,7 @@ static struct buffer_head *ext3_xattr_cache_find(struct inode *,
 						 struct mb_cache_entry **);
 static void ext3_xattr_rehash(struct ext3_xattr_header *,
 			      struct ext3_xattr_entry *);
-static int ext3_xattr_list(struct inode *inode, char *buffer,
+static int ext3_xattr_list(struct dentry *dentry, char *buffer,
 			   size_t buffer_size);
 
 static struct mb_cache *ext3_xattr_cache;
@@ -147,7 +147,7 @@ ext3_xattr_handler(int name_index)
 ssize_t
 ext3_listxattr(struct dentry *dentry, char *buffer, size_t size)
 {
-	return ext3_xattr_list(dentry->d_inode, buffer, size);
+	return ext3_xattr_list(dentry, buffer, size);
 }
 
 static int
@@ -332,7 +332,7 @@ ext3_xattr_get(struct inode *inode, int name_index, const char *name,
 }
 
 static int
-ext3_xattr_list_entries(struct inode *inode, struct ext3_xattr_entry *entry,
+ext3_xattr_list_entries(struct dentry *dentry, struct ext3_xattr_entry *entry,
 			char *buffer, size_t buffer_size)
 {
 	size_t rest = buffer_size;
@@ -342,9 +342,10 @@ ext3_xattr_list_entries(struct inode *inode, struct ext3_xattr_entry *entry,
 			ext3_xattr_handler(entry->e_name_index);
 
 		if (handler) {
-			size_t size = handler->list(inode, buffer, rest,
+			size_t size = handler->list(dentry, buffer, rest,
 						    entry->e_name,
-						    entry->e_name_len);
+						    entry->e_name_len,
+						    handler->flags);
 			if (buffer) {
 				if (size > rest)
 					return -ERANGE;
@@ -357,8 +358,9 @@ ext3_xattr_list_entries(struct inode *inode, struct ext3_xattr_entry *entry,
 }
 
 static int
-ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size)
+ext3_xattr_block_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
+	struct inode *inode = dentry->d_inode;
 	struct buffer_head *bh = NULL;
 	int error;
 
@@ -383,7 +385,7 @@ ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size)
 		goto cleanup;
 	}
 	ext3_xattr_cache_insert(bh);
-	error = ext3_xattr_list_entries(inode, BFIRST(bh), buffer, buffer_size);
+	error = ext3_xattr_list_entries(dentry, BFIRST(bh), buffer, buffer_size);
 
 cleanup:
 	brelse(bh);
@@ -392,8 +394,9 @@ cleanup:
 }
 
 static int
-ext3_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size)
+ext3_xattr_ibody_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
+	struct inode *inode = dentry->d_inode;
 	struct ext3_xattr_ibody_header *header;
 	struct ext3_inode *raw_inode;
 	struct ext3_iloc iloc;
@@ -411,7 +414,7 @@ ext3_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size)
 	error = ext3_xattr_check_names(IFIRST(header), end);
 	if (error)
 		goto cleanup;
-	error = ext3_xattr_list_entries(inode, IFIRST(header),
+	error = ext3_xattr_list_entries(dentry, IFIRST(header),
 					buffer, buffer_size);
 
 cleanup:
@@ -430,12 +433,12 @@ cleanup:
  * used / required on success.
  */
 static int
-ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size)
+ext3_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
 	int i_error, b_error;
 
-	down_read(&EXT3_I(inode)->xattr_sem);
-	i_error = ext3_xattr_ibody_list(inode, buffer, buffer_size);
+	down_read(&EXT3_I(dentry->d_inode)->xattr_sem);
+	i_error = ext3_xattr_ibody_list(dentry, buffer, buffer_size);
 	if (i_error < 0) {
 		b_error = 0;
 	} else {
@@ -443,11 +446,11 @@ ext3_xattr_list(struct inode *inode, char *buffer, size_t buffer_size)
 			buffer += i_error;
 			buffer_size -= i_error;
 		}
-		b_error = ext3_xattr_block_list(inode, buffer, buffer_size);
+		b_error = ext3_xattr_block_list(dentry, buffer, buffer_size);
 		if (b_error < 0)
 			i_error = 0;
 	}
-	up_read(&EXT3_I(inode)->xattr_sem);
+	up_read(&EXT3_I(dentry->d_inode)->xattr_sem);
 	return i_error + b_error;
 }
 
@@ -960,6 +963,10 @@ ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
 	if (error)
 		goto cleanup;
 
+	error = ext3_journal_get_write_access(handle, is.iloc.bh);
+	if (error)
+		goto cleanup;
+
 	if (EXT3_I(inode)->i_state & EXT3_STATE_NEW) {
 		struct ext3_inode *raw_inode = ext3_raw_inode(&is.iloc);
 		memset(raw_inode, 0, EXT3_SB(inode->i_sb)->s_inode_size);
@@ -985,9 +992,6 @@ ext3_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
 		if (flags & XATTR_CREATE)
 			goto cleanup;
 	}
-	error = ext3_journal_get_write_access(handle, is.iloc.bh);
-	if (error)
-		goto cleanup;
 	if (!value) {
 		if (!is.s.not_found)
 			error = ext3_xattr_ibody_set(handle, inode, &i, &is);
diff --git a/fs/ext3/xattr_security.c b/fs/ext3/xattr_security.c
index 37b81097bdf2..474348788dd9 100644
--- a/fs/ext3/xattr_security.c
+++ b/fs/ext3/xattr_security.c
@@ -12,8 +12,8 @@
 #include "xattr.h"
 
 static size_t
-ext3_xattr_security_list(struct inode *inode, char *list, size_t list_size,
-			 const char *name, size_t name_len)
+ext3_xattr_security_list(struct dentry *dentry, char *list, size_t list_size,
+			 const char *name, size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_SECURITY_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
@@ -28,23 +28,23 @@ ext3_xattr_security_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext3_xattr_security_get(struct inode *inode, const char *name,
-		       void *buffer, size_t size)
+ext3_xattr_security_get(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext3_xattr_get(inode, EXT3_XATTR_INDEX_SECURITY, name,
-			      buffer, size);
+	return ext3_xattr_get(dentry->d_inode, EXT3_XATTR_INDEX_SECURITY,
+			      name, buffer, size);
 }
 
 static int
-ext3_xattr_security_set(struct inode *inode, const char *name,
-		       const void *value, size_t size, int flags)
+ext3_xattr_security_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext3_xattr_set(inode, EXT3_XATTR_INDEX_SECURITY, name,
-			      value, size, flags);
+	return ext3_xattr_set(dentry->d_inode, EXT3_XATTR_INDEX_SECURITY,
+			      name, value, size, flags);
 }
 
 int
diff --git a/fs/ext3/xattr_trusted.c b/fs/ext3/xattr_trusted.c
index c7c41a410c4b..e5562845ed96 100644
--- a/fs/ext3/xattr_trusted.c
+++ b/fs/ext3/xattr_trusted.c
@@ -14,8 +14,8 @@
 #include "xattr.h"
 
 static size_t
-ext3_xattr_trusted_list(struct inode *inode, char *list, size_t list_size,
-			const char *name, size_t name_len)
+ext3_xattr_trusted_list(struct dentry *dentry, char *list, size_t list_size,
+		const char *name, size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_TRUSTED_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
@@ -32,22 +32,22 @@ ext3_xattr_trusted_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext3_xattr_trusted_get(struct inode *inode, const char *name,
-		       void *buffer, size_t size)
+ext3_xattr_trusted_get(struct dentry *dentry, const char *name,
+		       void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext3_xattr_get(inode, EXT3_XATTR_INDEX_TRUSTED, name,
-			      buffer, size);
+	return ext3_xattr_get(dentry->d_inode, EXT3_XATTR_INDEX_TRUSTED,
+			      name, buffer, size);
 }
 
 static int
-ext3_xattr_trusted_set(struct inode *inode, const char *name,
-		       const void *value, size_t size, int flags)
+ext3_xattr_trusted_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext3_xattr_set(inode, EXT3_XATTR_INDEX_TRUSTED, name,
+	return ext3_xattr_set(dentry->d_inode, EXT3_XATTR_INDEX_TRUSTED, name,
 			      value, size, flags);
 }
 
diff --git a/fs/ext3/xattr_user.c b/fs/ext3/xattr_user.c
index 430fe63b31b3..3bcfe9ee0a68 100644
--- a/fs/ext3/xattr_user.c
+++ b/fs/ext3/xattr_user.c
@@ -13,13 +13,13 @@
 #include "xattr.h"
 
 static size_t
-ext3_xattr_user_list(struct inode *inode, char *list, size_t list_size,
-		     const char *name, size_t name_len)
+ext3_xattr_user_list(struct dentry *dentry, char *list, size_t list_size,
+		const char *name, size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_USER_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return 0;
 
 	if (list && total_len <= list_size) {
@@ -31,26 +31,27 @@ ext3_xattr_user_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext3_xattr_user_get(struct inode *inode, const char *name,
-		    void *buffer, size_t size)
+ext3_xattr_user_get(struct dentry *dentry, const char *name, void *buffer,
+		size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return -EOPNOTSUPP;
-	return ext3_xattr_get(inode, EXT3_XATTR_INDEX_USER, name, buffer, size);
+	return ext3_xattr_get(dentry->d_inode, EXT3_XATTR_INDEX_USER,
+			      name, buffer, size);
 }
 
 static int
-ext3_xattr_user_set(struct inode *inode, const char *name,
-		    const void *value, size_t size, int flags)
+ext3_xattr_user_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return -EOPNOTSUPP;
-	return ext3_xattr_set(inode, EXT3_XATTR_INDEX_USER, name,
-			      value, size, flags);
+	return ext3_xattr_set(dentry->d_inode, EXT3_XATTR_INDEX_USER,
+			      name, value, size, flags);
 }
 
 struct xattr_handler ext3_xattr_user_handler = {
diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig
index 9f2d45d75b1a..9ed1bb1f319f 100644
--- a/fs/ext4/Kconfig
+++ b/fs/ext4/Kconfig
@@ -26,6 +26,17 @@ config EXT4_FS
 
 	  If unsure, say N.
 
+config EXT4_USE_FOR_EXT23
+	bool "Use ext4 for ext2/ext3 file systems"
+	depends on EXT4_FS
+	depends on EXT3_FS=n || EXT2_FS=n
+	default y
+	help
+	  Allow the ext4 file system driver code to be used for ext2 or
+	  ext3 file system mounts.  This allows users to reduce their
+	  compiled kernel size by using one file system driver for
+	  ext2, ext3, and ext4 file systems.
+
 config EXT4_FS_XATTR
 	bool "Ext4 extended attributes"
 	depends on EXT4_FS
diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c
index 0df88b2a69b0..8a2a29d35a6f 100644
--- a/fs/ext4/acl.c
+++ b/fs/ext4/acl.c
@@ -364,12 +364,12 @@ out:
  * Extended attribute handlers
  */
 static size_t
-ext4_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len,
-			   const char *name, size_t name_len)
+ext4_xattr_list_acl_access(struct dentry *dentry, char *list, size_t list_len,
+			   const char *name, size_t name_len, int type)
 {
 	const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS);
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_len)
 		memcpy(list, POSIX_ACL_XATTR_ACCESS, size);
@@ -377,12 +377,12 @@ ext4_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len,
 }
 
 static size_t
-ext4_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len,
-			    const char *name, size_t name_len)
+ext4_xattr_list_acl_default(struct dentry *dentry, char *list, size_t list_len,
+			    const char *name, size_t name_len, int type)
 {
 	const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT);
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return 0;
 	if (list && size <= list_len)
 		memcpy(list, POSIX_ACL_XATTR_DEFAULT, size);
@@ -390,15 +390,18 @@ ext4_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len,
 }
 
 static int
-ext4_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size)
+ext4_xattr_get_acl(struct dentry *dentry, const char *name, void *buffer,
+		   size_t size, int type)
 {
 	struct posix_acl *acl;
 	int error;
 
-	if (!test_opt(inode->i_sb, POSIX_ACL))
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+	if (!test_opt(dentry->d_sb, POSIX_ACL))
 		return -EOPNOTSUPP;
 
-	acl = ext4_get_acl(inode, type);
+	acl = ext4_get_acl(dentry->d_inode, type);
 	if (IS_ERR(acl))
 		return PTR_ERR(acl);
 	if (acl == NULL)
@@ -410,31 +413,16 @@ ext4_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size)
 }
 
 static int
-ext4_xattr_get_acl_access(struct inode *inode, const char *name,
-			  void *buffer, size_t size)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext4_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size);
-}
-
-static int
-ext4_xattr_get_acl_default(struct inode *inode, const char *name,
-			   void *buffer, size_t size)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext4_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size);
-}
-
-static int
-ext4_xattr_set_acl(struct inode *inode, int type, const void *value,
-		   size_t size)
+ext4_xattr_set_acl(struct dentry *dentry, const char *name, const void *value,
+		   size_t size, int flags, int type)
 {
+	struct inode *inode = dentry->d_inode;
 	handle_t *handle;
 	struct posix_acl *acl;
 	int error, retries = 0;
 
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
 	if (!test_opt(inode->i_sb, POSIX_ACL))
 		return -EOPNOTSUPP;
 	if (!is_owner_or_cap(inode))
@@ -466,34 +454,18 @@ release_and_out:
 	return error;
 }
 
-static int
-ext4_xattr_set_acl_access(struct inode *inode, const char *name,
-			  const void *value, size_t size, int flags)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext4_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size);
-}
-
-static int
-ext4_xattr_set_acl_default(struct inode *inode, const char *name,
-			   const void *value, size_t size, int flags)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ext4_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size);
-}
-
 struct xattr_handler ext4_xattr_acl_access_handler = {
 	.prefix	= POSIX_ACL_XATTR_ACCESS,
+	.flags	= ACL_TYPE_ACCESS,
 	.list	= ext4_xattr_list_acl_access,
-	.get	= ext4_xattr_get_acl_access,
-	.set	= ext4_xattr_set_acl_access,
+	.get	= ext4_xattr_get_acl,
+	.set	= ext4_xattr_set_acl,
 };
 
 struct xattr_handler ext4_xattr_acl_default_handler = {
 	.prefix	= POSIX_ACL_XATTR_DEFAULT,
+	.flags	= ACL_TYPE_DEFAULT,
 	.list	= ext4_xattr_list_acl_default,
-	.get	= ext4_xattr_get_acl_default,
-	.set	= ext4_xattr_set_acl_default,
+	.get	= ext4_xattr_get_acl,
+	.set	= ext4_xattr_set_acl,
 };
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 1d0418980f8d..22bc7435d913 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -499,44 +499,6 @@ error_return:
 }
 
 /**
- * ext4_free_blocks() -- Free given blocks and update quota
- * @handle:		handle for this transaction
- * @inode:		inode
- * @block:		start physical block to free
- * @count:		number of blocks to count
- * @metadata: 		Are these metadata blocks
- */
-void ext4_free_blocks(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t block, unsigned long count,
-			int metadata)
-{
-	struct super_block *sb;
-	unsigned long dquot_freed_blocks;
-
-	/* this isn't the right place to decide whether block is metadata
-	 * inode.c/extents.c knows better, but for safety ... */
-	if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
-		metadata = 1;
-
-	/* We need to make sure we don't reuse
-	 * block released untill the transaction commit.
-	 * writeback mode have weak data consistency so
-	 * don't force data as metadata when freeing block
-	 * for writeback mode.
-	 */
-	if (metadata == 0 && !ext4_should_writeback_data(inode))
-		metadata = 1;
-
-	sb = inode->i_sb;
-
-	ext4_mb_free_blocks(handle, inode, block, count,
-			    metadata, &dquot_freed_blocks);
-	if (dquot_freed_blocks)
-		vfs_dq_free_block(inode, dquot_freed_blocks);
-	return;
-}
-
-/**
  * ext4_has_free_blocks()
  * @sbi:	in-core super block structure.
  * @nblocks:	number of needed blocks
@@ -761,7 +723,13 @@ static unsigned long ext4_bg_num_gdb_meta(struct super_block *sb,
 static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb,
 					ext4_group_t group)
 {
-	return ext4_bg_has_super(sb, group) ? EXT4_SB(sb)->s_gdb_count : 0;
+	if (!ext4_bg_has_super(sb, group))
+		return 0;
+
+	if (EXT4_HAS_INCOMPAT_FEATURE(sb,EXT4_FEATURE_INCOMPAT_META_BG))
+		return le32_to_cpu(EXT4_SB(sb)->s_es->s_first_meta_bg);
+	else
+		return EXT4_SB(sb)->s_gdb_count;
 }
 
 /**
diff --git a/fs/ext4/block_validity.c b/fs/ext4/block_validity.c
index 50784ef07563..a60ab9aad57d 100644
--- a/fs/ext4/block_validity.c
+++ b/fs/ext4/block_validity.c
@@ -16,7 +16,6 @@
 #include <linux/module.h>
 #include <linux/swap.h>
 #include <linux/pagemap.h>
-#include <linux/version.h>
 #include <linux/blkdev.h>
 #include <linux/mutex.h>
 #include "ext4.h"
@@ -160,7 +159,7 @@ int ext4_setup_system_zone(struct super_block *sb)
 		if (ext4_bg_has_super(sb, i) &&
 		    ((i < 5) || ((i % flex_size) == 0)))
 			add_system_zone(sbi, ext4_group_first_block_no(sb, i),
-					sbi->s_gdb_count + 1);
+					ext4_bg_num_gdb(sb, i) + 1);
 		gdp = ext4_get_group_desc(sb, i, NULL);
 		ret = add_system_zone(sbi, ext4_block_bitmap(sb, gdp), 1);
 		if (ret)
@@ -228,6 +227,7 @@ int ext4_data_block_valid(struct ext4_sb_info *sbi, ext4_fsblk_t start_blk,
 	struct rb_node *n = sbi->system_blks.rb_node;
 
 	if ((start_blk <= le32_to_cpu(sbi->s_es->s_first_data_block)) ||
+	    (start_blk + count < start_blk) ||
 	    (start_blk + count > ext4_blocks_count(sbi->s_es)))
 		return 0;
 	while (n) {
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 8825515eeddd..874d169a193e 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -361,14 +361,11 @@ struct ext4_new_group_data {
 	   so set the magic i_delalloc_reserve_flag after taking the 
 	   inode allocation semaphore for */
 #define EXT4_GET_BLOCKS_DELALLOC_RESERVE	0x0004
-	/* Call ext4_da_update_reserve_space() after successfully 
-	   allocating the blocks */
-#define EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE	0x0008
 	/* caller is from the direct IO path, request to creation of an
 	unitialized extents if not allocated, split the uninitialized
 	extent if blocks has been preallocated already*/
-#define EXT4_GET_BLOCKS_DIO			0x0010
-#define EXT4_GET_BLOCKS_CONVERT			0x0020
+#define EXT4_GET_BLOCKS_DIO			0x0008
+#define EXT4_GET_BLOCKS_CONVERT			0x0010
 #define EXT4_GET_BLOCKS_DIO_CREATE_EXT		(EXT4_GET_BLOCKS_DIO|\
 					 EXT4_GET_BLOCKS_CREATE_UNINIT_EXT)
 	/* Convert extent to initialized after direct IO complete */
@@ -376,6 +373,12 @@ struct ext4_new_group_data {
 					 EXT4_GET_BLOCKS_DIO_CREATE_EXT)
 
 /*
+ * Flags used by ext4_free_blocks
+ */
+#define EXT4_FREE_BLOCKS_METADATA	0x0001
+#define EXT4_FREE_BLOCKS_FORGET		0x0002
+
+/*
  * ioctl commands
  */
 #define	EXT4_IOC_GETFLAGS		FS_IOC_GETFLAGS
@@ -693,16 +696,29 @@ struct ext4_inode_info {
 	unsigned int i_reserved_meta_blocks;
 	unsigned int i_allocated_meta_blocks;
 	unsigned short i_delalloc_reserved_flag;
+	sector_t i_da_metadata_calc_last_lblock;
+	int i_da_metadata_calc_len;
 
 	/* on-disk additional length */
 	__u16 i_extra_isize;
 
 	spinlock_t i_block_reservation_lock;
+#ifdef CONFIG_QUOTA
+	/* quota space reservation, managed internally by quota code */
+	qsize_t i_reserved_quota;
+#endif
 
 	/* completed async DIOs that might need unwritten extents handling */
 	struct list_head i_aio_dio_complete_list;
 	/* current io_end structure for async DIO write*/
 	ext4_io_end_t *cur_aio_dio;
+
+	/*
+	 * Transactions that contain inode's metadata needed to complete
+	 * fsync and fdatasync, respectively.
+	 */
+	tid_t i_sync_tid;
+	tid_t i_datasync_tid;
 };
 
 /*
@@ -750,6 +766,7 @@ struct ext4_inode_info {
 #define EXT4_MOUNT_DELALLOC		0x8000000 /* Delalloc support */
 #define EXT4_MOUNT_DATA_ERR_ABORT	0x10000000 /* Abort on file data write */
 #define EXT4_MOUNT_BLOCK_VALIDITY	0x20000000 /* Block validity checking */
+#define EXT4_MOUNT_DISCARD		0x40000000 /* Issue DISCARD requests */
 
 #define clear_opt(o, opt)		o &= ~EXT4_MOUNT_##opt
 #define set_opt(o, opt)			o |= EXT4_MOUNT_##opt
@@ -1324,8 +1341,6 @@ extern ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode,
 			ext4_fsblk_t goal, unsigned long *count, int *errp);
 extern int ext4_claim_free_blocks(struct ext4_sb_info *sbi, s64 nblocks);
 extern int ext4_has_free_blocks(struct ext4_sb_info *sbi, s64 nblocks);
-extern void ext4_free_blocks(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t block, unsigned long count, int metadata);
 extern void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
 				ext4_fsblk_t block, unsigned long count);
 extern ext4_fsblk_t ext4_count_free_blocks(struct super_block *);
@@ -1384,16 +1399,15 @@ extern int ext4_mb_reserve_blocks(struct super_block *, int);
 extern void ext4_discard_preallocations(struct inode *);
 extern int __init init_ext4_mballoc(void);
 extern void exit_ext4_mballoc(void);
-extern void ext4_mb_free_blocks(handle_t *, struct inode *,
-		ext4_fsblk_t, unsigned long, int, unsigned long *);
+extern void ext4_free_blocks(handle_t *handle, struct inode *inode,
+			     struct buffer_head *bh, ext4_fsblk_t block,
+			     unsigned long count, int flags);
 extern int ext4_mb_add_groupinfo(struct super_block *sb,
 		ext4_group_t i, struct ext4_group_desc *desc);
 extern int ext4_mb_get_buddy_cache_lock(struct super_block *, ext4_group_t);
 extern void ext4_mb_put_buddy_cache_lock(struct super_block *,
 						ext4_group_t, int);
 /* inode.c */
-int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
-		struct buffer_head *bh, ext4_fsblk_t blocknr);
 struct buffer_head *ext4_getblk(handle_t *, struct inode *,
 						ext4_lblk_t, int, int *);
 struct buffer_head *ext4_bread(handle_t *, struct inode *,
@@ -1424,8 +1438,10 @@ extern int ext4_chunk_trans_blocks(struct inode *, int nrblocks);
 extern int ext4_block_truncate_page(handle_t *handle,
 		struct address_space *mapping, loff_t from);
 extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf);
-extern qsize_t ext4_get_reserved_space(struct inode *inode);
+extern qsize_t *ext4_get_reserved_space(struct inode *inode);
 extern int flush_aio_dio_completed_IO(struct inode *inode);
+extern void ext4_da_update_reserve_space(struct inode *inode,
+					int used, int quota_claim);
 /* ioctl.c */
 extern long ext4_ioctl(struct file *, unsigned int, unsigned long);
 extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long);
diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h
index 2ca686454e87..bdb6ce7e2eb4 100644
--- a/fs/ext4/ext4_extents.h
+++ b/fs/ext4/ext4_extents.h
@@ -225,7 +225,8 @@ static inline void ext4_ext_mark_initialized(struct ext4_extent *ext)
 	ext->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ext));
 }
 
-extern int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks);
+extern int ext4_ext_calc_metadata_amount(struct inode *inode,
+					 sector_t lblocks);
 extern ext4_fsblk_t ext_pblock(struct ext4_extent *ex);
 extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *);
 extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t);
diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c
index 6a9409920dee..b57e5c711b6d 100644
--- a/fs/ext4/ext4_jbd2.c
+++ b/fs/ext4/ext4_jbd2.c
@@ -4,6 +4,8 @@
 
 #include "ext4_jbd2.h"
 
+#include <trace/events/ext4.h>
+
 int __ext4_journal_get_undo_access(const char *where, handle_t *handle,
 				struct buffer_head *bh)
 {
@@ -32,35 +34,69 @@ int __ext4_journal_get_write_access(const char *where, handle_t *handle,
 	return err;
 }
 
-int __ext4_journal_forget(const char *where, handle_t *handle,
-				struct buffer_head *bh)
+/*
+ * The ext4 forget function must perform a revoke if we are freeing data
+ * which has been journaled.  Metadata (eg. indirect blocks) must be
+ * revoked in all cases.
+ *
+ * "bh" may be NULL: a metadata block may have been freed from memory
+ * but there may still be a record of it in the journal, and that record
+ * still needs to be revoked.
+ *
+ * If the handle isn't valid we're not journaling, but we still need to
+ * call into ext4_journal_revoke() to put the buffer head.
+ */
+int __ext4_forget(const char *where, handle_t *handle, int is_metadata,
+		  struct inode *inode, struct buffer_head *bh,
+		  ext4_fsblk_t blocknr)
 {
-	int err = 0;
+	int err;
 
-	if (ext4_handle_valid(handle)) {
-		err = jbd2_journal_forget(handle, bh);
-		if (err)
-			ext4_journal_abort_handle(where, __func__, bh,
-						  handle, err);
-	}
-	else
+	might_sleep();
+
+	trace_ext4_forget(inode, is_metadata, blocknr);
+	BUFFER_TRACE(bh, "enter");
+
+	jbd_debug(4, "forgetting bh %p: is_metadata = %d, mode %o, "
+		  "data mode %x\n",
+		  bh, is_metadata, inode->i_mode,
+		  test_opt(inode->i_sb, DATA_FLAGS));
+
+	/* In the no journal case, we can just do a bforget and return */
+	if (!ext4_handle_valid(handle)) {
 		bforget(bh);
-	return err;
-}
+		return 0;
+	}
 
-int __ext4_journal_revoke(const char *where, handle_t *handle,
-				ext4_fsblk_t blocknr, struct buffer_head *bh)
-{
-	int err = 0;
+	/* Never use the revoke function if we are doing full data
+	 * journaling: there is no need to, and a V1 superblock won't
+	 * support it.  Otherwise, only skip the revoke on un-journaled
+	 * data blocks. */
 
-	if (ext4_handle_valid(handle)) {
-		err = jbd2_journal_revoke(handle, blocknr, bh);
-		if (err)
-			ext4_journal_abort_handle(where, __func__, bh,
-						  handle, err);
+	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ||
+	    (!is_metadata && !ext4_should_journal_data(inode))) {
+		if (bh) {
+			BUFFER_TRACE(bh, "call jbd2_journal_forget");
+			err = jbd2_journal_forget(handle, bh);
+			if (err)
+				ext4_journal_abort_handle(where, __func__, bh,
+							  handle, err);
+			return err;
+		}
+		return 0;
 	}
-	else
-		bforget(bh);
+
+	/*
+	 * data!=journal && (is_metadata || should_journal_data(inode))
+	 */
+	BUFFER_TRACE(bh, "call jbd2_journal_revoke");
+	err = jbd2_journal_revoke(handle, blocknr, bh);
+	if (err) {
+		ext4_journal_abort_handle(where, __func__, bh, handle, err);
+		ext4_abort(inode->i_sb, __func__,
+			   "error %d when attempting revoke", err);
+	}
+	BUFFER_TRACE(bh, "exit");
 	return err;
 }
 
diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h
index a2865980342f..05eca817d704 100644
--- a/fs/ext4/ext4_jbd2.h
+++ b/fs/ext4/ext4_jbd2.h
@@ -49,7 +49,7 @@
 
 #define EXT4_DATA_TRANS_BLOCKS(sb)	(EXT4_SINGLEDATA_TRANS_BLOCKS(sb) + \
 					 EXT4_XATTR_TRANS_BLOCKS - 2 + \
-					 2*EXT4_QUOTA_TRANS_BLOCKS(sb))
+					 EXT4_MAXQUOTAS_TRANS_BLOCKS(sb))
 
 /*
  * Define the number of metadata blocks we need to account to modify data.
@@ -57,7 +57,7 @@
  * This include super block, inode block, quota blocks and xattr blocks
  */
 #define EXT4_META_TRANS_BLOCKS(sb)	(EXT4_XATTR_TRANS_BLOCKS + \
-					2*EXT4_QUOTA_TRANS_BLOCKS(sb))
+					EXT4_MAXQUOTAS_TRANS_BLOCKS(sb))
 
 /* Delete operations potentially hit one directory's namespace plus an
  * entire inode, plus arbitrary amounts of bitmap/indirection data.  Be
@@ -92,6 +92,7 @@
  * but inode, sb and group updates are done only once */
 #define EXT4_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\
 		(EXT4_SINGLEDATA_TRANS_BLOCKS(sb)-3)+3+DQUOT_INIT_REWRITE) : 0)
+
 #define EXT4_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\
 		(EXT4_SINGLEDATA_TRANS_BLOCKS(sb)-3)+3+DQUOT_DEL_REWRITE) : 0)
 #else
@@ -99,6 +100,9 @@
 #define EXT4_QUOTA_INIT_BLOCKS(sb) 0
 #define EXT4_QUOTA_DEL_BLOCKS(sb) 0
 #endif
+#define EXT4_MAXQUOTAS_TRANS_BLOCKS(sb) (MAXQUOTAS*EXT4_QUOTA_TRANS_BLOCKS(sb))
+#define EXT4_MAXQUOTAS_INIT_BLOCKS(sb) (MAXQUOTAS*EXT4_QUOTA_INIT_BLOCKS(sb))
+#define EXT4_MAXQUOTAS_DEL_BLOCKS(sb) (MAXQUOTAS*EXT4_QUOTA_DEL_BLOCKS(sb))
 
 int
 ext4_mark_iloc_dirty(handle_t *handle,
@@ -116,12 +120,8 @@ int ext4_reserve_inode_write(handle_t *handle, struct inode *inode,
 int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode);
 
 /*
- * Wrapper functions with which ext4 calls into JBD.  The intent here is
- * to allow these to be turned into appropriate stubs so ext4 can control
- * ext2 filesystems, so ext2+ext4 systems only nee one fs.  This work hasn't
- * been done yet.
+ * Wrapper functions with which ext4 calls into JBD.
  */
-
 void ext4_journal_abort_handle(const char *caller, const char *err_fn,
 		struct buffer_head *bh, handle_t *handle, int err);
 
@@ -131,13 +131,9 @@ int __ext4_journal_get_undo_access(const char *where, handle_t *handle,
 int __ext4_journal_get_write_access(const char *where, handle_t *handle,
 				struct buffer_head *bh);
 
-/* When called with an invalid handle, this will still do a put on the BH */
-int __ext4_journal_forget(const char *where, handle_t *handle,
-				struct buffer_head *bh);
-
-/* When called with an invalid handle, this will still do a put on the BH */
-int __ext4_journal_revoke(const char *where, handle_t *handle,
-				ext4_fsblk_t blocknr, struct buffer_head *bh);
+int __ext4_forget(const char *where, handle_t *handle, int is_metadata,
+		  struct inode *inode, struct buffer_head *bh,
+		  ext4_fsblk_t blocknr);
 
 int __ext4_journal_get_create_access(const char *where,
 				handle_t *handle, struct buffer_head *bh);
@@ -149,12 +145,11 @@ int __ext4_handle_dirty_metadata(const char *where, handle_t *handle,
 	__ext4_journal_get_undo_access(__func__, (handle), (bh))
 #define ext4_journal_get_write_access(handle, bh) \
 	__ext4_journal_get_write_access(__func__, (handle), (bh))
-#define ext4_journal_revoke(handle, blocknr, bh) \
-	__ext4_journal_revoke(__func__, (handle), (blocknr), (bh))
+#define ext4_forget(handle, is_metadata, inode, bh, block_nr) \
+	__ext4_forget(__func__, (handle), (is_metadata), (inode), (bh),\
+		      (block_nr))
 #define ext4_journal_get_create_access(handle, bh) \
 	__ext4_journal_get_create_access(__func__, (handle), (bh))
-#define ext4_journal_forget(handle, bh) \
-	__ext4_journal_forget(__func__, (handle), (bh))
 #define ext4_handle_dirty_metadata(handle, inode, bh) \
 	__ext4_handle_dirty_metadata(__func__, (handle), (inode), (bh))
 
@@ -254,6 +249,19 @@ static inline int ext4_jbd2_file_inode(handle_t *handle, struct inode *inode)
 	return 0;
 }
 
+static inline void ext4_update_inode_fsync_trans(handle_t *handle,
+						 struct inode *inode,
+						 int datasync)
+{
+	struct ext4_inode_info *ei = EXT4_I(inode);
+
+	if (ext4_handle_valid(handle)) {
+		ei->i_sync_tid = handle->h_transaction->t_tid;
+		if (datasync)
+			ei->i_datasync_tid = handle->h_transaction->t_tid;
+	}
+}
+
 /* super.c */
 int ext4_force_commit(struct super_block *sb);
 
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 715264b4bae4..765a4826b118 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -296,29 +296,44 @@ static inline int ext4_ext_space_root_idx(struct inode *inode, int check)
  * to allocate @blocks
  * Worse case is one block per extent
  */
-int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks)
+int ext4_ext_calc_metadata_amount(struct inode *inode, sector_t lblock)
 {
-	int lcap, icap, rcap, leafs, idxs, num;
-	int newextents = blocks;
-
-	rcap = ext4_ext_space_root_idx(inode, 0);
-	lcap = ext4_ext_space_block(inode, 0);
-	icap = ext4_ext_space_block_idx(inode, 0);
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	int idxs, num = 0;
 
-	/* number of new leaf blocks needed */
-	num = leafs = (newextents + lcap - 1) / lcap;
+	idxs = ((inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header))
+		/ sizeof(struct ext4_extent_idx));
 
 	/*
-	 * Worse case, we need separate index block(s)
-	 * to link all new leaf blocks
+	 * If the new delayed allocation block is contiguous with the
+	 * previous da block, it can share index blocks with the
+	 * previous block, so we only need to allocate a new index
+	 * block every idxs leaf blocks.  At ldxs**2 blocks, we need
+	 * an additional index block, and at ldxs**3 blocks, yet
+	 * another index blocks.
 	 */
-	idxs = (leafs + icap - 1) / icap;
-	do {
-		num += idxs;
-		idxs = (idxs + icap - 1) / icap;
-	} while (idxs > rcap);
+	if (ei->i_da_metadata_calc_len &&
+	    ei->i_da_metadata_calc_last_lblock+1 == lblock) {
+		if ((ei->i_da_metadata_calc_len % idxs) == 0)
+			num++;
+		if ((ei->i_da_metadata_calc_len % (idxs*idxs)) == 0)
+			num++;
+		if ((ei->i_da_metadata_calc_len % (idxs*idxs*idxs)) == 0) {
+			num++;
+			ei->i_da_metadata_calc_len = 0;
+		} else
+			ei->i_da_metadata_calc_len++;
+		ei->i_da_metadata_calc_last_lblock++;
+		return num;
+	}
 
-	return num;
+	/*
+	 * In the worst case we need a new set of index blocks at
+	 * every level of the inode's extent tree.
+	 */
+	ei->i_da_metadata_calc_len = 1;
+	ei->i_da_metadata_calc_last_lblock = lblock;
+	return ext_depth(inode) + 1;
 }
 
 static int
@@ -1007,7 +1022,8 @@ cleanup:
 		for (i = 0; i < depth; i++) {
 			if (!ablocks[i])
 				continue;
-			ext4_free_blocks(handle, inode, ablocks[i], 1, 1);
+			ext4_free_blocks(handle, inode, 0, ablocks[i], 1,
+					 EXT4_FREE_BLOCKS_METADATA);
 		}
 	}
 	kfree(ablocks);
@@ -1761,7 +1777,9 @@ int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block,
 	while (block < last && block != EXT_MAX_BLOCK) {
 		num = last - block;
 		/* find extent for this block */
+		down_read(&EXT4_I(inode)->i_data_sem);
 		path = ext4_ext_find_extent(inode, block, path);
+		up_read(&EXT4_I(inode)->i_data_sem);
 		if (IS_ERR(path)) {
 			err = PTR_ERR(path);
 			path = NULL;
@@ -1957,7 +1975,6 @@ errout:
 static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
 			struct ext4_ext_path *path)
 {
-	struct buffer_head *bh;
 	int err;
 	ext4_fsblk_t leaf;
 
@@ -1973,9 +1990,8 @@ static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
 	if (err)
 		return err;
 	ext_debug("index is empty, remove it, free block %llu\n", leaf);
-	bh = sb_find_get_block(inode->i_sb, leaf);
-	ext4_forget(handle, 1, inode, bh, leaf);
-	ext4_free_blocks(handle, inode, leaf, 1, 1);
+	ext4_free_blocks(handle, inode, 0, leaf, 1,
+			 EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET);
 	return err;
 }
 
@@ -2042,12 +2058,11 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
 				struct ext4_extent *ex,
 				ext4_lblk_t from, ext4_lblk_t to)
 {
-	struct buffer_head *bh;
 	unsigned short ee_len =  ext4_ext_get_actual_len(ex);
-	int i, metadata = 0;
+	int flags = EXT4_FREE_BLOCKS_FORGET;
 
 	if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
-		metadata = 1;
+		flags |= EXT4_FREE_BLOCKS_METADATA;
 #ifdef EXTENTS_STATS
 	{
 		struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
@@ -2072,11 +2087,7 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode,
 		num = le32_to_cpu(ex->ee_block) + ee_len - from;
 		start = ext_pblock(ex) + ee_len - num;
 		ext_debug("free last %u blocks starting %llu\n", num, start);
-		for (i = 0; i < num; i++) {
-			bh = sb_find_get_block(inode->i_sb, start + i);
-			ext4_forget(handle, 0, inode, bh, start + i);
-		}
-		ext4_free_blocks(handle, inode, start, num, metadata);
+		ext4_free_blocks(handle, inode, 0, start, num, flags);
 	} else if (from == le32_to_cpu(ex->ee_block)
 		   && to <= le32_to_cpu(ex->ee_block) + ee_len - 1) {
 		printk(KERN_INFO "strange request: removal %u-%u from %u:%u\n",
@@ -2167,7 +2178,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
 			correct_index = 1;
 			credits += (ext_depth(inode)) + 1;
 		}
-		credits += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb);
+		credits += EXT4_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb);
 
 		err = ext4_ext_truncate_extend_restart(handle, inode, credits);
 		if (err)
@@ -3027,6 +3038,14 @@ out:
 	return err;
 }
 
+static void unmap_underlying_metadata_blocks(struct block_device *bdev,
+			sector_t block, int count)
+{
+	int i;
+	for (i = 0; i < count; i++)
+                unmap_underlying_metadata(bdev, block + i);
+}
+
 static int
 ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
 			ext4_lblk_t iblock, unsigned int max_blocks,
@@ -3064,6 +3083,8 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
 	if (flags == EXT4_GET_BLOCKS_DIO_CONVERT_EXT) {
 		ret = ext4_convert_unwritten_extents_dio(handle, inode,
 							path);
+		if (ret >= 0)
+			ext4_update_inode_fsync_trans(handle, inode, 1);
 		goto out2;
 	}
 	/* buffered IO case */
@@ -3091,6 +3112,8 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
 	ret = ext4_ext_convert_to_initialized(handle, inode,
 						path, iblock,
 						max_blocks);
+	if (ret >= 0)
+		ext4_update_inode_fsync_trans(handle, inode, 1);
 out:
 	if (ret <= 0) {
 		err = ret;
@@ -3098,6 +3121,30 @@ out:
 	} else
 		allocated = ret;
 	set_buffer_new(bh_result);
+	/*
+	 * if we allocated more blocks than requested
+	 * we need to make sure we unmap the extra block
+	 * allocated. The actual needed block will get
+	 * unmapped later when we find the buffer_head marked
+	 * new.
+	 */
+	if (allocated > max_blocks) {
+		unmap_underlying_metadata_blocks(inode->i_sb->s_bdev,
+					newblock + max_blocks,
+					allocated - max_blocks);
+		allocated = max_blocks;
+	}
+
+	/*
+	 * If we have done fallocate with the offset that is already
+	 * delayed allocated, we would have block reservation
+	 * and quota reservation done in the delayed write path.
+	 * But fallocate would have already updated quota and block
+	 * count for this offset. So cancel these reservation
+	 */
+	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
+		ext4_da_update_reserve_space(inode, allocated, 0);
+
 map_out:
 	set_buffer_mapped(bh_result);
 out1:
@@ -3190,7 +3237,13 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 	 * this situation is possible, though, _during_ tree modification;
 	 * this is why assert can't be put in ext4_ext_find_extent()
 	 */
-	BUG_ON(path[depth].p_ext == NULL && depth != 0);
+	if (path[depth].p_ext == NULL && depth != 0) {
+		ext4_error(inode->i_sb, __func__, "bad extent address "
+			   "inode: %lu, iblock: %d, depth: %d",
+			   inode->i_ino, iblock, depth);
+		err = -EIO;
+		goto out2;
+	}
 	eh = path[depth].p_hdr;
 
 	ex = path[depth].p_ext;
@@ -3319,20 +3372,35 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 		/* not a good idea to call discard here directly,
 		 * but otherwise we'd need to call it every free() */
 		ext4_discard_preallocations(inode);
-		ext4_free_blocks(handle, inode, ext_pblock(&newex),
-					ext4_ext_get_actual_len(&newex), 0);
+		ext4_free_blocks(handle, inode, 0, ext_pblock(&newex),
+				 ext4_ext_get_actual_len(&newex), 0);
 		goto out2;
 	}
 
 	/* previous routine could use block we allocated */
 	newblock = ext_pblock(&newex);
 	allocated = ext4_ext_get_actual_len(&newex);
+	if (allocated > max_blocks)
+		allocated = max_blocks;
 	set_buffer_new(bh_result);
 
-	/* Cache only when it is _not_ an uninitialized extent */
-	if ((flags & EXT4_GET_BLOCKS_UNINIT_EXT) == 0)
+	/*
+	 * Update reserved blocks/metadata blocks after successful
+	 * block allocation which had been deferred till now.
+	 */
+	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
+		ext4_da_update_reserve_space(inode, allocated, 1);
+
+	/*
+	 * Cache the extent and update transaction to commit on fdatasync only
+	 * when it is _not_ an uninitialized extent.
+	 */
+	if ((flags & EXT4_GET_BLOCKS_UNINIT_EXT) == 0) {
 		ext4_ext_put_in_cache(inode, iblock, allocated, newblock,
 						EXT4_EXT_CACHE_EXTENT);
+		ext4_update_inode_fsync_trans(handle, inode, 1);
+	} else
+		ext4_update_inode_fsync_trans(handle, inode, 0);
 out:
 	if (allocated > max_blocks)
 		allocated = max_blocks;
@@ -3720,10 +3788,8 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		 * Walk the extent tree gathering extent information.
 		 * ext4_ext_fiemap_cb will push extents back to user.
 		 */
-		down_read(&EXT4_I(inode)->i_data_sem);
 		error = ext4_ext_walk_space(inode, start_blk, len_blks,
 					  ext4_ext_fiemap_cb, fieinfo);
-		up_read(&EXT4_I(inode)->i_data_sem);
 	}
 
 	return error;
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c
index 2b1531266ee2..98bd140aad01 100644
--- a/fs/ext4/fsync.c
+++ b/fs/ext4/fsync.c
@@ -51,25 +51,30 @@
 int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync)
 {
 	struct inode *inode = dentry->d_inode;
+	struct ext4_inode_info *ei = EXT4_I(inode);
 	journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
-	int err, ret = 0;
+	int ret;
+	tid_t commit_tid;
 
 	J_ASSERT(ext4_journal_current_handle() == NULL);
 
 	trace_ext4_sync_file(file, dentry, datasync);
 
+	if (inode->i_sb->s_flags & MS_RDONLY)
+		return 0;
+
 	ret = flush_aio_dio_completed_IO(inode);
 	if (ret < 0)
-		goto out;
+		return ret;
+	
+	if (!journal)
+		return simple_fsync(file, dentry, datasync);
+
 	/*
-	 * data=writeback:
+	 * data=writeback,ordered:
 	 *  The caller's filemap_fdatawrite()/wait will sync the data.
-	 *  sync_inode() will sync the metadata
-	 *
-	 * data=ordered:
-	 *  The caller's filemap_fdatawrite() will write the data and
-	 *  sync_inode() will write the inode if it is dirty.  Then the caller's
-	 *  filemap_fdatawait() will wait on the pages.
+	 *  Metadata is in the journal, we wait for proper transaction to
+	 *  commit here.
 	 *
 	 * data=journal:
 	 *  filemap_fdatawrite won't do anything (the buffers are clean).
@@ -79,32 +84,25 @@ int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync)
 	 *  (they were dirtied by commit).  But that's OK - the blocks are
 	 *  safe in-journal, which is all fsync() needs to ensure.
 	 */
-	if (ext4_should_journal_data(inode)) {
-		ret = ext4_force_commit(inode->i_sb);
-		goto out;
-	}
+	if (ext4_should_journal_data(inode))
+		return ext4_force_commit(inode->i_sb);
 
-	if (!journal)
-		ret = sync_mapping_buffers(inode->i_mapping);
-
-	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
-		goto out;
-
-	/*
-	 * The VFS has written the file data.  If the inode is unaltered
-	 * then we need not start a commit.
-	 */
-	if (inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC)) {
-		struct writeback_control wbc = {
-			.sync_mode = WB_SYNC_ALL,
-			.nr_to_write = 0, /* sys_fsync did this */
-		};
-		err = sync_inode(inode, &wbc);
-		if (ret == 0)
-			ret = err;
-	}
-out:
-	if (journal && (journal->j_flags & JBD2_BARRIER))
+	commit_tid = datasync ? ei->i_datasync_tid : ei->i_sync_tid;
+	if (jbd2_log_start_commit(journal, commit_tid)) {
+		/*
+		 * When the journal is on a different device than the
+		 * fs data disk, we need to issue the barrier in
+		 * writeback mode.  (In ordered mode, the jbd2 layer
+		 * will take care of issuing the barrier.  In
+		 * data=journal, all of the data blocks are written to
+		 * the journal device.)
+		 */
+		if (ext4_should_writeback_data(inode) &&
+		    (journal->j_fs_dev != journal->j_dev) &&
+		    (journal->j_flags & JBD2_BARRIER))
+			blkdev_issue_flush(inode->i_sb->s_bdev, NULL);
+		jbd2_log_wait_commit(journal, commit_tid);
+	} else if (journal->j_flags & JBD2_BARRIER)
 		blkdev_issue_flush(inode->i_sb->s_bdev, NULL);
 	return ret;
 }
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 4e8e2f15b8bd..e11952404e02 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -71,58 +71,6 @@ static int ext4_inode_is_fast_symlink(struct inode *inode)
 }
 
 /*
- * The ext4 forget function must perform a revoke if we are freeing data
- * which has been journaled.  Metadata (eg. indirect blocks) must be
- * revoked in all cases.
- *
- * "bh" may be NULL: a metadata block may have been freed from memory
- * but there may still be a record of it in the journal, and that record
- * still needs to be revoked.
- *
- * If the handle isn't valid we're not journaling, but we still need to
- * call into ext4_journal_revoke() to put the buffer head.
- */
-int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
-		struct buffer_head *bh, ext4_fsblk_t blocknr)
-{
-	int err;
-
-	might_sleep();
-
-	BUFFER_TRACE(bh, "enter");
-
-	jbd_debug(4, "forgetting bh %p: is_metadata = %d, mode %o, "
-		  "data mode %x\n",
-		  bh, is_metadata, inode->i_mode,
-		  test_opt(inode->i_sb, DATA_FLAGS));
-
-	/* Never use the revoke function if we are doing full data
-	 * journaling: there is no need to, and a V1 superblock won't
-	 * support it.  Otherwise, only skip the revoke on un-journaled
-	 * data blocks. */
-
-	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ||
-	    (!is_metadata && !ext4_should_journal_data(inode))) {
-		if (bh) {
-			BUFFER_TRACE(bh, "call jbd2_journal_forget");
-			return ext4_journal_forget(handle, bh);
-		}
-		return 0;
-	}
-
-	/*
-	 * data!=journal && (is_metadata || should_journal_data(inode))
-	 */
-	BUFFER_TRACE(bh, "call ext4_journal_revoke");
-	err = ext4_journal_revoke(handle, blocknr, bh);
-	if (err)
-		ext4_abort(inode->i_sb, __func__,
-			   "error %d when attempting revoke", err);
-	BUFFER_TRACE(bh, "exit");
-	return err;
-}
-
-/*
  * Work out how many blocks we need to proceed with the next chunk of a
  * truncate transaction.
  */
@@ -721,7 +669,7 @@ allocated:
 	return ret;
 failed_out:
 	for (i = 0; i < index; i++)
-		ext4_free_blocks(handle, inode, new_blocks[i], 1, 0);
+		ext4_free_blocks(handle, inode, 0, new_blocks[i], 1, 0);
 	return ret;
 }
 
@@ -817,14 +765,20 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
 	return err;
 failed:
 	/* Allocation failed, free what we already allocated */
+	ext4_free_blocks(handle, inode, 0, new_blocks[0], 1, 0);
 	for (i = 1; i <= n ; i++) {
-		BUFFER_TRACE(branch[i].bh, "call jbd2_journal_forget");
-		ext4_journal_forget(handle, branch[i].bh);
+		/* 
+		 * branch[i].bh is newly allocated, so there is no
+		 * need to revoke the block, which is why we don't
+		 * need to set EXT4_FREE_BLOCKS_METADATA.
+		 */
+		ext4_free_blocks(handle, inode, 0, new_blocks[i], 1,
+				 EXT4_FREE_BLOCKS_FORGET);
 	}
-	for (i = 0; i < indirect_blks; i++)
-		ext4_free_blocks(handle, inode, new_blocks[i], 1, 0);
+	for (i = n+1; i < indirect_blks; i++)
+		ext4_free_blocks(handle, inode, 0, new_blocks[i], 1, 0);
 
-	ext4_free_blocks(handle, inode, new_blocks[i], num, 0);
+	ext4_free_blocks(handle, inode, 0, new_blocks[i], num, 0);
 
 	return err;
 }
@@ -903,12 +857,16 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode,
 
 err_out:
 	for (i = 1; i <= num; i++) {
-		BUFFER_TRACE(where[i].bh, "call jbd2_journal_forget");
-		ext4_journal_forget(handle, where[i].bh);
-		ext4_free_blocks(handle, inode,
-					le32_to_cpu(where[i-1].key), 1, 0);
+		/* 
+		 * branch[i].bh is newly allocated, so there is no
+		 * need to revoke the block, which is why we don't
+		 * need to set EXT4_FREE_BLOCKS_METADATA.
+		 */
+		ext4_free_blocks(handle, inode, where[i].bh, 0, 1,
+				 EXT4_FREE_BLOCKS_FORGET);
 	}
-	ext4_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks, 0);
+	ext4_free_blocks(handle, inode, 0, le32_to_cpu(where[num].key),
+			 blks, 0);
 
 	return err;
 }
@@ -1021,10 +979,12 @@ static int ext4_ind_get_blocks(handle_t *handle, struct inode *inode,
 	if (!err)
 		err = ext4_splice_branch(handle, inode, iblock,
 					 partial, indirect_blks, count);
-	else
+	if (err)
 		goto cleanup;
 
 	set_buffer_new(bh_result);
+
+	ext4_update_inode_fsync_trans(handle, inode, 1);
 got_it:
 	map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key));
 	if (count > blocks_to_boundary)
@@ -1043,92 +1003,120 @@ out:
 	return err;
 }
 
-qsize_t ext4_get_reserved_space(struct inode *inode)
+#ifdef CONFIG_QUOTA
+qsize_t *ext4_get_reserved_space(struct inode *inode)
 {
-	unsigned long long total;
-
-	spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
-	total = EXT4_I(inode)->i_reserved_data_blocks +
-		EXT4_I(inode)->i_reserved_meta_blocks;
-	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
-
-	return total;
+	return &EXT4_I(inode)->i_reserved_quota;
 }
+#endif
+
 /*
  * Calculate the number of metadata blocks need to reserve
- * to allocate @blocks for non extent file based file
+ * to allocate a new block at @lblocks for non extent file based file
  */
-static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks)
+static int ext4_indirect_calc_metadata_amount(struct inode *inode,
+					      sector_t lblock)
 {
-	int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb);
-	int ind_blks, dind_blks, tind_blks;
-
-	/* number of new indirect blocks needed */
-	ind_blks = (blocks + icap - 1) / icap;
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	int dind_mask = EXT4_ADDR_PER_BLOCK(inode->i_sb) - 1;
+	int blk_bits;
 
-	dind_blks = (ind_blks + icap - 1) / icap;
+	if (lblock < EXT4_NDIR_BLOCKS)
+		return 0;
 
-	tind_blks = 1;
+	lblock -= EXT4_NDIR_BLOCKS;
 
-	return ind_blks + dind_blks + tind_blks;
+	if (ei->i_da_metadata_calc_len &&
+	    (lblock & dind_mask) == ei->i_da_metadata_calc_last_lblock) {
+		ei->i_da_metadata_calc_len++;
+		return 0;
+	}
+	ei->i_da_metadata_calc_last_lblock = lblock & dind_mask;
+	ei->i_da_metadata_calc_len = 1;
+	blk_bits = roundup_pow_of_two(lblock + 1);
+	return (blk_bits / EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb)) + 1;
 }
 
 /*
  * Calculate the number of metadata blocks need to reserve
- * to allocate given number of blocks
+ * to allocate a block located at @lblock
  */
-static int ext4_calc_metadata_amount(struct inode *inode, int blocks)
+static int ext4_calc_metadata_amount(struct inode *inode, sector_t lblock)
 {
-	if (!blocks)
-		return 0;
-
 	if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)
-		return ext4_ext_calc_metadata_amount(inode, blocks);
+		return ext4_ext_calc_metadata_amount(inode, lblock);
 
-	return ext4_indirect_calc_metadata_amount(inode, blocks);
+	return ext4_indirect_calc_metadata_amount(inode, lblock);
 }
 
-static void ext4_da_update_reserve_space(struct inode *inode, int used)
+/*
+ * Called with i_data_sem down, which is important since we can call
+ * ext4_discard_preallocations() from here.
+ */
+void ext4_da_update_reserve_space(struct inode *inode,
+					int used, int quota_claim)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
-	int total, mdb, mdb_free;
-
-	spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
-	/* recalculate the number of metablocks still need to be reserved */
-	total = EXT4_I(inode)->i_reserved_data_blocks - used;
-	mdb = ext4_calc_metadata_amount(inode, total);
-
-	/* figure out how many metablocks to release */
-	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
-	mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb;
-
-	if (mdb_free) {
-		/* Account for allocated meta_blocks */
-		mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks;
-
-		/* update fs dirty blocks counter */
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	int mdb_free = 0, allocated_meta_blocks = 0;
+
+	spin_lock(&ei->i_block_reservation_lock);
+	if (unlikely(used > ei->i_reserved_data_blocks)) {
+		ext4_msg(inode->i_sb, KERN_NOTICE, "%s: ino %lu, used %d "
+			 "with only %d reserved data blocks\n",
+			 __func__, inode->i_ino, used,
+			 ei->i_reserved_data_blocks);
+		WARN_ON(1);
+		used = ei->i_reserved_data_blocks;
+	}
+
+	/* Update per-inode reservations */
+	ei->i_reserved_data_blocks -= used;
+	used += ei->i_allocated_meta_blocks;
+	ei->i_reserved_meta_blocks -= ei->i_allocated_meta_blocks;
+	allocated_meta_blocks = ei->i_allocated_meta_blocks;
+	ei->i_allocated_meta_blocks = 0;
+	percpu_counter_sub(&sbi->s_dirtyblocks_counter, used);
+
+	if (ei->i_reserved_data_blocks == 0) {
+		/*
+		 * We can release all of the reserved metadata blocks
+		 * only when we have written all of the delayed
+		 * allocation blocks.
+		 */
+		mdb_free = ei->i_reserved_meta_blocks;
+		ei->i_reserved_meta_blocks = 0;
+		ei->i_da_metadata_calc_len = 0;
 		percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free);
-		EXT4_I(inode)->i_allocated_meta_blocks = 0;
-		EXT4_I(inode)->i_reserved_meta_blocks = mdb;
 	}
-
-	/* update per-inode reservations */
-	BUG_ON(used  > EXT4_I(inode)->i_reserved_data_blocks);
-	EXT4_I(inode)->i_reserved_data_blocks -= used;
 	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 
-	/*
-	 * free those over-booking quota for metadata blocks
-	 */
-	if (mdb_free)
-		vfs_dq_release_reservation_block(inode, mdb_free);
+	/* Update quota subsystem */
+	if (quota_claim) {
+		vfs_dq_claim_block(inode, used);
+		if (mdb_free)
+			vfs_dq_release_reservation_block(inode, mdb_free);
+	} else {
+		/*
+		 * We did fallocate with an offset that is already delayed
+		 * allocated. So on delayed allocated writeback we should
+		 * not update the quota for allocated blocks. But then
+		 * converting an fallocate region to initialized region would
+		 * have caused a metadata allocation. So claim quota for
+		 * that
+		 */
+		if (allocated_meta_blocks)
+			vfs_dq_claim_block(inode, allocated_meta_blocks);
+		vfs_dq_release_reservation_block(inode, mdb_free + used);
+	}
 
 	/*
 	 * If we have done all the pending block allocations and if
 	 * there aren't any writers on the inode, we can discard the
 	 * inode's preallocations.
 	 */
-	if (!total && (atomic_read(&inode->i_writecount) == 0))
+	if ((ei->i_reserved_data_blocks == 0) &&
+	    (atomic_read(&inode->i_writecount) == 0))
 		ext4_discard_preallocations(inode);
 }
 
@@ -1320,18 +1308,20 @@ int ext4_get_blocks(handle_t *handle, struct inode *inode, sector_t block,
 			 */
 			EXT4_I(inode)->i_state &= ~EXT4_STATE_EXT_MIGRATE;
 		}
-	}
 
+		/*
+		 * Update reserved blocks/metadata blocks after successful
+		 * block allocation which had been deferred till now. We don't
+		 * support fallocate for non extent files. So we can update
+		 * reserve space here.
+		 */
+		if ((retval > 0) &&
+			(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE))
+			ext4_da_update_reserve_space(inode, retval, 1);
+	}
 	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
 		EXT4_I(inode)->i_delalloc_reserved_flag = 0;
 
-	/*
-	 * Update reserved blocks/metadata blocks after successful
-	 * block allocation which had been deferred till now.
-	 */
-	if ((retval > 0) && (flags & EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE))
-		ext4_da_update_reserve_space(inode, retval);
-
 	up_write((&EXT4_I(inode)->i_data_sem));
 	if (retval > 0 && buffer_mapped(bh)) {
 		int ret = check_block_validity(inode, "file system "
@@ -1534,6 +1524,16 @@ static int do_journal_get_write_access(handle_t *handle,
 	return ext4_journal_get_write_access(handle, bh);
 }
 
+/*
+ * Truncate blocks that were not used by write. We have to truncate the
+ * pagecache as well so that corresponding buffers get properly unmapped.
+ */
+static void ext4_truncate_failed_write(struct inode *inode)
+{
+	truncate_inode_pages(inode->i_mapping, inode->i_size);
+	ext4_truncate(inode);
+}
+
 static int ext4_write_begin(struct file *file, struct address_space *mapping,
 			    loff_t pos, unsigned len, unsigned flags,
 			    struct page **pagep, void **fsdata)
@@ -1599,7 +1599,7 @@ retry:
 
 		ext4_journal_stop(handle);
 		if (pos + len > inode->i_size) {
-			ext4_truncate(inode);
+			ext4_truncate_failed_write(inode);
 			/*
 			 * If truncate failed early the inode might
 			 * still be on the orphan list; we need to
@@ -1709,7 +1709,7 @@ static int ext4_ordered_write_end(struct file *file,
 		ret = ret2;
 
 	if (pos + len > inode->i_size) {
-		ext4_truncate(inode);
+		ext4_truncate_failed_write(inode);
 		/*
 		 * If truncate failed early the inode might still be
 		 * on the orphan list; we need to make sure the inode
@@ -1751,7 +1751,7 @@ static int ext4_writeback_write_end(struct file *file,
 		ret = ret2;
 
 	if (pos + len > inode->i_size) {
-		ext4_truncate(inode);
+		ext4_truncate_failed_write(inode);
 		/*
 		 * If truncate failed early the inode might still be
 		 * on the orphan list; we need to make sure the inode
@@ -1814,7 +1814,7 @@ static int ext4_journalled_write_end(struct file *file,
 	if (!ret)
 		ret = ret2;
 	if (pos + len > inode->i_size) {
-		ext4_truncate(inode);
+		ext4_truncate_failed_write(inode);
 		/*
 		 * If truncate failed early the inode might still be
 		 * on the orphan list; we need to make sure the inode
@@ -1827,11 +1827,15 @@ static int ext4_journalled_write_end(struct file *file,
 	return ret ? ret : copied;
 }
 
-static int ext4_da_reserve_space(struct inode *inode, int nrblocks)
+/*
+ * Reserve a single block located at lblock
+ */
+static int ext4_da_reserve_space(struct inode *inode, sector_t lblock)
 {
 	int retries = 0;
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
-	unsigned long md_needed, mdblocks, total = 0;
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	unsigned long md_needed, md_reserved;
 
 	/*
 	 * recalculate the amount of metadata blocks to reserve
@@ -1839,86 +1843,78 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks)
 	 * worse case is one extent per block
 	 */
 repeat:
-	spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
-	total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks;
-	mdblocks = ext4_calc_metadata_amount(inode, total);
-	BUG_ON(mdblocks < EXT4_I(inode)->i_reserved_meta_blocks);
-
-	md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks;
-	total = md_needed + nrblocks;
+	spin_lock(&ei->i_block_reservation_lock);
+	md_reserved = ei->i_reserved_meta_blocks;
+	md_needed = ext4_calc_metadata_amount(inode, lblock);
+	spin_unlock(&ei->i_block_reservation_lock);
 
 	/*
 	 * Make quota reservation here to prevent quota overflow
 	 * later. Real quota accounting is done at pages writeout
 	 * time.
 	 */
-	if (vfs_dq_reserve_block(inode, total)) {
-		spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
+	if (vfs_dq_reserve_block(inode, md_needed + 1))
 		return -EDQUOT;
-	}
 
-	if (ext4_claim_free_blocks(sbi, total)) {
-		spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
-		vfs_dq_release_reservation_block(inode, total);
+	if (ext4_claim_free_blocks(sbi, md_needed + 1)) {
+		vfs_dq_release_reservation_block(inode, md_needed + 1);
 		if (ext4_should_retry_alloc(inode->i_sb, &retries)) {
 			yield();
 			goto repeat;
 		}
 		return -ENOSPC;
 	}
-	EXT4_I(inode)->i_reserved_data_blocks += nrblocks;
-	EXT4_I(inode)->i_reserved_meta_blocks = mdblocks;
+	spin_lock(&ei->i_block_reservation_lock);
+	ei->i_reserved_data_blocks++;
+	ei->i_reserved_meta_blocks += md_needed;
+	spin_unlock(&ei->i_block_reservation_lock);
 
-	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 	return 0;       /* success */
 }
 
 static void ext4_da_release_space(struct inode *inode, int to_free)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
-	int total, mdb, mdb_free, release;
+	struct ext4_inode_info *ei = EXT4_I(inode);
 
 	if (!to_free)
 		return;		/* Nothing to release, exit */
 
 	spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
 
-	if (!EXT4_I(inode)->i_reserved_data_blocks) {
+	if (unlikely(to_free > ei->i_reserved_data_blocks)) {
 		/*
-		 * if there is no reserved blocks, but we try to free some
-		 * then the counter is messed up somewhere.
-		 * but since this function is called from invalidate
-		 * page, it's harmless to return without any action
+		 * if there aren't enough reserved blocks, then the
+		 * counter is messed up somewhere.  Since this
+		 * function is called from invalidate page, it's
+		 * harmless to return without any action.
 		 */
-		printk(KERN_INFO "ext4 delalloc try to release %d reserved "
-			    "blocks for inode %lu, but there is no reserved "
-			    "data blocks\n", to_free, inode->i_ino);
-		spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
-		return;
+		ext4_msg(inode->i_sb, KERN_NOTICE, "ext4_da_release_space: "
+			 "ino %lu, to_free %d with only %d reserved "
+			 "data blocks\n", inode->i_ino, to_free,
+			 ei->i_reserved_data_blocks);
+		WARN_ON(1);
+		to_free = ei->i_reserved_data_blocks;
 	}
+	ei->i_reserved_data_blocks -= to_free;
 
-	/* recalculate the number of metablocks still need to be reserved */
-	total = EXT4_I(inode)->i_reserved_data_blocks - to_free;
-	mdb = ext4_calc_metadata_amount(inode, total);
-
-	/* figure out how many metablocks to release */
-	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
-	mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb;
-
-	release = to_free + mdb_free;
-
-	/* update fs dirty blocks counter for truncate case */
-	percpu_counter_sub(&sbi->s_dirtyblocks_counter, release);
+	if (ei->i_reserved_data_blocks == 0) {
+		/*
+		 * We can release all of the reserved metadata blocks
+		 * only when we have written all of the delayed
+		 * allocation blocks.
+		 */
+		to_free += ei->i_reserved_meta_blocks;
+		ei->i_reserved_meta_blocks = 0;
+		ei->i_da_metadata_calc_len = 0;
+	}
 
-	/* update per-inode reservations */
-	BUG_ON(to_free > EXT4_I(inode)->i_reserved_data_blocks);
-	EXT4_I(inode)->i_reserved_data_blocks -= to_free;
+	/* update fs dirty blocks counter */
+	percpu_counter_sub(&sbi->s_dirtyblocks_counter, to_free);
 
-	BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks);
-	EXT4_I(inode)->i_reserved_meta_blocks = mdb;
 	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 
-	vfs_dq_release_reservation_block(inode, release);
+	vfs_dq_release_reservation_block(inode, to_free);
 }
 
 static void ext4_da_page_release_reservation(struct page *page,
@@ -2223,10 +2219,10 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
 	 * variables are updated after the blocks have been allocated.
 	 */
 	new.b_state = 0;
-	get_blocks_flags = (EXT4_GET_BLOCKS_CREATE |
-			    EXT4_GET_BLOCKS_DELALLOC_RESERVE);
+	get_blocks_flags = EXT4_GET_BLOCKS_CREATE;
 	if (mpd->b_state & (1 << BH_Delay))
-		get_blocks_flags |= EXT4_GET_BLOCKS_UPDATE_RESERVE_SPACE;
+		get_blocks_flags |= EXT4_GET_BLOCKS_DELALLOC_RESERVE;
+
 	blks = ext4_get_blocks(handle, mpd->inode, next, max_blocks,
 			       &new, get_blocks_flags);
 	if (blks < 0) {
@@ -2524,7 +2520,7 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock,
 		 * XXX: __block_prepare_write() unmaps passed block,
 		 * is it OK?
 		 */
-		ret = ext4_da_reserve_space(inode, 1);
+		ret = ext4_da_reserve_space(inode, iblock);
 		if (ret)
 			/* not enough space to reserve */
 			return ret;
@@ -2600,7 +2596,6 @@ static int bput_one(handle_t *handle, struct buffer_head *bh)
 }
 
 static int __ext4_journalled_writepage(struct page *page,
-				       struct writeback_control *wbc,
 				       unsigned int len)
 {
 	struct address_space *mapping = page->mapping;
@@ -2758,7 +2753,7 @@ static int ext4_writepage(struct page *page,
 		 * doesn't seem much point in redirtying the page here.
 		 */
 		ClearPageChecked(page);
-		return __ext4_journalled_writepage(page, wbc, len);
+		return __ext4_journalled_writepage(page, len);
 	}
 
 	if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode))
@@ -2788,7 +2783,7 @@ static int ext4_da_writepages_trans_blocks(struct inode *inode)
 	 * number of contiguous block. So we will limit
 	 * number of contiguous block to a sane value
 	 */
-	if (!(inode->i_flags & EXT4_EXTENTS_FL) &&
+	if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) &&
 	    (max_blocks > EXT4_MAX_TRANS_DATA))
 		max_blocks = EXT4_MAX_TRANS_DATA;
 
@@ -2999,8 +2994,7 @@ retry:
 out_writepages:
 	if (!no_nrwrite_index_update)
 		wbc->no_nrwrite_index_update = 0;
-	if (wbc->nr_to_write > nr_to_writebump)
-		wbc->nr_to_write -= nr_to_writebump;
+	wbc->nr_to_write -= nr_to_writebump;
 	wbc->range_start = range_start;
 	trace_ext4_da_writepages_result(inode, wbc, ret, pages_written);
 	return ret;
@@ -3025,11 +3019,18 @@ static int ext4_nonda_switch(struct super_block *sb)
 	if (2 * free_blocks < 3 * dirty_blocks ||
 		free_blocks < (dirty_blocks + EXT4_FREEBLOCKS_WATERMARK)) {
 		/*
-		 * free block count is less that 150% of dirty blocks
-		 * or free blocks is less that watermark
+		 * free block count is less than 150% of dirty blocks
+		 * or free blocks is less than watermark
 		 */
 		return 1;
 	}
+	/*
+	 * Even if we don't switch but are nearing capacity,
+	 * start pushing delalloc when 1/2 of free blocks are dirty.
+	 */
+	if (free_blocks < 2 * dirty_blocks)
+		writeback_inodes_sb_if_idle(sb);
+
 	return 0;
 }
 
@@ -3037,7 +3038,7 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
 			       loff_t pos, unsigned len, unsigned flags,
 			       struct page **pagep, void **fsdata)
 {
-	int ret, retries = 0;
+	int ret, retries = 0, quota_retries = 0;
 	struct page *page;
 	pgoff_t index;
 	unsigned from, to;
@@ -3091,11 +3092,27 @@ retry:
 		 * i_size_read because we hold i_mutex.
 		 */
 		if (pos + len > inode->i_size)
-			ext4_truncate(inode);
+			ext4_truncate_failed_write(inode);
 	}
 
 	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
 		goto retry;
+
+	if ((ret == -EDQUOT) &&
+	    EXT4_I(inode)->i_reserved_meta_blocks &&
+	    (quota_retries++ < 3)) {
+		/*
+		 * Since we often over-estimate the number of meta
+		 * data blocks required, we may sometimes get a
+		 * spurios out of quota error even though there would
+		 * be enough space once we write the data blocks and
+		 * find out how many meta data blocks were _really_
+		 * required.  So try forcing the inode write to see if
+		 * that helps.
+		 */
+		write_inode_now(inode, (quota_retries == 3));
+		goto retry;
+	}
 out:
 	return ret;
 }
@@ -4120,6 +4137,11 @@ static void ext4_clear_blocks(handle_t *handle, struct inode *inode,
 			      __le32 *last)
 {
 	__le32 *p;
+	int	flags = EXT4_FREE_BLOCKS_FORGET;
+
+	if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
+		flags |= EXT4_FREE_BLOCKS_METADATA;
+
 	if (try_to_extend_transaction(handle, inode)) {
 		if (bh) {
 			BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
@@ -4134,27 +4156,10 @@ static void ext4_clear_blocks(handle_t *handle, struct inode *inode,
 		}
 	}
 
-	/*
-	 * Any buffers which are on the journal will be in memory. We
-	 * find them on the hash table so jbd2_journal_revoke() will
-	 * run jbd2_journal_forget() on them.  We've already detached
-	 * each block from the file, so bforget() in
-	 * jbd2_journal_forget() should be safe.
-	 *
-	 * AKPM: turn on bforget in jbd2_journal_forget()!!!
-	 */
-	for (p = first; p < last; p++) {
-		u32 nr = le32_to_cpu(*p);
-		if (nr) {
-			struct buffer_head *tbh;
-
-			*p = 0;
-			tbh = sb_find_get_block(inode->i_sb, nr);
-			ext4_forget(handle, 0, inode, tbh, nr);
-		}
-	}
+	for (p = first; p < last; p++)
+		*p = 0;
 
-	ext4_free_blocks(handle, inode, block_to_free, count, 0);
+	ext4_free_blocks(handle, inode, 0, block_to_free, count, flags);
 }
 
 /**
@@ -4342,7 +4347,8 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 					    blocks_for_truncate(inode));
 			}
 
-			ext4_free_blocks(handle, inode, nr, 1, 1);
+			ext4_free_blocks(handle, inode, 0, nr, 1,
+					 EXT4_FREE_BLOCKS_METADATA);
 
 			if (parent_bh) {
 				/*
@@ -4781,8 +4787,8 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	struct ext4_iloc iloc;
 	struct ext4_inode *raw_inode;
 	struct ext4_inode_info *ei;
-	struct buffer_head *bh;
 	struct inode *inode;
+	journal_t *journal = EXT4_SB(sb)->s_journal;
 	long ret;
 	int block;
 
@@ -4793,11 +4799,11 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		return inode;
 
 	ei = EXT4_I(inode);
+	iloc.bh = 0;
 
 	ret = __ext4_get_inode_loc(inode, &iloc, 0);
 	if (ret < 0)
 		goto bad_inode;
-	bh = iloc.bh;
 	raw_inode = ext4_raw_inode(&iloc);
 	inode->i_mode = le16_to_cpu(raw_inode->i_mode);
 	inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low);
@@ -4820,7 +4826,6 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		if (inode->i_mode == 0 ||
 		    !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) {
 			/* this inode is deleted */
-			brelse(bh);
 			ret = -ESTALE;
 			goto bad_inode;
 		}
@@ -4837,6 +4842,9 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 			((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32;
 	inode->i_size = ext4_isize(raw_inode);
 	ei->i_disksize = inode->i_size;
+#ifdef CONFIG_QUOTA
+	ei->i_reserved_quota = 0;
+#endif
 	inode->i_generation = le32_to_cpu(raw_inode->i_generation);
 	ei->i_block_group = iloc.block_group;
 	ei->i_last_alloc_group = ~0;
@@ -4848,11 +4856,35 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		ei->i_data[block] = raw_inode->i_block[block];
 	INIT_LIST_HEAD(&ei->i_orphan);
 
+	/*
+	 * Set transaction id's of transactions that have to be committed
+	 * to finish f[data]sync. We set them to currently running transaction
+	 * as we cannot be sure that the inode or some of its metadata isn't
+	 * part of the transaction - the inode could have been reclaimed and
+	 * now it is reread from disk.
+	 */
+	if (journal) {
+		transaction_t *transaction;
+		tid_t tid;
+
+		spin_lock(&journal->j_state_lock);
+		if (journal->j_running_transaction)
+			transaction = journal->j_running_transaction;
+		else
+			transaction = journal->j_committing_transaction;
+		if (transaction)
+			tid = transaction->t_tid;
+		else
+			tid = journal->j_commit_sequence;
+		spin_unlock(&journal->j_state_lock);
+		ei->i_sync_tid = tid;
+		ei->i_datasync_tid = tid;
+	}
+
 	if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) {
 		ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize);
 		if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize >
 		    EXT4_INODE_SIZE(inode->i_sb)) {
-			brelse(bh);
 			ret = -EIO;
 			goto bad_inode;
 		}
@@ -4884,10 +4916,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 
 	ret = 0;
 	if (ei->i_file_acl &&
-	    ((ei->i_file_acl <
-	      (le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block) +
-	       EXT4_SB(sb)->s_gdb_count)) ||
-	     (ei->i_file_acl >= ext4_blocks_count(EXT4_SB(sb)->s_es)))) {
+	    !ext4_data_block_valid(EXT4_SB(sb), ei->i_file_acl, 1)) {
 		ext4_error(sb, __func__,
 			   "bad extended attribute block %llu in inode #%lu",
 			   ei->i_file_acl, inode->i_ino);
@@ -4905,10 +4934,8 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		/* Validate block references which are part of inode */
 		ret = ext4_check_inode_blockref(inode);
 	}
-	if (ret) {
-		brelse(bh);
+	if (ret)
 		goto bad_inode;
-	}
 
 	if (S_ISREG(inode->i_mode)) {
 		inode->i_op = &ext4_file_inode_operations;
@@ -4936,7 +4963,6 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 			init_special_inode(inode, inode->i_mode,
 			   new_decode_dev(le32_to_cpu(raw_inode->i_block[1])));
 	} else {
-		brelse(bh);
 		ret = -EIO;
 		ext4_error(inode->i_sb, __func__,
 			   "bogus i_mode (%o) for inode=%lu",
@@ -4949,6 +4975,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	return inode;
 
 bad_inode:
+	brelse(iloc.bh);
 	iget_failed(inode);
 	return ERR_PTR(ret);
 }
@@ -5108,6 +5135,7 @@ static int ext4_do_update_inode(handle_t *handle,
 		err = rc;
 	ei->i_state &= ~EXT4_STATE_NEW;
 
+	ext4_update_inode_fsync_trans(handle, inode, 0);
 out_brelse:
 	brelse(bh);
 	ext4_std_error(inode->i_sb, err);
@@ -5227,8 +5255,8 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
 
 		/* (user+group)*(old+new) structure, inode write (sb,
 		 * inode block, ? - but truncate inode update has it) */
-		handle = ext4_journal_start(inode, 2*(EXT4_QUOTA_INIT_BLOCKS(inode->i_sb)+
-					EXT4_QUOTA_DEL_BLOCKS(inode->i_sb))+3);
+		handle = ext4_journal_start(inode, (EXT4_MAXQUOTAS_INIT_BLOCKS(inode->i_sb)+
+					EXT4_MAXQUOTAS_DEL_BLOCKS(inode->i_sb))+3);
 		if (IS_ERR(handle)) {
 			error = PTR_ERR(handle);
 			goto err_out;
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index c1cdf613e725..b63d193126db 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -221,31 +221,38 @@ setversion_out:
 		struct file *donor_filp;
 		int err;
 
+		if (!(filp->f_mode & FMODE_READ) ||
+		    !(filp->f_mode & FMODE_WRITE))
+			return -EBADF;
+
 		if (copy_from_user(&me,
 			(struct move_extent __user *)arg, sizeof(me)))
 			return -EFAULT;
+		me.moved_len = 0;
 
 		donor_filp = fget(me.donor_fd);
 		if (!donor_filp)
 			return -EBADF;
 
-		if (!capable(CAP_DAC_OVERRIDE)) {
-			if ((current->real_cred->fsuid != inode->i_uid) ||
-				!(inode->i_mode & S_IRUSR) ||
-				!(donor_filp->f_dentry->d_inode->i_mode &
-				S_IRUSR)) {
-				fput(donor_filp);
-				return -EACCES;
-			}
+		if (!(donor_filp->f_mode & FMODE_WRITE)) {
+			err = -EBADF;
+			goto mext_out;
 		}
 
+		err = mnt_want_write(filp->f_path.mnt);
+		if (err)
+			goto mext_out;
+
 		err = ext4_move_extents(filp, donor_filp, me.orig_start,
 					me.donor_start, me.len, &me.moved_len);
-		fput(donor_filp);
+		mnt_drop_write(filp->f_path.mnt);
+		if (me.moved_len > 0)
+			file_remove_suid(donor_filp);
 
 		if (copy_to_user((struct move_extent *)arg, &me, sizeof(me)))
-			return -EFAULT;
-
+			err = -EFAULT;
+mext_out:
+		fput(donor_filp);
 		return err;
 	}
 
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 74e495dabe09..d34afad3e137 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2529,7 +2529,6 @@ static void release_blocks_on_commit(journal_t *journal, transaction_t *txn)
 	struct ext4_group_info *db;
 	int err, count = 0, count2 = 0;
 	struct ext4_free_data *entry;
-	ext4_fsblk_t discard_block;
 	struct list_head *l, *ltmp;
 
 	list_for_each_safe(l, ltmp, &txn->t_private_list) {
@@ -2559,13 +2558,19 @@ static void release_blocks_on_commit(journal_t *journal, transaction_t *txn)
 			page_cache_release(e4b.bd_bitmap_page);
 		}
 		ext4_unlock_group(sb, entry->group);
-		discard_block = (ext4_fsblk_t) entry->group * EXT4_BLOCKS_PER_GROUP(sb)
-			+ entry->start_blk
-			+ le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
-		trace_ext4_discard_blocks(sb, (unsigned long long)discard_block,
-					  entry->count);
-		sb_issue_discard(sb, discard_block, entry->count);
-
+		if (test_opt(sb, DISCARD)) {
+			ext4_fsblk_t discard_block;
+			struct ext4_super_block *es = EXT4_SB(sb)->s_es;
+
+			discard_block = (ext4_fsblk_t)entry->group *
+						EXT4_BLOCKS_PER_GROUP(sb)
+					+ entry->start_blk
+					+ le32_to_cpu(es->s_first_data_block);
+			trace_ext4_discard_blocks(sb,
+					(unsigned long long)discard_block,
+					entry->count);
+			sb_issue_discard(sb, discard_block, entry->count);
+		}
 		kmem_cache_free(ext4_free_ext_cachep, entry);
 		ext4_mb_release_desc(&e4b);
 	}
@@ -2750,12 +2755,6 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 	if (!(ac->ac_flags & EXT4_MB_DELALLOC_RESERVED))
 		/* release all the reserved blocks if non delalloc */
 		percpu_counter_sub(&sbi->s_dirtyblocks_counter, reserv_blks);
-	else {
-		percpu_counter_sub(&sbi->s_dirtyblocks_counter,
-						ac->ac_b_ex.fe_len);
-		/* convert reserved quota blocks to real quota blocks */
-		vfs_dq_claim_block(ac->ac_inode, ac->ac_b_ex.fe_len);
-	}
 
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi,
@@ -3006,6 +3005,24 @@ static void ext4_mb_collect_stats(struct ext4_allocation_context *ac)
 }
 
 /*
+ * Called on failure; free up any blocks from the inode PA for this
+ * context.  We don't need this for MB_GROUP_PA because we only change
+ * pa_free in ext4_mb_release_context(), but on failure, we've already
+ * zeroed out ac->ac_b_ex.fe_len, so group_pa->pa_free is not changed.
+ */
+static void ext4_discard_allocated_blocks(struct ext4_allocation_context *ac)
+{
+	struct ext4_prealloc_space *pa = ac->ac_pa;
+	int len;
+
+	if (pa && pa->pa_type == MB_INODE_PA) {
+		len = ac->ac_b_ex.fe_len;
+		pa->pa_free += len;
+	}
+
+}
+
+/*
  * use blocks preallocated to inode
  */
 static void ext4_mb_use_inode_pa(struct ext4_allocation_context *ac,
@@ -3932,7 +3949,7 @@ static void ext4_mb_group_or_file(struct ext4_allocation_context *ac)
 	 * per cpu locality group is to reduce the contention between block
 	 * request from multiple CPUs.
 	 */
-	ac->ac_lg = per_cpu_ptr(sbi->s_locality_groups, raw_smp_processor_id());
+	ac->ac_lg = __this_cpu_ptr(sbi->s_locality_groups);
 
 	/* we're going to use group allocation */
 	ac->ac_flags |= EXT4_MB_HINT_GROUP_ALLOC;
@@ -4290,6 +4307,7 @@ repeat:
 			ac->ac_status = AC_STATUS_CONTINUE;
 			goto repeat;
 		} else if (*errp) {
+			ext4_discard_allocated_blocks(ac);
 			ac->ac_b_ex.fe_len = 0;
 			ar->len = 0;
 			ext4_mb_show_ac(ac);
@@ -4422,18 +4440,24 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b,
 	return 0;
 }
 
-/*
- * Main entry point into mballoc to free blocks
+/**
+ * ext4_free_blocks() -- Free given blocks and update quota
+ * @handle:		handle for this transaction
+ * @inode:		inode
+ * @block:		start physical block to free
+ * @count:		number of blocks to count
+ * @metadata: 		Are these metadata blocks
  */
-void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t block, unsigned long count,
-			int metadata, unsigned long *freed)
+void ext4_free_blocks(handle_t *handle, struct inode *inode,
+		      struct buffer_head *bh, ext4_fsblk_t block,
+		      unsigned long count, int flags)
 {
 	struct buffer_head *bitmap_bh = NULL;
 	struct super_block *sb = inode->i_sb;
 	struct ext4_allocation_context *ac = NULL;
 	struct ext4_group_desc *gdp;
 	struct ext4_super_block *es;
+	unsigned long freed = 0;
 	unsigned int overflow;
 	ext4_grpblk_t bit;
 	struct buffer_head *gd_bh;
@@ -4443,13 +4467,16 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
 	int err = 0;
 	int ret;
 
-	*freed = 0;
+	if (bh) {
+		if (block)
+			BUG_ON(block != bh->b_blocknr);
+		else
+			block = bh->b_blocknr;
+	}
 
 	sbi = EXT4_SB(sb);
 	es = EXT4_SB(sb)->s_es;
-	if (block < le32_to_cpu(es->s_first_data_block) ||
-	    block + count < block ||
-	    block + count > ext4_blocks_count(es)) {
+	if (!ext4_data_block_valid(sbi, block, count)) {
 		ext4_error(sb, __func__,
 			    "Freeing blocks not in datazone - "
 			    "block = %llu, count = %lu", block, count);
@@ -4457,7 +4484,32 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
 	}
 
 	ext4_debug("freeing block %llu\n", block);
-	trace_ext4_free_blocks(inode, block, count, metadata);
+	trace_ext4_free_blocks(inode, block, count, flags);
+
+	if (flags & EXT4_FREE_BLOCKS_FORGET) {
+		struct buffer_head *tbh = bh;
+		int i;
+
+		BUG_ON(bh && (count > 1));
+
+		for (i = 0; i < count; i++) {
+			if (!bh)
+				tbh = sb_find_get_block(inode->i_sb,
+							block + i);
+			ext4_forget(handle, flags & EXT4_FREE_BLOCKS_METADATA, 
+				    inode, tbh, block + i);
+		}
+	}
+
+	/* 
+	 * We need to make sure we don't reuse the freed block until
+	 * after the transaction is committed, which we can do by
+	 * treating the block as metadata, below.  We make an
+	 * exception if the inode is to be written in writeback mode
+	 * since writeback mode has weak data consistency guarantees.
+	 */
+	if (!ext4_should_writeback_data(inode))
+		flags |= EXT4_FREE_BLOCKS_METADATA;
 
 	ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS);
 	if (ac) {
@@ -4533,7 +4585,8 @@ do_more:
 	err = ext4_mb_load_buddy(sb, block_group, &e4b);
 	if (err)
 		goto error_return;
-	if (metadata && ext4_handle_valid(handle)) {
+
+	if ((flags & EXT4_FREE_BLOCKS_METADATA) && ext4_handle_valid(handle)) {
 		struct ext4_free_data *new_entry;
 		/*
 		 * blocks being freed are metadata. these blocks shouldn't
@@ -4572,7 +4625,7 @@ do_more:
 
 	ext4_mb_release_desc(&e4b);
 
-	*freed += count;
+	freed += count;
 
 	/* We dirtied the bitmap block */
 	BUFFER_TRACE(bitmap_bh, "dirtied bitmap block");
@@ -4592,6 +4645,8 @@ do_more:
 	}
 	sb->s_dirt = 1;
 error_return:
+	if (freed)
+		vfs_dq_free_block(inode, freed);
 	brelse(bitmap_bh);
 	ext4_std_error(sb, err);
 	if (ac)
diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h
index 0ca811061bc7..436521cae456 100644
--- a/fs/ext4/mballoc.h
+++ b/fs/ext4/mballoc.h
@@ -17,7 +17,6 @@
 #include <linux/proc_fs.h>
 #include <linux/pagemap.h>
 #include <linux/seq_file.h>
-#include <linux/version.h>
 #include <linux/blkdev.h>
 #include <linux/mutex.h>
 #include "ext4_jbd2.h"
diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c
index a93d5b80f3e2..81415814b00b 100644
--- a/fs/ext4/migrate.c
+++ b/fs/ext4/migrate.c
@@ -238,7 +238,7 @@ static int extend_credit_for_blkdel(handle_t *handle, struct inode *inode)
 	 * So allocate a credit of 3. We may update
 	 * quota (user and group).
 	 */
-	needed = 3 + 2*EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb);
+	needed = 3 + EXT4_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb);
 
 	if (ext4_journal_extend(handle, needed) != 0)
 		retval = ext4_journal_restart(handle, needed);
@@ -262,13 +262,17 @@ static int free_dind_blocks(handle_t *handle,
 	for (i = 0; i < max_entries; i++) {
 		if (tmp_idata[i]) {
 			extend_credit_for_blkdel(handle, inode);
-			ext4_free_blocks(handle, inode,
-					le32_to_cpu(tmp_idata[i]), 1, 1);
+			ext4_free_blocks(handle, inode, 0,
+					 le32_to_cpu(tmp_idata[i]), 1,
+					 EXT4_FREE_BLOCKS_METADATA |
+					 EXT4_FREE_BLOCKS_FORGET);
 		}
 	}
 	put_bh(bh);
 	extend_credit_for_blkdel(handle, inode);
-	ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1, 1);
+	ext4_free_blocks(handle, inode, 0, le32_to_cpu(i_data), 1,
+			 EXT4_FREE_BLOCKS_METADATA |
+			 EXT4_FREE_BLOCKS_FORGET);
 	return 0;
 }
 
@@ -297,7 +301,9 @@ static int free_tind_blocks(handle_t *handle,
 	}
 	put_bh(bh);
 	extend_credit_for_blkdel(handle, inode);
-	ext4_free_blocks(handle, inode, le32_to_cpu(i_data), 1, 1);
+	ext4_free_blocks(handle, inode, 0, le32_to_cpu(i_data), 1,
+			 EXT4_FREE_BLOCKS_METADATA |
+			 EXT4_FREE_BLOCKS_FORGET);
 	return 0;
 }
 
@@ -308,8 +314,10 @@ static int free_ind_block(handle_t *handle, struct inode *inode, __le32 *i_data)
 	/* ei->i_data[EXT4_IND_BLOCK] */
 	if (i_data[0]) {
 		extend_credit_for_blkdel(handle, inode);
-		ext4_free_blocks(handle, inode,
-				le32_to_cpu(i_data[0]), 1, 1);
+		ext4_free_blocks(handle, inode, 0,
+				le32_to_cpu(i_data[0]), 1,
+				 EXT4_FREE_BLOCKS_METADATA |
+				 EXT4_FREE_BLOCKS_FORGET);
 	}
 
 	/* ei->i_data[EXT4_DIND_BLOCK] */
@@ -419,7 +427,8 @@ static int free_ext_idx(handle_t *handle, struct inode *inode,
 	}
 	put_bh(bh);
 	extend_credit_for_blkdel(handle, inode);
-	ext4_free_blocks(handle, inode, block, 1, 1);
+	ext4_free_blocks(handle, inode, 0, block, 1,
+			 EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET);
 	return retval;
 }
 
@@ -477,7 +486,7 @@ int ext4_ext_migrate(struct inode *inode)
 	handle = ext4_journal_start(inode,
 					EXT4_DATA_TRANS_BLOCKS(inode->i_sb) +
 					EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2 * EXT4_QUOTA_INIT_BLOCKS(inode->i_sb)
+					EXT4_MAXQUOTAS_INIT_BLOCKS(inode->i_sb)
 					+ 1);
 	if (IS_ERR(handle)) {
 		retval = PTR_ERR(handle);
diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c
index 25b6b1457360..82c415be87a4 100644
--- a/fs/ext4/move_extent.c
+++ b/fs/ext4/move_extent.c
@@ -77,12 +77,14 @@ static int
 mext_next_extent(struct inode *inode, struct ext4_ext_path *path,
 		      struct ext4_extent **extent)
 {
+	struct ext4_extent_header *eh;
 	int ppos, leaf_ppos = path->p_depth;
 
 	ppos = leaf_ppos;
 	if (EXT_LAST_EXTENT(path[ppos].p_hdr) > path[ppos].p_ext) {
 		/* leaf block */
 		*extent = ++path[ppos].p_ext;
+		path[ppos].p_block = ext_pblock(path[ppos].p_ext);
 		return 0;
 	}
 
@@ -119,9 +121,18 @@ mext_next_extent(struct inode *inode, struct ext4_ext_path *path,
 					ext_block_hdr(path[cur_ppos+1].p_bh);
 			}
 
+			path[leaf_ppos].p_ext = *extent = NULL;
+
+			eh = path[leaf_ppos].p_hdr;
+			if (le16_to_cpu(eh->eh_entries) == 0)
+				/* empty leaf is found */
+				return -ENODATA;
+
 			/* leaf block */
 			path[leaf_ppos].p_ext = *extent =
 				EXT_FIRST_EXTENT(path[leaf_ppos].p_hdr);
+			path[leaf_ppos].p_block =
+					ext_pblock(path[leaf_ppos].p_ext);
 			return 0;
 		}
 	}
@@ -155,40 +166,15 @@ mext_check_null_inode(struct inode *inode1, struct inode *inode2,
 }
 
 /**
- * mext_double_down_read - Acquire two inodes' read semaphore
- *
- * @orig_inode:		original inode structure
- * @donor_inode:	donor inode structure
- * Acquire read semaphore of the two inodes (orig and donor) by i_ino order.
- */
-static void
-mext_double_down_read(struct inode *orig_inode, struct inode *donor_inode)
-{
-	struct inode *first = orig_inode, *second = donor_inode;
-
-	/*
-	 * Use the inode number to provide the stable locking order instead
-	 * of its address, because the C language doesn't guarantee you can
-	 * compare pointers that don't come from the same array.
-	 */
-	if (donor_inode->i_ino < orig_inode->i_ino) {
-		first = donor_inode;
-		second = orig_inode;
-	}
-
-	down_read(&EXT4_I(first)->i_data_sem);
-	down_read(&EXT4_I(second)->i_data_sem);
-}
-
-/**
- * mext_double_down_write - Acquire two inodes' write semaphore
+ * double_down_write_data_sem - Acquire two inodes' write lock of i_data_sem
  *
  * @orig_inode:		original inode structure
  * @donor_inode:	donor inode structure
- * Acquire write semaphore of the two inodes (orig and donor) by i_ino order.
+ * Acquire write lock of i_data_sem of the two inodes (orig and donor) by
+ * i_ino order.
  */
 static void
-mext_double_down_write(struct inode *orig_inode, struct inode *donor_inode)
+double_down_write_data_sem(struct inode *orig_inode, struct inode *donor_inode)
 {
 	struct inode *first = orig_inode, *second = donor_inode;
 
@@ -203,32 +189,18 @@ mext_double_down_write(struct inode *orig_inode, struct inode *donor_inode)
 	}
 
 	down_write(&EXT4_I(first)->i_data_sem);
-	down_write(&EXT4_I(second)->i_data_sem);
+	down_write_nested(&EXT4_I(second)->i_data_sem, SINGLE_DEPTH_NESTING);
 }
 
 /**
- * mext_double_up_read - Release two inodes' read semaphore
+ * double_up_write_data_sem - Release two inodes' write lock of i_data_sem
  *
  * @orig_inode:		original inode structure to be released its lock first
  * @donor_inode:	donor inode structure to be released its lock second
- * Release read semaphore of two inodes (orig and donor).
+ * Release write lock of i_data_sem of two inodes (orig and donor).
  */
 static void
-mext_double_up_read(struct inode *orig_inode, struct inode *donor_inode)
-{
-	up_read(&EXT4_I(orig_inode)->i_data_sem);
-	up_read(&EXT4_I(donor_inode)->i_data_sem);
-}
-
-/**
- * mext_double_up_write - Release two inodes' write semaphore
- *
- * @orig_inode:		original inode structure to be released its lock first
- * @donor_inode:	donor inode structure to be released its lock second
- * Release write semaphore of two inodes (orig and donor).
- */
-static void
-mext_double_up_write(struct inode *orig_inode, struct inode *donor_inode)
+double_up_write_data_sem(struct inode *orig_inode, struct inode *donor_inode)
 {
 	up_write(&EXT4_I(orig_inode)->i_data_sem);
 	up_write(&EXT4_I(donor_inode)->i_data_sem);
@@ -596,7 +568,7 @@ out:
  * @tmp_oext:		the extent that will belong to the donor inode
  * @orig_off:		block offset of original inode
  * @donor_off:		block offset of donor inode
- * @max_count:		the maximun length of extents
+ * @max_count:		the maximum length of extents
  *
  * Return 0 on success, or a negative error value on failure.
  */
@@ -661,6 +633,7 @@ mext_calc_swap_extents(struct ext4_extent *tmp_dext,
  * @donor_inode:	donor inode
  * @from:		block offset of orig_inode
  * @count:		block count to be replaced
+ * @err:		pointer to save return value
  *
  * Replace original inode extents and donor inode extents page by page.
  * We implement this replacement in the following three steps:
@@ -671,33 +644,33 @@ mext_calc_swap_extents(struct ext4_extent *tmp_dext,
  * 3. Change the block information of donor inode to point at the saved
  *    original inode blocks in the dummy extents.
  *
- * Return 0 on success, or a negative error value on failure.
+ * Return replaced block count.
  */
 static int
 mext_replace_branches(handle_t *handle, struct inode *orig_inode,
 			   struct inode *donor_inode, ext4_lblk_t from,
-			   ext4_lblk_t count)
+			   ext4_lblk_t count, int *err)
 {
 	struct ext4_ext_path *orig_path = NULL;
 	struct ext4_ext_path *donor_path = NULL;
 	struct ext4_extent *oext, *dext;
 	struct ext4_extent tmp_dext, tmp_oext;
 	ext4_lblk_t orig_off = from, donor_off = from;
-	int err = 0;
 	int depth;
 	int replaced_count = 0;
 	int dext_alen;
 
-	mext_double_down_write(orig_inode, donor_inode);
+	/* Protect extent trees against block allocations via delalloc */
+	double_down_write_data_sem(orig_inode, donor_inode);
 
 	/* Get the original extent for the block "orig_off" */
-	err = get_ext_path(orig_inode, orig_off, &orig_path);
-	if (err)
+	*err = get_ext_path(orig_inode, orig_off, &orig_path);
+	if (*err)
 		goto out;
 
 	/* Get the donor extent for the head */
-	err = get_ext_path(donor_inode, donor_off, &donor_path);
-	if (err)
+	*err = get_ext_path(donor_inode, donor_off, &donor_path);
+	if (*err)
 		goto out;
 	depth = ext_depth(orig_inode);
 	oext = orig_path[depth].p_ext;
@@ -707,9 +680,9 @@ mext_replace_branches(handle_t *handle, struct inode *orig_inode,
 	dext = donor_path[depth].p_ext;
 	tmp_dext = *dext;
 
-	err = mext_calc_swap_extents(&tmp_dext, &tmp_oext, orig_off,
+	*err = mext_calc_swap_extents(&tmp_dext, &tmp_oext, orig_off,
 				      donor_off, count);
-	if (err)
+	if (*err)
 		goto out;
 
 	/* Loop for the donor extents */
@@ -718,7 +691,7 @@ mext_replace_branches(handle_t *handle, struct inode *orig_inode,
 		if (!dext) {
 			ext4_error(donor_inode->i_sb, __func__,
 				   "The extent for donor must be found");
-			err = -EIO;
+			*err = -EIO;
 			goto out;
 		} else if (donor_off != le32_to_cpu(tmp_dext.ee_block)) {
 			ext4_error(donor_inode->i_sb, __func__,
@@ -726,20 +699,20 @@ mext_replace_branches(handle_t *handle, struct inode *orig_inode,
 				"extent(%u) should be equal",
 				donor_off,
 				le32_to_cpu(tmp_dext.ee_block));
-			err = -EIO;
+			*err = -EIO;
 			goto out;
 		}
 
 		/* Set donor extent to orig extent */
-		err = mext_leaf_block(handle, orig_inode,
+		*err = mext_leaf_block(handle, orig_inode,
 					   orig_path, &tmp_dext, &orig_off);
-		if (err < 0)
+		if (*err)
 			goto out;
 
 		/* Set orig extent to donor extent */
-		err = mext_leaf_block(handle, donor_inode,
+		*err = mext_leaf_block(handle, donor_inode,
 					   donor_path, &tmp_oext, &donor_off);
-		if (err < 0)
+		if (*err)
 			goto out;
 
 		dext_alen = ext4_ext_get_actual_len(&tmp_dext);
@@ -753,35 +726,25 @@ mext_replace_branches(handle_t *handle, struct inode *orig_inode,
 
 		if (orig_path)
 			ext4_ext_drop_refs(orig_path);
-		err = get_ext_path(orig_inode, orig_off, &orig_path);
-		if (err)
+		*err = get_ext_path(orig_inode, orig_off, &orig_path);
+		if (*err)
 			goto out;
 		depth = ext_depth(orig_inode);
 		oext = orig_path[depth].p_ext;
-		if (le32_to_cpu(oext->ee_block) +
-				ext4_ext_get_actual_len(oext) <= orig_off) {
-			err = 0;
-			goto out;
-		}
 		tmp_oext = *oext;
 
 		if (donor_path)
 			ext4_ext_drop_refs(donor_path);
-		err = get_ext_path(donor_inode, donor_off, &donor_path);
-		if (err)
+		*err = get_ext_path(donor_inode, donor_off, &donor_path);
+		if (*err)
 			goto out;
 		depth = ext_depth(donor_inode);
 		dext = donor_path[depth].p_ext;
-		if (le32_to_cpu(dext->ee_block) +
-				ext4_ext_get_actual_len(dext) <= donor_off) {
-			err = 0;
-			goto out;
-		}
 		tmp_dext = *dext;
 
-		err = mext_calc_swap_extents(&tmp_dext, &tmp_oext, orig_off,
+		*err = mext_calc_swap_extents(&tmp_dext, &tmp_oext, orig_off,
 					   donor_off, count - replaced_count);
-		if (err)
+		if (*err)
 			goto out;
 	}
 
@@ -795,8 +758,12 @@ out:
 		kfree(donor_path);
 	}
 
-	mext_double_up_write(orig_inode, donor_inode);
-	return err;
+	ext4_ext_invalidate_cache(orig_inode);
+	ext4_ext_invalidate_cache(donor_inode);
+
+	double_up_write_data_sem(orig_inode, donor_inode);
+
+	return replaced_count;
 }
 
 /**
@@ -808,16 +775,17 @@ out:
  * @data_offset_in_page:	block index where data swapping starts
  * @block_len_in_page:		the number of blocks to be swapped
  * @uninit:			orig extent is uninitialized or not
+ * @err:			pointer to save return value
  *
  * Save the data in original inode blocks and replace original inode extents
  * with donor inode extents by calling mext_replace_branches().
- * Finally, write out the saved data in new original inode blocks. Return 0
- * on success, or a negative error value on failure.
+ * Finally, write out the saved data in new original inode blocks. Return
+ * replaced block count.
  */
 static int
 move_extent_per_page(struct file *o_filp, struct inode *donor_inode,
 		  pgoff_t orig_page_offset, int data_offset_in_page,
-		  int block_len_in_page, int uninit)
+		  int block_len_in_page, int uninit, int *err)
 {
 	struct inode *orig_inode = o_filp->f_dentry->d_inode;
 	struct address_space *mapping = orig_inode->i_mapping;
@@ -829,9 +797,11 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode,
 	long long offs = orig_page_offset << PAGE_CACHE_SHIFT;
 	unsigned long blocksize = orig_inode->i_sb->s_blocksize;
 	unsigned int w_flags = 0;
-	unsigned int tmp_data_len, data_len;
+	unsigned int tmp_data_size, data_size, replaced_size;
 	void *fsdata;
-	int ret, i, jblocks;
+	int i, jblocks;
+	int err2 = 0;
+	int replaced_count = 0;
 	int blocks_per_page = PAGE_CACHE_SIZE >> orig_inode->i_blkbits;
 
 	/*
@@ -841,8 +811,8 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode,
 	jblocks = ext4_writepage_trans_blocks(orig_inode) * 2;
 	handle = ext4_journal_start(orig_inode, jblocks);
 	if (IS_ERR(handle)) {
-		ret = PTR_ERR(handle);
-		return ret;
+		*err = PTR_ERR(handle);
+		return 0;
 	}
 
 	if (segment_eq(get_fs(), KERNEL_DS))
@@ -858,39 +828,36 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode,
 	 * Just swap data blocks between orig and donor.
 	 */
 	if (uninit) {
-		ret = mext_replace_branches(handle, orig_inode,
-						 donor_inode, orig_blk_offset,
-						 block_len_in_page);
-
-		/* Clear the inode cache not to refer to the old data */
-		ext4_ext_invalidate_cache(orig_inode);
-		ext4_ext_invalidate_cache(donor_inode);
+		replaced_count = mext_replace_branches(handle, orig_inode,
+						donor_inode, orig_blk_offset,
+						block_len_in_page, err);
 		goto out2;
 	}
 
 	offs = (long long)orig_blk_offset << orig_inode->i_blkbits;
 
-	/* Calculate data_len */
+	/* Calculate data_size */
 	if ((orig_blk_offset + block_len_in_page - 1) ==
 	    ((orig_inode->i_size - 1) >> orig_inode->i_blkbits)) {
 		/* Replace the last block */
-		tmp_data_len = orig_inode->i_size & (blocksize - 1);
+		tmp_data_size = orig_inode->i_size & (blocksize - 1);
 		/*
-		 * If data_len equal zero, it shows data_len is multiples of
+		 * If data_size equal zero, it shows data_size is multiples of
 		 * blocksize. So we set appropriate value.
 		 */
-		if (tmp_data_len == 0)
-			tmp_data_len = blocksize;
+		if (tmp_data_size == 0)
+			tmp_data_size = blocksize;
 
-		data_len = tmp_data_len +
+		data_size = tmp_data_size +
 			((block_len_in_page - 1) << orig_inode->i_blkbits);
-	} else {
-		data_len = block_len_in_page << orig_inode->i_blkbits;
-	}
+	} else
+		data_size = block_len_in_page << orig_inode->i_blkbits;
+
+	replaced_size = data_size;
 
-	ret = a_ops->write_begin(o_filp, mapping, offs, data_len, w_flags,
+	*err = a_ops->write_begin(o_filp, mapping, offs, data_size, w_flags,
 				 &page, &fsdata);
-	if (unlikely(ret < 0))
+	if (unlikely(*err < 0))
 		goto out;
 
 	if (!PageUptodate(page)) {
@@ -911,14 +878,17 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode,
 	/* Release old bh and drop refs */
 	try_to_release_page(page, 0);
 
-	ret = mext_replace_branches(handle, orig_inode, donor_inode,
-					 orig_blk_offset, block_len_in_page);
-	if (ret < 0)
-		goto out;
-
-	/* Clear the inode cache not to refer to the old data */
-	ext4_ext_invalidate_cache(orig_inode);
-	ext4_ext_invalidate_cache(donor_inode);
+	replaced_count = mext_replace_branches(handle, orig_inode, donor_inode,
+					orig_blk_offset, block_len_in_page,
+					&err2);
+	if (err2) {
+		if (replaced_count) {
+			block_len_in_page = replaced_count;
+			replaced_size =
+				block_len_in_page << orig_inode->i_blkbits;
+		} else
+			goto out;
+	}
 
 	if (!page_has_buffers(page))
 		create_empty_buffers(page, 1 << orig_inode->i_blkbits, 0);
@@ -928,16 +898,16 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode,
 		bh = bh->b_this_page;
 
 	for (i = 0; i < block_len_in_page; i++) {
-		ret = ext4_get_block(orig_inode,
+		*err = ext4_get_block(orig_inode,
 				(sector_t)(orig_blk_offset + i), bh, 0);
-		if (ret < 0)
+		if (*err < 0)
 			goto out;
 
 		if (bh->b_this_page != NULL)
 			bh = bh->b_this_page;
 	}
 
-	ret = a_ops->write_end(o_filp, mapping, offs, data_len, data_len,
+	*err = a_ops->write_end(o_filp, mapping, offs, data_size, replaced_size,
 			       page, fsdata);
 	page = NULL;
 
@@ -951,7 +921,10 @@ out:
 out2:
 	ext4_journal_stop(handle);
 
-	return ret < 0 ? ret : 0;
+	if (err2)
+		*err = err2;
+
+	return replaced_count;
 }
 
 /**
@@ -962,7 +935,6 @@ out2:
  * @orig_start:		logical start offset in block for orig
  * @donor_start:	logical start offset in block for donor
  * @len:		the number of blocks to be moved
- * @moved_len:		moved block length
  *
  * Check the arguments of ext4_move_extents() whether the files can be
  * exchanged with each other.
@@ -970,8 +942,8 @@ out2:
  */
 static int
 mext_check_arguments(struct inode *orig_inode,
-			  struct inode *donor_inode, __u64 orig_start,
-			  __u64 donor_start, __u64 *len, __u64 moved_len)
+		     struct inode *donor_inode, __u64 orig_start,
+		     __u64 donor_start, __u64 *len)
 {
 	ext4_lblk_t orig_blocks, donor_blocks;
 	unsigned int blkbits = orig_inode->i_blkbits;
@@ -985,6 +957,13 @@ mext_check_arguments(struct inode *orig_inode,
 		return -EINVAL;
 	}
 
+	if (donor_inode->i_mode & (S_ISUID|S_ISGID)) {
+		ext4_debug("ext4 move extent: suid or sgid is set"
+			   " to donor file [ino:orig %lu, donor %lu]\n",
+			   orig_inode->i_ino, donor_inode->i_ino);
+		return -EINVAL;
+	}
+
 	/* Ext4 move extent does not support swapfile */
 	if (IS_SWAPFILE(orig_inode) || IS_SWAPFILE(donor_inode)) {
 		ext4_debug("ext4 move extent: The argument files should "
@@ -1025,13 +1004,6 @@ mext_check_arguments(struct inode *orig_inode,
 		return -EINVAL;
 	}
 
-	if (moved_len) {
-		ext4_debug("ext4 move extent: moved_len should be 0 "
-			"[ino:orig %lu, donor %lu]\n", orig_inode->i_ino,
-			donor_inode->i_ino);
-		return -EINVAL;
-	}
-
 	if ((orig_start > EXT_MAX_BLOCK) ||
 	    (donor_start > EXT_MAX_BLOCK) ||
 	    (*len > EXT_MAX_BLOCK) ||
@@ -1088,7 +1060,7 @@ mext_check_arguments(struct inode *orig_inode,
 	}
 
 	if (!*len) {
-		ext4_debug("ext4 move extent: len shoudld not be 0 "
+		ext4_debug("ext4 move extent: len should not be 0 "
 			"[ino:orig %lu, donor %lu]\n", orig_inode->i_ino,
 			donor_inode->i_ino);
 		return -EINVAL;
@@ -1232,16 +1204,16 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp,
 		return -EINVAL;
 	}
 
-	/* protect orig and donor against a truncate */
+	/* Protect orig and donor inodes against a truncate */
 	ret1 = mext_inode_double_lock(orig_inode, donor_inode);
 	if (ret1 < 0)
 		return ret1;
 
-	mext_double_down_read(orig_inode, donor_inode);
+	/* Protect extent tree against block allocations via delalloc */
+	double_down_write_data_sem(orig_inode, donor_inode);
 	/* Check the filesystem environment whether move_extent can be done */
 	ret1 = mext_check_arguments(orig_inode, donor_inode, orig_start,
-					donor_start, &len, *moved_len);
-	mext_double_up_read(orig_inode, donor_inode);
+				    donor_start, &len);
 	if (ret1)
 		goto out;
 
@@ -1355,36 +1327,39 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp,
 		seq_start = le32_to_cpu(ext_cur->ee_block);
 		rest_blocks = seq_blocks;
 
-		/* Discard preallocations of two inodes */
-		down_write(&EXT4_I(orig_inode)->i_data_sem);
-		ext4_discard_preallocations(orig_inode);
-		up_write(&EXT4_I(orig_inode)->i_data_sem);
-
-		down_write(&EXT4_I(donor_inode)->i_data_sem);
-		ext4_discard_preallocations(donor_inode);
-		up_write(&EXT4_I(donor_inode)->i_data_sem);
+		/*
+		 * Up semaphore to avoid following problems:
+		 * a. transaction deadlock among ext4_journal_start,
+		 *    ->write_begin via pagefault, and jbd2_journal_commit
+		 * b. racing with ->readpage, ->write_begin, and ext4_get_block
+		 *    in move_extent_per_page
+		 */
+		double_up_write_data_sem(orig_inode, donor_inode);
 
 		while (orig_page_offset <= seq_end_page) {
 
 			/* Swap original branches with new branches */
-			ret1 = move_extent_per_page(o_filp, donor_inode,
+			block_len_in_page = move_extent_per_page(
+						o_filp, donor_inode,
 						orig_page_offset,
 						data_offset_in_page,
-						block_len_in_page, uninit);
-			if (ret1 < 0)
-				goto out;
-			orig_page_offset++;
+						block_len_in_page, uninit,
+						&ret1);
+
 			/* Count how many blocks we have exchanged */
 			*moved_len += block_len_in_page;
+			if (ret1 < 0)
+				break;
 			if (*moved_len > len) {
 				ext4_error(orig_inode->i_sb, __func__,
 					"We replaced blocks too much! "
 					"sum of replaced: %llu requested: %llu",
 					*moved_len, len);
 				ret1 = -EIO;
-				goto out;
+				break;
 			}
 
+			orig_page_offset++;
 			data_offset_in_page = 0;
 			rest_blocks -= block_len_in_page;
 			if (rest_blocks > blocks_per_page)
@@ -1393,6 +1368,10 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp,
 				block_len_in_page = rest_blocks;
 		}
 
+		double_down_write_data_sem(orig_inode, donor_inode);
+		if (ret1 < 0)
+			break;
+
 		/* Decrease buffer counter */
 		if (holecheck_path)
 			ext4_ext_drop_refs(holecheck_path);
@@ -1414,6 +1393,11 @@ ext4_move_extents(struct file *o_filp, struct file *d_filp,
 
 	}
 out:
+	if (*moved_len) {
+		ext4_discard_preallocations(orig_inode);
+		ext4_discard_preallocations(donor_inode);
+	}
+
 	if (orig_path) {
 		ext4_ext_drop_refs(orig_path);
 		kfree(orig_path);
@@ -1422,7 +1406,7 @@ out:
 		ext4_ext_drop_refs(holecheck_path);
 		kfree(holecheck_path);
 	}
-
+	double_up_write_data_sem(orig_inode, donor_inode);
 	ret2 = mext_inode_double_unlock(orig_inode, donor_inode);
 
 	if (ret1)
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 6d2c1b897fc7..17a17e10dd60 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -1292,9 +1292,6 @@ errout:
  * add_dirent_to_buf will attempt search the directory block for
  * space.  It will return -ENOSPC if no space is available, and -EIO
  * and -EEXIST if directory entry already exists.
- *
- * NOTE!  bh is NOT released in the case where ENOSPC is returned.  In
- * all other cases bh is released.
  */
 static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 			     struct inode *inode, struct ext4_dir_entry_2 *de,
@@ -1315,14 +1312,10 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 		top = bh->b_data + blocksize - reclen;
 		while ((char *) de <= top) {
 			if (!ext4_check_dir_entry("ext4_add_entry", dir, de,
-						  bh, offset)) {
-				brelse(bh);
+						  bh, offset))
 				return -EIO;
-			}
-			if (ext4_match(namelen, name, de)) {
-				brelse(bh);
+			if (ext4_match(namelen, name, de))
 				return -EEXIST;
-			}
 			nlen = EXT4_DIR_REC_LEN(de->name_len);
 			rlen = ext4_rec_len_from_disk(de->rec_len, blocksize);
 			if ((de->inode? rlen - nlen: rlen) >= reclen)
@@ -1337,7 +1330,6 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 	err = ext4_journal_get_write_access(handle, bh);
 	if (err) {
 		ext4_std_error(dir->i_sb, err);
-		brelse(bh);
 		return err;
 	}
 
@@ -1377,7 +1369,6 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 	err = ext4_handle_dirty_metadata(handle, dir, bh);
 	if (err)
 		ext4_std_error(dir->i_sb, err);
-	brelse(bh);
 	return 0;
 }
 
@@ -1471,7 +1462,9 @@ static int make_indexed_dir(handle_t *handle, struct dentry *dentry,
 	if (!(de))
 		return retval;
 
-	return add_dirent_to_buf(handle, dentry, inode, de, bh);
+	retval = add_dirent_to_buf(handle, dentry, inode, de, bh);
+	brelse(bh);
+	return retval;
 }
 
 /*
@@ -1514,8 +1507,10 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
 		if(!bh)
 			return retval;
 		retval = add_dirent_to_buf(handle, dentry, inode, NULL, bh);
-		if (retval != -ENOSPC)
+		if (retval != -ENOSPC) {
+			brelse(bh);
 			return retval;
+		}
 
 		if (blocks == 1 && !dx_fallback &&
 		    EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_DIR_INDEX))
@@ -1528,7 +1523,9 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
 	de = (struct ext4_dir_entry_2 *) bh->b_data;
 	de->inode = 0;
 	de->rec_len = ext4_rec_len_to_disk(blocksize, blocksize);
-	return add_dirent_to_buf(handle, dentry, inode, de, bh);
+	retval = add_dirent_to_buf(handle, dentry, inode, de, bh);
+	brelse(bh);
+	return retval;
 }
 
 /*
@@ -1561,10 +1558,8 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 		goto journal_error;
 
 	err = add_dirent_to_buf(handle, dentry, inode, NULL, bh);
-	if (err != -ENOSPC) {
-		bh = NULL;
+	if (err != -ENOSPC)
 		goto cleanup;
-	}
 
 	/* Block full, should compress but for now just split */
 	dxtrace(printk(KERN_DEBUG "using %u of %u node entries\n",
@@ -1657,7 +1652,6 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 	if (!de)
 		goto cleanup;
 	err = add_dirent_to_buf(handle, dentry, inode, de, bh);
-	bh = NULL;
 	goto cleanup;
 
 journal_error:
@@ -1775,7 +1769,7 @@ static int ext4_create(struct inode *dir, struct dentry *dentry, int mode,
 retry:
 	handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb));
+					EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -1809,7 +1803,7 @@ static int ext4_mknod(struct inode *dir, struct dentry *dentry,
 retry:
 	handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb));
+					EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -1846,7 +1840,7 @@ static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 retry:
 	handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 +
-					2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb));
+					EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
@@ -2259,7 +2253,7 @@ static int ext4_symlink(struct inode *dir,
 retry:
 	handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) +
 					EXT4_INDEX_EXTRA_TRANS_BLOCKS + 5 +
-					2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb));
+					EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index 3cfc343c41b5..3b2c5541d8a6 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -247,7 +247,7 @@ static int setup_new_group_blocks(struct super_block *sb,
 			goto exit_bh;
 
 		if (IS_ERR(gdb = bclean(handle, sb, block))) {
-			err = PTR_ERR(bh);
+			err = PTR_ERR(gdb);
 			goto exit_bh;
 		}
 		ext4_handle_dirty_metadata(handle, NULL, gdb);
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index d4ca92aab514..735c20d5fd56 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -603,10 +603,6 @@ static void ext4_put_super(struct super_block *sb)
 	if (sb->s_dirt)
 		ext4_commit_super(sb, 1);
 
-	ext4_release_system_zone(sb);
-	ext4_mb_release(sb);
-	ext4_ext_release(sb);
-	ext4_xattr_put_super(sb);
 	if (sbi->s_journal) {
 		err = jbd2_journal_destroy(sbi->s_journal);
 		sbi->s_journal = NULL;
@@ -614,6 +610,12 @@ static void ext4_put_super(struct super_block *sb)
 			ext4_abort(sb, __func__,
 				   "Couldn't clean up the journal");
 	}
+
+	ext4_release_system_zone(sb);
+	ext4_mb_release(sb);
+	ext4_ext_release(sb);
+	ext4_xattr_put_super(sb);
+
 	if (!(sb->s_flags & MS_RDONLY)) {
 		EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
 		es->s_state = cpu_to_le16(sbi->s_mount_state);
@@ -700,10 +702,16 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	ei->i_reserved_data_blocks = 0;
 	ei->i_reserved_meta_blocks = 0;
 	ei->i_allocated_meta_blocks = 0;
+	ei->i_da_metadata_calc_len = 0;
 	ei->i_delalloc_reserved_flag = 0;
 	spin_lock_init(&(ei->i_block_reservation_lock));
+#ifdef CONFIG_QUOTA
+	ei->i_reserved_quota = 0;
+#endif
 	INIT_LIST_HEAD(&ei->i_aio_dio_complete_list);
 	ei->cur_aio_dio = NULL;
+	ei->i_sync_tid = 0;
+	ei->i_datasync_tid = 0;
 
 	return &ei->vfs_inode;
 }
@@ -765,9 +773,22 @@ static inline void ext4_show_quota_options(struct seq_file *seq,
 #if defined(CONFIG_QUOTA)
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 
-	if (sbi->s_jquota_fmt)
-		seq_printf(seq, ",jqfmt=%s",
-		(sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold" : "vfsv0");
+	if (sbi->s_jquota_fmt) {
+		char *fmtname = "";
+
+		switch (sbi->s_jquota_fmt) {
+		case QFMT_VFS_OLD:
+			fmtname = "vfsold";
+			break;
+		case QFMT_VFS_V0:
+			fmtname = "vfsv0";
+			break;
+		case QFMT_VFS_V1:
+			fmtname = "vfsv1";
+			break;
+		}
+		seq_printf(seq, ",jqfmt=%s", fmtname);
+	}
 
 	if (sbi->s_qf_names[USRQUOTA])
 		seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]);
@@ -899,6 +920,12 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs)
 	if (test_opt(sb, NO_AUTO_DA_ALLOC))
 		seq_puts(seq, ",noauto_da_alloc");
 
+	if (test_opt(sb, DISCARD))
+		seq_puts(seq, ",discard");
+
+	if (test_opt(sb, NOLOAD))
+		seq_puts(seq, ",norecovery");
+
 	ext4_show_quota_options(seq, sb);
 
 	return 0;
@@ -991,7 +1018,9 @@ static const struct dquot_operations ext4_quota_operations = {
 	.reserve_space	= dquot_reserve_space,
 	.claim_space	= dquot_claim_space,
 	.release_rsv	= dquot_release_reserved_space,
+#ifdef CONFIG_QUOTA
 	.get_reserved_space = ext4_get_reserved_space,
+#endif
 	.alloc_inode	= dquot_alloc_inode,
 	.free_space	= dquot_free_space,
 	.free_inode	= dquot_free_inode,
@@ -1074,12 +1103,13 @@ enum {
 	Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback,
 	Opt_data_err_abort, Opt_data_err_ignore,
 	Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota,
-	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota,
-	Opt_ignore, Opt_barrier, Opt_nobarrier, Opt_err, Opt_resize,
-	Opt_usrquota, Opt_grpquota, Opt_i_version,
+	Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota,
+	Opt_noquota, Opt_ignore, Opt_barrier, Opt_nobarrier, Opt_err,
+	Opt_resize, Opt_usrquota, Opt_grpquota, Opt_i_version,
 	Opt_stripe, Opt_delalloc, Opt_nodelalloc,
 	Opt_block_validity, Opt_noblock_validity,
-	Opt_inode_readahead_blks, Opt_journal_ioprio
+	Opt_inode_readahead_blks, Opt_journal_ioprio,
+	Opt_discard, Opt_nodiscard,
 };
 
 static const match_table_t tokens = {
@@ -1104,6 +1134,7 @@ static const match_table_t tokens = {
 	{Opt_acl, "acl"},
 	{Opt_noacl, "noacl"},
 	{Opt_noload, "noload"},
+	{Opt_noload, "norecovery"},
 	{Opt_nobh, "nobh"},
 	{Opt_bh, "bh"},
 	{Opt_commit, "commit=%u"},
@@ -1125,6 +1156,7 @@ static const match_table_t tokens = {
 	{Opt_grpjquota, "grpjquota=%s"},
 	{Opt_jqfmt_vfsold, "jqfmt=vfsold"},
 	{Opt_jqfmt_vfsv0, "jqfmt=vfsv0"},
+	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
 	{Opt_grpquota, "grpquota"},
 	{Opt_noquota, "noquota"},
 	{Opt_quota, "quota"},
@@ -1144,6 +1176,8 @@ static const match_table_t tokens = {
 	{Opt_auto_da_alloc, "auto_da_alloc=%u"},
 	{Opt_auto_da_alloc, "auto_da_alloc"},
 	{Opt_noauto_da_alloc, "noauto_da_alloc"},
+	{Opt_discard, "discard"},
+	{Opt_nodiscard, "nodiscard"},
 	{Opt_err, NULL},
 };
 
@@ -1425,6 +1459,9 @@ clear_qf_name:
 			goto set_qf_format;
 		case Opt_jqfmt_vfsv0:
 			qfmt = QFMT_VFS_V0;
+			goto set_qf_format;
+		case Opt_jqfmt_vfsv1:
+			qfmt = QFMT_VFS_V1;
 set_qf_format:
 			if (sb_any_quota_loaded(sb) &&
 			    sbi->s_jquota_fmt != qfmt) {
@@ -1467,6 +1504,7 @@ set_qf_format:
 		case Opt_offgrpjquota:
 		case Opt_jqfmt_vfsold:
 		case Opt_jqfmt_vfsv0:
+		case Opt_jqfmt_vfsv1:
 			ext4_msg(sb, KERN_ERR,
 				"journaled quota options not supported");
 			break;
@@ -1565,6 +1603,12 @@ set_qf_format:
 			else
 				set_opt(sbi->s_mount_opt,NO_AUTO_DA_ALLOC);
 			break;
+		case Opt_discard:
+			set_opt(sbi->s_mount_opt, DISCARD);
+			break;
+		case Opt_nodiscard:
+			clear_opt(sbi->s_mount_opt, DISCARD);
+			break;
 		default:
 			ext4_msg(sb, KERN_ERR,
 			       "Unrecognized mount option \"%s\" "
@@ -1673,14 +1717,14 @@ static int ext4_fill_flex_info(struct super_block *sb)
 	size_t size;
 	int i;
 
-	if (!sbi->s_es->s_log_groups_per_flex) {
+	sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
+	groups_per_flex = 1 << sbi->s_log_groups_per_flex;
+
+	if (groups_per_flex < 2) {
 		sbi->s_log_groups_per_flex = 0;
 		return 1;
 	}
 
-	sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex;
-	groups_per_flex = 1 << sbi->s_log_groups_per_flex;
-
 	/* We allocate both existing and potentially added groups */
 	flex_group_count = ((sbi->s_groups_count + groups_per_flex - 1) +
 			((le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks) + 1) <<
@@ -2099,11 +2143,8 @@ static int parse_strtoul(const char *buf,
 {
 	char *endp;
 
-	while (*buf && isspace(*buf))
-		buf++;
-	*value = simple_strtoul(buf, &endp, 0);
-	while (*endp && isspace(*endp))
-		endp++;
+	*value = simple_strtoul(skip_spaces(buf), &endp, 0);
+	endp = skip_spaces(endp);
 	if (*endp || *value > max)
 		return -EINVAL;
 
@@ -2134,9 +2175,9 @@ static ssize_t lifetime_write_kbytes_show(struct ext4_attr *a,
 	struct super_block *sb = sbi->s_buddy_cache->i_sb;
 
 	return snprintf(buf, PAGE_SIZE, "%llu\n",
-			sbi->s_kbytes_written + 
+			(unsigned long long)(sbi->s_kbytes_written +
 			((part_stat_read(sb->s_bdev->bd_part, sectors[1]) -
-			  EXT4_SB(sb)->s_sectors_written_start) >> 1));
+			  EXT4_SB(sb)->s_sectors_written_start) >> 1)));
 }
 
 static ssize_t inode_readahead_blks_store(struct ext4_attr *a,
@@ -2721,26 +2762,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 	    EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) {
 		if (ext4_load_journal(sb, es, journal_devnum))
 			goto failed_mount3;
-		if (!(sb->s_flags & MS_RDONLY) &&
-		    EXT4_SB(sb)->s_journal->j_failed_commit) {
-			ext4_msg(sb, KERN_CRIT, "error: "
-			       "ext4_fill_super: Journal transaction "
-			       "%u is corrupt",
-			       EXT4_SB(sb)->s_journal->j_failed_commit);
-			if (test_opt(sb, ERRORS_RO)) {
-				ext4_msg(sb, KERN_CRIT,
-				       "Mounting filesystem read-only");
-				sb->s_flags |= MS_RDONLY;
-				EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
-				es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
-			}
-			if (test_opt(sb, ERRORS_PANIC)) {
-				EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
-				es->s_state |= cpu_to_le16(EXT4_ERROR_FS);
-				ext4_commit_super(sb, 1);
-				goto failed_mount4;
-			}
-		}
 	} else if (test_opt(sb, NOLOAD) && !(sb->s_flags & MS_RDONLY) &&
 	      EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) {
 		ext4_msg(sb, KERN_ERR, "required journal recovery "
@@ -3668,13 +3689,11 @@ static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_blocks = ext4_blocks_count(es) - sbi->s_overhead_last;
 	buf->f_bfree = percpu_counter_sum_positive(&sbi->s_freeblocks_counter) -
 		       percpu_counter_sum_positive(&sbi->s_dirtyblocks_counter);
-	ext4_free_blocks_count_set(es, buf->f_bfree);
 	buf->f_bavail = buf->f_bfree - ext4_r_blocks_count(es);
 	if (buf->f_bfree < ext4_r_blocks_count(es))
 		buf->f_bavail = 0;
 	buf->f_files = le32_to_cpu(es->s_inodes_count);
 	buf->f_ffree = percpu_counter_sum_positive(&sbi->s_freeinodes_counter);
-	es->s_free_inodes_count = cpu_to_le32(buf->f_ffree);
 	buf->f_namelen = EXT4_NAME_LEN;
 	fsid = le64_to_cpup((void *)es->s_uuid) ^
 	       le64_to_cpup((void *)es->s_uuid + sizeof(u64));
@@ -3966,6 +3985,60 @@ static int ext4_get_sb(struct file_system_type *fs_type, int flags,
 	return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super,mnt);
 }
 
+#if !defined(CONTIG_EXT2_FS) && !defined(CONFIG_EXT2_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23)
+static struct file_system_type ext2_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "ext2",
+	.get_sb		= ext4_get_sb,
+	.kill_sb	= kill_block_super,
+	.fs_flags	= FS_REQUIRES_DEV,
+};
+
+static inline void register_as_ext2(void)
+{
+	int err = register_filesystem(&ext2_fs_type);
+	if (err)
+		printk(KERN_WARNING
+		       "EXT4-fs: Unable to register as ext2 (%d)\n", err);
+}
+
+static inline void unregister_as_ext2(void)
+{
+	unregister_filesystem(&ext2_fs_type);
+}
+MODULE_ALIAS("ext2");
+#else
+static inline void register_as_ext2(void) { }
+static inline void unregister_as_ext2(void) { }
+#endif
+
+#if !defined(CONTIG_EXT3_FS) && !defined(CONFIG_EXT3_FS_MODULE) && defined(CONFIG_EXT4_USE_FOR_EXT23)
+static struct file_system_type ext3_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= "ext3",
+	.get_sb		= ext4_get_sb,
+	.kill_sb	= kill_block_super,
+	.fs_flags	= FS_REQUIRES_DEV,
+};
+
+static inline void register_as_ext3(void)
+{
+	int err = register_filesystem(&ext3_fs_type);
+	if (err)
+		printk(KERN_WARNING
+		       "EXT4-fs: Unable to register as ext3 (%d)\n", err);
+}
+
+static inline void unregister_as_ext3(void)
+{
+	unregister_filesystem(&ext3_fs_type);
+}
+MODULE_ALIAS("ext3");
+#else
+static inline void register_as_ext3(void) { }
+static inline void unregister_as_ext3(void) { }
+#endif
+
 static struct file_system_type ext4_fs_type = {
 	.owner		= THIS_MODULE,
 	.name		= "ext4",
@@ -3995,11 +4068,15 @@ static int __init init_ext4_fs(void)
 	err = init_inodecache();
 	if (err)
 		goto out1;
+	register_as_ext2();
+	register_as_ext3();
 	err = register_filesystem(&ext4_fs_type);
 	if (err)
 		goto out;
 	return 0;
 out:
+	unregister_as_ext2();
+	unregister_as_ext3();
 	destroy_inodecache();
 out1:
 	exit_ext4_xattr();
@@ -4015,6 +4092,8 @@ out4:
 
 static void __exit exit_ext4_fs(void)
 {
+	unregister_as_ext2();
+	unregister_as_ext3();
 	unregister_filesystem(&ext4_fs_type);
 	destroy_inodecache();
 	exit_ext4_xattr();
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index fed5b01d7a8d..f3a2f7ed45aa 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -92,7 +92,7 @@ static struct buffer_head *ext4_xattr_cache_find(struct inode *,
 						 struct mb_cache_entry **);
 static void ext4_xattr_rehash(struct ext4_xattr_header *,
 			      struct ext4_xattr_entry *);
-static int ext4_xattr_list(struct inode *inode, char *buffer,
+static int ext4_xattr_list(struct dentry *dentry, char *buffer,
 			   size_t buffer_size);
 
 static struct mb_cache *ext4_xattr_cache;
@@ -140,7 +140,7 @@ ext4_xattr_handler(int name_index)
 ssize_t
 ext4_listxattr(struct dentry *dentry, char *buffer, size_t size)
 {
-	return ext4_xattr_list(dentry->d_inode, buffer, size);
+	return ext4_xattr_list(dentry, buffer, size);
 }
 
 static int
@@ -325,7 +325,7 @@ ext4_xattr_get(struct inode *inode, int name_index, const char *name,
 }
 
 static int
-ext4_xattr_list_entries(struct inode *inode, struct ext4_xattr_entry *entry,
+ext4_xattr_list_entries(struct dentry *dentry, struct ext4_xattr_entry *entry,
 			char *buffer, size_t buffer_size)
 {
 	size_t rest = buffer_size;
@@ -335,9 +335,10 @@ ext4_xattr_list_entries(struct inode *inode, struct ext4_xattr_entry *entry,
 			ext4_xattr_handler(entry->e_name_index);
 
 		if (handler) {
-			size_t size = handler->list(inode, buffer, rest,
+			size_t size = handler->list(dentry, buffer, rest,
 						    entry->e_name,
-						    entry->e_name_len);
+						    entry->e_name_len,
+						    handler->flags);
 			if (buffer) {
 				if (size > rest)
 					return -ERANGE;
@@ -350,8 +351,9 @@ ext4_xattr_list_entries(struct inode *inode, struct ext4_xattr_entry *entry,
 }
 
 static int
-ext4_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size)
+ext4_xattr_block_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
+	struct inode *inode = dentry->d_inode;
 	struct buffer_head *bh = NULL;
 	int error;
 
@@ -376,7 +378,7 @@ ext4_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size)
 		goto cleanup;
 	}
 	ext4_xattr_cache_insert(bh);
-	error = ext4_xattr_list_entries(inode, BFIRST(bh), buffer, buffer_size);
+	error = ext4_xattr_list_entries(dentry, BFIRST(bh), buffer, buffer_size);
 
 cleanup:
 	brelse(bh);
@@ -385,8 +387,9 @@ cleanup:
 }
 
 static int
-ext4_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size)
+ext4_xattr_ibody_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
+	struct inode *inode = dentry->d_inode;
 	struct ext4_xattr_ibody_header *header;
 	struct ext4_inode *raw_inode;
 	struct ext4_iloc iloc;
@@ -404,7 +407,7 @@ ext4_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size)
 	error = ext4_xattr_check_names(IFIRST(header), end);
 	if (error)
 		goto cleanup;
-	error = ext4_xattr_list_entries(inode, IFIRST(header),
+	error = ext4_xattr_list_entries(dentry, IFIRST(header),
 					buffer, buffer_size);
 
 cleanup:
@@ -423,12 +426,12 @@ cleanup:
  * used / required on success.
  */
 static int
-ext4_xattr_list(struct inode *inode, char *buffer, size_t buffer_size)
+ext4_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
 	int i_error, b_error;
 
-	down_read(&EXT4_I(inode)->xattr_sem);
-	i_error = ext4_xattr_ibody_list(inode, buffer, buffer_size);
+	down_read(&EXT4_I(dentry->d_inode)->xattr_sem);
+	i_error = ext4_xattr_ibody_list(dentry, buffer, buffer_size);
 	if (i_error < 0) {
 		b_error = 0;
 	} else {
@@ -436,11 +439,11 @@ ext4_xattr_list(struct inode *inode, char *buffer, size_t buffer_size)
 			buffer += i_error;
 			buffer_size -= i_error;
 		}
-		b_error = ext4_xattr_block_list(inode, buffer, buffer_size);
+		b_error = ext4_xattr_block_list(dentry, buffer, buffer_size);
 		if (b_error < 0)
 			i_error = 0;
 	}
-	up_read(&EXT4_I(inode)->xattr_sem);
+	up_read(&EXT4_I(dentry->d_inode)->xattr_sem);
 	return i_error + b_error;
 }
 
@@ -482,9 +485,10 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode,
 		ea_bdebug(bh, "refcount now=0; freeing");
 		if (ce)
 			mb_cache_entry_free(ce);
-		ext4_free_blocks(handle, inode, bh->b_blocknr, 1, 1);
 		get_bh(bh);
-		ext4_forget(handle, 1, inode, bh, bh->b_blocknr);
+		ext4_free_blocks(handle, inode, bh, 0, 1,
+				 EXT4_FREE_BLOCKS_METADATA |
+				 EXT4_FREE_BLOCKS_FORGET);
 	} else {
 		le32_add_cpu(&BHDR(bh)->h_refcount, -1);
 		error = ext4_handle_dirty_metadata(handle, inode, bh);
@@ -832,7 +836,8 @@ inserted:
 			new_bh = sb_getblk(sb, block);
 			if (!new_bh) {
 getblk_failed:
-				ext4_free_blocks(handle, inode, block, 1, 1);
+				ext4_free_blocks(handle, inode, 0, block, 1,
+						 EXT4_FREE_BLOCKS_METADATA);
 				error = -EIO;
 				goto cleanup;
 			}
@@ -988,6 +993,10 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
 	if (error)
 		goto cleanup;
 
+	error = ext4_journal_get_write_access(handle, is.iloc.bh);
+	if (error)
+		goto cleanup;
+
 	if (EXT4_I(inode)->i_state & EXT4_STATE_NEW) {
 		struct ext4_inode *raw_inode = ext4_raw_inode(&is.iloc);
 		memset(raw_inode, 0, EXT4_SB(inode->i_sb)->s_inode_size);
@@ -1013,9 +1022,6 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
 		if (flags & XATTR_CREATE)
 			goto cleanup;
 	}
-	error = ext4_journal_get_write_access(handle, is.iloc.bh);
-	if (error)
-		goto cleanup;
 	if (!value) {
 		if (!is.s.not_found)
 			error = ext4_xattr_ibody_set(handle, inode, &i, &is);
@@ -1326,6 +1332,8 @@ retry:
 			goto cleanup;
 		kfree(b_entry_name);
 		kfree(buffer);
+		b_entry_name = NULL;
+		buffer = NULL;
 		brelse(is->iloc.bh);
 		kfree(is);
 		kfree(bs);
diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c
index ca5f89fc6cae..983c253999a7 100644
--- a/fs/ext4/xattr_security.c
+++ b/fs/ext4/xattr_security.c
@@ -12,8 +12,8 @@
 #include "xattr.h"
 
 static size_t
-ext4_xattr_security_list(struct inode *inode, char *list, size_t list_size,
-			 const char *name, size_t name_len)
+ext4_xattr_security_list(struct dentry *dentry, char *list, size_t list_size,
+		const char *name, size_t name_len, int type)
 {
 	const size_t prefix_len = sizeof(XATTR_SECURITY_PREFIX)-1;
 	const size_t total_len = prefix_len + name_len + 1;
@@ -28,23 +28,23 @@ ext4_xattr_security_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext4_xattr_security_get(struct inode *inode, const char *name,
-		       void *buffer, size_t size)
+ext4_xattr_security_get(struct dentry *dentry, const char *name,
+		       void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext4_xattr_get(inode, EXT4_XATTR_INDEX_SECURITY, name,
-			      buffer, size);
+	return ext4_xattr_get(dentry->d_inode, EXT4_XATTR_INDEX_SECURITY,
+			      name, buffer, size);
 }
 
 static int
-ext4_xattr_security_set(struct inode *inode, const char *name,
-		       const void *value, size_t size, int flags)
+ext4_xattr_security_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext4_xattr_set(inode, EXT4_XATTR_INDEX_SECURITY, name,
-			      value, size, flags);
+	return ext4_xattr_set(dentry->d_inode, EXT4_XATTR_INDEX_SECURITY,
+			      name, value, size, flags);
 }
 
 int
diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c
index ac1a52cf2a37..15b50edc6587 100644
--- a/fs/ext4/xattr_trusted.c
+++ b/fs/ext4/xattr_trusted.c
@@ -14,8 +14,8 @@
 #include "xattr.h"
 
 static size_t
-ext4_xattr_trusted_list(struct inode *inode, char *list, size_t list_size,
-			const char *name, size_t name_len)
+ext4_xattr_trusted_list(struct dentry *dentry, char *list, size_t list_size,
+		const char *name, size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_TRUSTED_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
@@ -32,23 +32,23 @@ ext4_xattr_trusted_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext4_xattr_trusted_get(struct inode *inode, const char *name,
-		       void *buffer, size_t size)
+ext4_xattr_trusted_get(struct dentry *dentry, const char *name, void *buffer,
+		size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext4_xattr_get(inode, EXT4_XATTR_INDEX_TRUSTED, name,
-			      buffer, size);
+	return ext4_xattr_get(dentry->d_inode, EXT4_XATTR_INDEX_TRUSTED,
+			      name, buffer, size);
 }
 
 static int
-ext4_xattr_trusted_set(struct inode *inode, const char *name,
-		       const void *value, size_t size, int flags)
+ext4_xattr_trusted_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ext4_xattr_set(inode, EXT4_XATTR_INDEX_TRUSTED, name,
-			      value, size, flags);
+	return ext4_xattr_set(dentry->d_inode, EXT4_XATTR_INDEX_TRUSTED,
+			      name, value, size, flags);
 }
 
 struct xattr_handler ext4_xattr_trusted_handler = {
diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c
index d91aa61b42aa..c4ce05746ce1 100644
--- a/fs/ext4/xattr_user.c
+++ b/fs/ext4/xattr_user.c
@@ -13,13 +13,13 @@
 #include "xattr.h"
 
 static size_t
-ext4_xattr_user_list(struct inode *inode, char *list, size_t list_size,
-		     const char *name, size_t name_len)
+ext4_xattr_user_list(struct dentry *dentry, char *list, size_t list_size,
+		     const char *name, size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_USER_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return 0;
 
 	if (list && total_len <= list_size) {
@@ -31,26 +31,27 @@ ext4_xattr_user_list(struct inode *inode, char *list, size_t list_size,
 }
 
 static int
-ext4_xattr_user_get(struct inode *inode, const char *name,
-		    void *buffer, size_t size)
+ext4_xattr_user_get(struct dentry *dentry, const char *name,
+		    void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return -EOPNOTSUPP;
-	return ext4_xattr_get(inode, EXT4_XATTR_INDEX_USER, name, buffer, size);
+	return ext4_xattr_get(dentry->d_inode, EXT4_XATTR_INDEX_USER,
+			      name, buffer, size);
 }
 
 static int
-ext4_xattr_user_set(struct inode *inode, const char *name,
-		    const void *value, size_t size, int flags)
+ext4_xattr_user_set(struct dentry *dentry, const char *name,
+		    const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	if (!test_opt(inode->i_sb, XATTR_USER))
+	if (!test_opt(dentry->d_sb, XATTR_USER))
 		return -EOPNOTSUPP;
-	return ext4_xattr_set(inode, EXT4_XATTR_INDEX_USER, name,
-			      value, size, flags);
+	return ext4_xattr_set(dentry->d_inode, EXT4_XATTR_INDEX_USER,
+			      name, value, size, flags);
 }
 
 struct xattr_handler ext4_xattr_user_handler = {
diff --git a/fs/fat/fat.h b/fs/fat/fat.h
index 7db0979c6b72..e6efdfa0f6db 100644
--- a/fs/fat/fat.h
+++ b/fs/fat/fat.h
@@ -44,7 +44,8 @@ struct fat_mount_options {
 		 nocase:1,	  /* Does this need case conversion? 0=need case conversion*/
 		 usefree:1,	  /* Use free_clusters for FAT32 */
 		 tz_utc:1,	  /* Filesystem timestamps are in UTC */
-		 rodir:1;	  /* allow ATTR_RO for directory */
+		 rodir:1,	  /* allow ATTR_RO for directory */
+		 discard:1;	  /* Issue discard requests on deletions */
 };
 
 #define FAT_HASH_BITS	8
diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c
index a81037721a6f..81184d3b75a3 100644
--- a/fs/fat/fatent.c
+++ b/fs/fat/fatent.c
@@ -566,16 +566,21 @@ int fat_free_clusters(struct inode *inode, int cluster)
 			goto error;
 		}
 
-		/* 
-		 * Issue discard for the sectors we no longer care about,
-		 * batching contiguous clusters into one request
-		 */
-		if (cluster != fatent.entry + 1) {
-			int nr_clus = fatent.entry - first_cl + 1;
-
-			sb_issue_discard(sb, fat_clus_to_blknr(sbi, first_cl),
-					 nr_clus * sbi->sec_per_clus);
-			first_cl = cluster;
+		if (sbi->options.discard) {
+			/*
+			 * Issue discard for the sectors we no longer
+			 * care about, batching contiguous clusters
+			 * into one request
+			 */
+			if (cluster != fatent.entry + 1) {
+				int nr_clus = fatent.entry - first_cl + 1;
+
+				sb_issue_discard(sb,
+					fat_clus_to_blknr(sbi, first_cl),
+					nr_clus * sbi->sec_per_clus);
+
+				first_cl = cluster;
+			}
 		}
 
 		ops->ent_put(&fatent, FAT_ENT_FREE);
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 76b7961ab663..14da530b05ca 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -858,6 +858,8 @@ static int fat_show_options(struct seq_file *m, struct vfsmount *mnt)
 		seq_puts(m, ",errors=panic");
 	else
 		seq_puts(m, ",errors=remount-ro");
+	if (opts->discard)
+		seq_puts(m, ",discard");
 
 	return 0;
 }
@@ -871,7 +873,7 @@ enum {
 	Opt_shortname_winnt, Opt_shortname_mixed, Opt_utf8_no, Opt_utf8_yes,
 	Opt_uni_xl_no, Opt_uni_xl_yes, Opt_nonumtail_no, Opt_nonumtail_yes,
 	Opt_obsolate, Opt_flush, Opt_tz_utc, Opt_rodir, Opt_err_cont,
-	Opt_err_panic, Opt_err_ro, Opt_err,
+	Opt_err_panic, Opt_err_ro, Opt_discard, Opt_err,
 };
 
 static const match_table_t fat_tokens = {
@@ -899,6 +901,7 @@ static const match_table_t fat_tokens = {
 	{Opt_err_cont, "errors=continue"},
 	{Opt_err_panic, "errors=panic"},
 	{Opt_err_ro, "errors=remount-ro"},
+	{Opt_discard, "discard"},
 	{Opt_obsolate, "conv=binary"},
 	{Opt_obsolate, "conv=text"},
 	{Opt_obsolate, "conv=auto"},
@@ -1136,6 +1139,9 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug,
 		case Opt_rodir:
 			opts->rodir = 1;
 			break;
+		case Opt_discard:
+			opts->discard = 1;
+			break;
 
 		/* obsolete mount options */
 		case Opt_obsolate:
diff --git a/fs/fat/misc.c b/fs/fat/misc.c
index 0f55f5cb732f..d3da05f26465 100644
--- a/fs/fat/misc.c
+++ b/fs/fat/misc.c
@@ -9,6 +9,7 @@
 #include <linux/module.h>
 #include <linux/fs.h>
 #include <linux/buffer_head.h>
+#include <linux/time.h>
 #include "fat.h"
 
 /*
@@ -157,10 +158,6 @@ extern struct timezone sys_tz;
 #define SECS_PER_MIN	60
 #define SECS_PER_HOUR	(60 * 60)
 #define SECS_PER_DAY	(SECS_PER_HOUR * 24)
-#define UNIX_SECS_1980	315532800L
-#if BITS_PER_LONG == 64
-#define UNIX_SECS_2108	4354819200L
-#endif
 /* days between 1.1.70 and 1.1.80 (2 leap days) */
 #define DAYS_DELTA	(365 * 10 + 2)
 /* 120 (2100 - 1980) isn't leap year */
@@ -213,58 +210,35 @@ void fat_time_fat2unix(struct msdos_sb_info *sbi, struct timespec *ts,
 void fat_time_unix2fat(struct msdos_sb_info *sbi, struct timespec *ts,
 		       __le16 *time, __le16 *date, u8 *time_cs)
 {
-	time_t second = ts->tv_sec;
-	time_t day, leap_day, month, year;
+	struct tm tm;
+	time_to_tm(ts->tv_sec, sbi->options.tz_utc ? 0 :
+		   -sys_tz.tz_minuteswest * 60, &tm);
 
-	if (!sbi->options.tz_utc)
-		second -= sys_tz.tz_minuteswest * SECS_PER_MIN;
-
-	/* Jan 1 GMT 00:00:00 1980. But what about another time zone? */
-	if (second < UNIX_SECS_1980) {
+	/*  FAT can only support year between 1980 to 2107 */
+	if (tm.tm_year < 1980 - 1900) {
 		*time = 0;
 		*date = cpu_to_le16((0 << 9) | (1 << 5) | 1);
 		if (time_cs)
 			*time_cs = 0;
 		return;
 	}
-#if BITS_PER_LONG == 64
-	if (second >= UNIX_SECS_2108) {
+	if (tm.tm_year > 2107 - 1900) {
 		*time = cpu_to_le16((23 << 11) | (59 << 5) | 29);
 		*date = cpu_to_le16((127 << 9) | (12 << 5) | 31);
 		if (time_cs)
 			*time_cs = 199;
 		return;
 	}
-#endif
 
-	day = second / SECS_PER_DAY - DAYS_DELTA;
-	year = day / 365;
-	leap_day = (year + 3) / 4;
-	if (year > YEAR_2100)		/* 2100 isn't leap year */
-		leap_day--;
-	if (year * 365 + leap_day > day)
-		year--;
-	leap_day = (year + 3) / 4;
-	if (year > YEAR_2100)		/* 2100 isn't leap year */
-		leap_day--;
-	day -= year * 365 + leap_day;
-
-	if (IS_LEAP_YEAR(year) && day == days_in_year[3]) {
-		month = 2;
-	} else {
-		if (IS_LEAP_YEAR(year) && day > days_in_year[3])
-			day--;
-		for (month = 1; month < 12; month++) {
-			if (days_in_year[month + 1] > day)
-				break;
-		}
-	}
-	day -= days_in_year[month];
+	/* from 1900 -> from 1980 */
+	tm.tm_year -= 80;
+	/* 0~11 -> 1~12 */
+	tm.tm_mon++;
+	/* 0~59 -> 0~29(2sec counts) */
+	tm.tm_sec >>= 1;
 
-	*time = cpu_to_le16(((second / SECS_PER_HOUR) % 24) << 11
-			    | ((second / SECS_PER_MIN) % 60) << 5
-			    | (second % SECS_PER_MIN) >> 1);
-	*date = cpu_to_le16((year << 9) | (month << 5) | (day + 1));
+	*time = cpu_to_le16(tm.tm_hour << 11 | tm.tm_min << 5 | tm.tm_sec);
+	*date = cpu_to_le16(tm.tm_year << 9 | tm.tm_mon << 5 | tm.tm_mday);
 	if (time_cs)
 		*time_cs = (ts->tv_sec & 1) * 100 + ts->tv_nsec / 10000000;
 }
@@ -285,4 +259,3 @@ int fat_sync_bhs(struct buffer_head **bhs, int nr_bhs)
 	}
 	return err;
 }
-
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 2cf93ec40a67..5ef953e6f908 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -199,7 +199,9 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
 static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
                      int force)
 {
-	write_lock_irq(&filp->f_owner.lock);
+	unsigned long flags;
+
+	write_lock_irqsave(&filp->f_owner.lock, flags);
 	if (force || !filp->f_owner.pid) {
 		put_pid(filp->f_owner.pid);
 		filp->f_owner.pid = get_pid(pid);
@@ -211,7 +213,7 @@ static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
 			filp->f_owner.euid = cred->euid;
 		}
 	}
-	write_unlock_irq(&filp->f_owner.lock);
+	write_unlock_irqrestore(&filp->f_owner.lock, flags);
 }
 
 int __f_setown(struct file *filp, struct pid *pid, enum pid_type type,
@@ -618,60 +620,90 @@ static DEFINE_RWLOCK(fasync_lock);
 static struct kmem_cache *fasync_cache __read_mostly;
 
 /*
- * fasync_helper() is used by almost all character device drivers
- * to set up the fasync queue. It returns negative on error, 0 if it did
- * no changes and positive if it added/deleted the entry.
+ * Remove a fasync entry. If successfully removed, return
+ * positive and clear the FASYNC flag. If no entry exists,
+ * do nothing and return 0.
+ *
+ * NOTE! It is very important that the FASYNC flag always
+ * match the state "is the filp on a fasync list".
+ *
+ * We always take the 'filp->f_lock', in since fasync_lock
+ * needs to be irq-safe.
  */
-int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fapp)
+static int fasync_remove_entry(struct file *filp, struct fasync_struct **fapp)
 {
 	struct fasync_struct *fa, **fp;
-	struct fasync_struct *new = NULL;
 	int result = 0;
 
-	if (on) {
-		new = kmem_cache_alloc(fasync_cache, GFP_KERNEL);
-		if (!new)
-			return -ENOMEM;
+	spin_lock(&filp->f_lock);
+	write_lock_irq(&fasync_lock);
+	for (fp = fapp; (fa = *fp) != NULL; fp = &fa->fa_next) {
+		if (fa->fa_file != filp)
+			continue;
+		*fp = fa->fa_next;
+		kmem_cache_free(fasync_cache, fa);
+		filp->f_flags &= ~FASYNC;
+		result = 1;
+		break;
 	}
+	write_unlock_irq(&fasync_lock);
+	spin_unlock(&filp->f_lock);
+	return result;
+}
+
+/*
+ * Add a fasync entry. Return negative on error, positive if
+ * added, and zero if did nothing but change an existing one.
+ *
+ * NOTE! It is very important that the FASYNC flag always
+ * match the state "is the filp on a fasync list".
+ */
+static int fasync_add_entry(int fd, struct file *filp, struct fasync_struct **fapp)
+{
+	struct fasync_struct *new, *fa, **fp;
+	int result = 0;
+
+	new = kmem_cache_alloc(fasync_cache, GFP_KERNEL);
+	if (!new)
+		return -ENOMEM;
 
-	/*
-	 * We need to take f_lock first since it's not an IRQ-safe
-	 * lock.
-	 */
 	spin_lock(&filp->f_lock);
 	write_lock_irq(&fasync_lock);
 	for (fp = fapp; (fa = *fp) != NULL; fp = &fa->fa_next) {
-		if (fa->fa_file == filp) {
-			if(on) {
-				fa->fa_fd = fd;
-				kmem_cache_free(fasync_cache, new);
-			} else {
-				*fp = fa->fa_next;
-				kmem_cache_free(fasync_cache, fa);
-				result = 1;
-			}
-			goto out;
-		}
+		if (fa->fa_file != filp)
+			continue;
+		fa->fa_fd = fd;
+		kmem_cache_free(fasync_cache, new);
+		goto out;
 	}
 
-	if (on) {
-		new->magic = FASYNC_MAGIC;
-		new->fa_file = filp;
-		new->fa_fd = fd;
-		new->fa_next = *fapp;
-		*fapp = new;
-		result = 1;
-	}
+	new->magic = FASYNC_MAGIC;
+	new->fa_file = filp;
+	new->fa_fd = fd;
+	new->fa_next = *fapp;
+	*fapp = new;
+	result = 1;
+	filp->f_flags |= FASYNC;
+
 out:
-	if (on)
-		filp->f_flags |= FASYNC;
-	else
-		filp->f_flags &= ~FASYNC;
 	write_unlock_irq(&fasync_lock);
 	spin_unlock(&filp->f_lock);
 	return result;
 }
 
+/*
+ * fasync_helper() is used by almost all character device drivers
+ * to set up the fasync queue, and for regular files by the file
+ * lease code. It returns negative on error, 0 if it did no changes
+ * and positive if it added/deleted the entry.
+ */
+int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fapp)
+{
+	if (!on)
+		return fasync_remove_entry(filp, fapp);
+	return fasync_add_entry(fd, filp, fapp);
+}
+
 EXPORT_SYMBOL(fasync_helper);
 
 void __kill_fasync(struct fasync_struct *fa, int sig, int band)
diff --git a/fs/file_table.c b/fs/file_table.c
index 4bef4c01ec6f..69652c5bd5f0 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -21,9 +21,12 @@
 #include <linux/fsnotify.h>
 #include <linux/sysctl.h>
 #include <linux/percpu_counter.h>
+#include <linux/ima.h>
 
 #include <asm/atomic.h>
 
+#include "internal.h"
+
 /* sysctl tunables... */
 struct files_stat_struct files_stat = {
 	.max_files = NR_FILE
@@ -147,8 +150,6 @@ fail:
 	return NULL;
 }
 
-EXPORT_SYMBOL(get_empty_filp);
-
 /**
  * alloc_file - allocate and initialize a 'struct file'
  * @mnt: the vfsmount on which the file will reside
@@ -164,8 +165,8 @@ EXPORT_SYMBOL(get_empty_filp);
  * If all the callers of init_file() are eliminated, its
  * code should be moved into this function.
  */
-struct file *alloc_file(struct vfsmount *mnt, struct dentry *dentry,
-		fmode_t mode, const struct file_operations *fop)
+struct file *alloc_file(struct path *path, fmode_t mode,
+		const struct file_operations *fop)
 {
 	struct file *file;
 
@@ -173,35 +174,8 @@ struct file *alloc_file(struct vfsmount *mnt, struct dentry *dentry,
 	if (!file)
 		return NULL;
 
-	init_file(file, mnt, dentry, mode, fop);
-	return file;
-}
-EXPORT_SYMBOL(alloc_file);
-
-/**
- * init_file - initialize a 'struct file'
- * @file: the already allocated 'struct file' to initialized
- * @mnt: the vfsmount on which the file resides
- * @dentry: the dentry representing this file
- * @mode: the mode the file is opened with
- * @fop: the 'struct file_operations' for this file
- *
- * Use this instead of setting the members directly.  Doing so
- * avoids making mistakes like forgetting the mntget() or
- * forgetting to take a write on the mnt.
- *
- * Note: This is a crappy interface.  It is here to make
- * merging with the existing users of get_empty_filp()
- * who have complex failure logic easier.  All users
- * of this should be moving to alloc_file().
- */
-int init_file(struct file *file, struct vfsmount *mnt, struct dentry *dentry,
-	   fmode_t mode, const struct file_operations *fop)
-{
-	int error = 0;
-	file->f_path.dentry = dentry;
-	file->f_path.mnt = mntget(mnt);
-	file->f_mapping = dentry->d_inode->i_mapping;
+	file->f_path = *path;
+	file->f_mapping = path->dentry->d_inode->i_mapping;
 	file->f_mode = mode;
 	file->f_op = fop;
 
@@ -211,14 +185,14 @@ int init_file(struct file *file, struct vfsmount *mnt, struct dentry *dentry,
 	 * visible.  We do this for consistency, and so
 	 * that we can do debugging checks at __fput()
 	 */
-	if ((mode & FMODE_WRITE) && !special_file(dentry->d_inode->i_mode)) {
+	if ((mode & FMODE_WRITE) && !special_file(path->dentry->d_inode->i_mode)) {
 		file_take_write(file);
-		error = mnt_clone_write(mnt);
-		WARN_ON(error);
+		WARN_ON(mnt_clone_write(path->mnt));
 	}
-	return error;
+	ima_counts_get(file);
+	return file;
 }
-EXPORT_SYMBOL(init_file);
+EXPORT_SYMBOL(alloc_file);
 
 void fput(struct file *file)
 {
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 49bc1b8e8f19..1a7c42c64ff4 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -242,6 +242,7 @@ static void bdi_sync_writeback(struct backing_dev_info *bdi,
 /**
  * bdi_start_writeback - start writeback
  * @bdi: the backing device to write from
+ * @sb: write inodes from this super_block
  * @nr_pages: the number of pages to write
  *
  * Description:
@@ -1187,6 +1188,23 @@ void writeback_inodes_sb(struct super_block *sb)
 EXPORT_SYMBOL(writeback_inodes_sb);
 
 /**
+ * writeback_inodes_sb_if_idle	-	start writeback if none underway
+ * @sb: the superblock
+ *
+ * Invoke writeback_inodes_sb if no writeback is currently underway.
+ * Returns 1 if writeback was started, 0 if not.
+ */
+int writeback_inodes_sb_if_idle(struct super_block *sb)
+{
+	if (!writeback_in_progress(sb->s_bdi)) {
+		writeback_inodes_sb(sb);
+		return 1;
+	} else
+		return 0;
+}
+EXPORT_SYMBOL(writeback_inodes_sb_if_idle);
+
+/**
  * sync_inodes_sb	-	sync sb inode pages
  * @sb: the superblock
  *
diff --git a/fs/fscache/object-list.c b/fs/fscache/object-list.c
index e590242fa41a..3221a0c7944e 100644
--- a/fs/fscache/object-list.c
+++ b/fs/fscache/object-list.c
@@ -91,7 +91,7 @@ EXPORT_SYMBOL(fscache_object_destroy);
  */
 static struct fscache_object *fscache_objlist_lookup(loff_t *_pos)
 {
-	struct fscache_object *pobj, *obj, *minobj = NULL;
+	struct fscache_object *pobj, *obj = NULL, *minobj = NULL;
 	struct rb_node *p;
 	unsigned long pos;
 
diff --git a/fs/generic_acl.c b/fs/generic_acl.c
index e0b53aa7bbec..55458031e501 100644
--- a/fs/generic_acl.c
+++ b/fs/generic_acl.c
@@ -1,62 +1,58 @@
 /*
- * fs/generic_acl.c
- *
  * (C) 2005 Andreas Gruenbacher <agruen@suse.de>
  *
  * This file is released under the GPL.
+ *
+ * Generic ACL support for in-memory filesystems.
  */
 
 #include <linux/sched.h>
 #include <linux/fs.h>
 #include <linux/generic_acl.h>
+#include <linux/posix_acl.h>
+#include <linux/posix_acl_xattr.h>
 
-/**
- * generic_acl_list  -  Generic xattr_handler->list() operation
- * @ops:	Filesystem specific getacl and setacl callbacks
- */
-size_t
-generic_acl_list(struct inode *inode, struct generic_acl_operations *ops,
-		 int type, char *list, size_t list_size)
+
+static size_t
+generic_acl_list(struct dentry *dentry, char *list, size_t list_size,
+		const char *name, size_t name_len, int type)
 {
 	struct posix_acl *acl;
-	const char *name;
+	const char *xname;
 	size_t size;
 
-	acl = ops->getacl(inode, type);
+	acl = get_cached_acl(dentry->d_inode, type);
 	if (!acl)
 		return 0;
 	posix_acl_release(acl);
 
-	switch(type) {
-		case ACL_TYPE_ACCESS:
-			name = POSIX_ACL_XATTR_ACCESS;
-			break;
-
-		case ACL_TYPE_DEFAULT:
-			name = POSIX_ACL_XATTR_DEFAULT;
-			break;
-
-		default:
-			return 0;
+	switch (type) {
+	case ACL_TYPE_ACCESS:
+		xname = POSIX_ACL_XATTR_ACCESS;
+		break;
+	case ACL_TYPE_DEFAULT:
+		xname = POSIX_ACL_XATTR_DEFAULT;
+		break;
+	default:
+		return 0;
 	}
-	size = strlen(name) + 1;
+	size = strlen(xname) + 1;
 	if (list && size <= list_size)
-		memcpy(list, name, size);
+		memcpy(list, xname, size);
 	return size;
 }
 
-/**
- * generic_acl_get  -  Generic xattr_handler->get() operation
- * @ops:	Filesystem specific getacl and setacl callbacks
- */
-int
-generic_acl_get(struct inode *inode, struct generic_acl_operations *ops,
-		int type, void *buffer, size_t size)
+static int
+generic_acl_get(struct dentry *dentry, const char *name, void *buffer,
+		     size_t size, int type)
 {
 	struct posix_acl *acl;
 	int error;
 
-	acl = ops->getacl(inode, type);
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
+
+	acl = get_cached_acl(dentry->d_inode, type);
 	if (!acl)
 		return -ENODATA;
 	error = posix_acl_to_xattr(acl, buffer, size);
@@ -65,17 +61,16 @@ generic_acl_get(struct inode *inode, struct generic_acl_operations *ops,
 	return error;
 }
 
-/**
- * generic_acl_set  -  Generic xattr_handler->set() operation
- * @ops:	Filesystem specific getacl and setacl callbacks
- */
-int
-generic_acl_set(struct inode *inode, struct generic_acl_operations *ops,
-		int type, const void *value, size_t size)
+static int
+generic_acl_set(struct dentry *dentry, const char *name, const void *value,
+		     size_t size, int flags, int type)
 {
+	struct inode *inode = dentry->d_inode;
 	struct posix_acl *acl = NULL;
 	int error;
 
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
 	if (S_ISLNK(inode->i_mode))
 		return -EOPNOTSUPP;
 	if (!is_owner_or_cap(inode))
@@ -91,28 +86,27 @@ generic_acl_set(struct inode *inode, struct generic_acl_operations *ops,
 		error = posix_acl_valid(acl);
 		if (error)
 			goto failed;
-		switch(type) {
-			case ACL_TYPE_ACCESS:
-				mode = inode->i_mode;
-				error = posix_acl_equiv_mode(acl, &mode);
-				if (error < 0)
-					goto failed;
-				inode->i_mode = mode;
-				if (error == 0) {
-					posix_acl_release(acl);
-					acl = NULL;
-				}
-				break;
-
-			case ACL_TYPE_DEFAULT:
-				if (!S_ISDIR(inode->i_mode)) {
-					error = -EINVAL;
-					goto failed;
-				}
-				break;
+		switch (type) {
+		case ACL_TYPE_ACCESS:
+			mode = inode->i_mode;
+			error = posix_acl_equiv_mode(acl, &mode);
+			if (error < 0)
+				goto failed;
+			inode->i_mode = mode;
+			if (error == 0) {
+				posix_acl_release(acl);
+				acl = NULL;
+			}
+			break;
+		case ACL_TYPE_DEFAULT:
+			if (!S_ISDIR(inode->i_mode)) {
+				error = -EINVAL;
+				goto failed;
+			}
+			break;
 		}
 	}
-	ops->setacl(inode, type, acl);
+	set_cached_acl(inode, type, acl);
 	error = 0;
 failed:
 	posix_acl_release(acl);
@@ -121,14 +115,12 @@ failed:
 
 /**
  * generic_acl_init  -  Take care of acl inheritance at @inode create time
- * @ops:	Filesystem specific getacl and setacl callbacks
  *
  * Files created inside a directory with a default ACL inherit the
  * directory's default ACL.
  */
 int
-generic_acl_init(struct inode *inode, struct inode *dir,
-		 struct generic_acl_operations *ops)
+generic_acl_init(struct inode *inode, struct inode *dir)
 {
 	struct posix_acl *acl = NULL;
 	mode_t mode = inode->i_mode;
@@ -136,7 +128,7 @@ generic_acl_init(struct inode *inode, struct inode *dir,
 
 	inode->i_mode = mode & ~current_umask();
 	if (!S_ISLNK(inode->i_mode))
-		acl = ops->getacl(dir, ACL_TYPE_DEFAULT);
+		acl = get_cached_acl(dir, ACL_TYPE_DEFAULT);
 	if (acl) {
 		struct posix_acl *clone;
 
@@ -145,7 +137,7 @@ generic_acl_init(struct inode *inode, struct inode *dir,
 			error = -ENOMEM;
 			if (!clone)
 				goto cleanup;
-			ops->setacl(inode, ACL_TYPE_DEFAULT, clone);
+			set_cached_acl(inode, ACL_TYPE_DEFAULT, clone);
 			posix_acl_release(clone);
 		}
 		clone = posix_acl_clone(acl, GFP_KERNEL);
@@ -156,7 +148,7 @@ generic_acl_init(struct inode *inode, struct inode *dir,
 		if (error >= 0) {
 			inode->i_mode = mode;
 			if (error > 0)
-				ops->setacl(inode, ACL_TYPE_ACCESS, clone);
+				set_cached_acl(inode, ACL_TYPE_ACCESS, clone);
 		}
 		posix_acl_release(clone);
 	}
@@ -169,20 +161,19 @@ cleanup:
 
 /**
  * generic_acl_chmod  -  change the access acl of @inode upon chmod()
- * @ops:	FIlesystem specific getacl and setacl callbacks
  *
  * A chmod also changes the permissions of the owner, group/mask, and
  * other ACL entries.
  */
 int
-generic_acl_chmod(struct inode *inode, struct generic_acl_operations *ops)
+generic_acl_chmod(struct inode *inode)
 {
 	struct posix_acl *acl, *clone;
 	int error = 0;
 
 	if (S_ISLNK(inode->i_mode))
 		return -EOPNOTSUPP;
-	acl = ops->getacl(inode, ACL_TYPE_ACCESS);
+	acl = get_cached_acl(inode, ACL_TYPE_ACCESS);
 	if (acl) {
 		clone = posix_acl_clone(acl, GFP_KERNEL);
 		posix_acl_release(acl);
@@ -190,8 +181,37 @@ generic_acl_chmod(struct inode *inode, struct generic_acl_operations *ops)
 			return -ENOMEM;
 		error = posix_acl_chmod_masq(clone, inode->i_mode);
 		if (!error)
-			ops->setacl(inode, ACL_TYPE_ACCESS, clone);
+			set_cached_acl(inode, ACL_TYPE_ACCESS, clone);
 		posix_acl_release(clone);
 	}
 	return error;
 }
+
+int
+generic_check_acl(struct inode *inode, int mask)
+{
+	struct posix_acl *acl = get_cached_acl(inode, ACL_TYPE_ACCESS);
+
+	if (acl) {
+		int error = posix_acl_permission(inode, acl, mask);
+		posix_acl_release(acl);
+		return error;
+	}
+	return -EAGAIN;
+}
+
+struct xattr_handler generic_acl_access_handler = {
+	.prefix = POSIX_ACL_XATTR_ACCESS,
+	.flags	= ACL_TYPE_ACCESS,
+	.list	= generic_acl_list,
+	.get	= generic_acl_get,
+	.set	= generic_acl_set,
+};
+
+struct xattr_handler generic_acl_default_handler = {
+	.prefix = POSIX_ACL_XATTR_DEFAULT,
+	.flags	= ACL_TYPE_DEFAULT,
+	.list	= generic_acl_list,
+	.get	= generic_acl_get,
+	.set	= generic_acl_set,
+};
diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c
index 3eb1ea846173..87ee309d4c24 100644
--- a/fs/gfs2/acl.c
+++ b/fs/gfs2/acl.c
@@ -126,7 +126,7 @@ static int gfs2_acl_set(struct inode *inode, int type, struct posix_acl *acl)
 	error = posix_acl_to_xattr(acl, data, len);
 	if (error < 0)
 		goto out;
-	error = gfs2_xattr_set(inode, GFS2_EATYPE_SYS, name, data, len, 0);
+	error = __gfs2_xattr_set(inode, name, data, len, 0, GFS2_EATYPE_SYS);
 	if (!error)
 		set_cached_acl(inode, type, acl);
 out:
@@ -232,9 +232,10 @@ static int gfs2_acl_type(const char *name)
 	return -EINVAL;
 }
 
-static int gfs2_xattr_system_get(struct inode *inode, const char *name,
-				 void *buffer, size_t size)
+static int gfs2_xattr_system_get(struct dentry *dentry, const char *name,
+				 void *buffer, size_t size, int xtype)
 {
+	struct inode *inode = dentry->d_inode;
 	struct posix_acl *acl;
 	int type;
 	int error;
@@ -255,9 +256,11 @@ static int gfs2_xattr_system_get(struct inode *inode, const char *name,
 	return error;
 }
 
-static int gfs2_xattr_system_set(struct inode *inode, const char *name,
-				 const void *value, size_t size, int flags)
+static int gfs2_xattr_system_set(struct dentry *dentry, const char *name,
+				 const void *value, size_t size, int flags,
+				 int xtype)
 {
+	struct inode *inode = dentry->d_inode;
 	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct posix_acl *acl = NULL;
 	int error = 0, type;
@@ -319,7 +322,7 @@ static int gfs2_xattr_system_set(struct inode *inode, const char *name,
 	}
 
 set_acl:
-	error = gfs2_xattr_set(inode, GFS2_EATYPE_SYS, name, value, size, 0);
+	error = __gfs2_xattr_set(inode, name, value, size, 0, GFS2_EATYPE_SYS);
 	if (!error) {
 		if (acl)
 			set_cached_acl(inode, type, acl);
@@ -334,6 +337,7 @@ out:
 
 struct xattr_handler gfs2_xattr_system_handler = {
 	.prefix = XATTR_SYSTEM_PREFIX,
+	.flags  = GFS2_EATYPE_SYS,
 	.get    = gfs2_xattr_system_get,
 	.set    = gfs2_xattr_system_set,
 };
diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c
index 4eb308aa3234..a6abbae8a278 100644
--- a/fs/gfs2/file.c
+++ b/fs/gfs2/file.c
@@ -569,6 +569,40 @@ static int gfs2_fsync(struct file *file, struct dentry *dentry, int datasync)
 	return ret;
 }
 
+/**
+ * gfs2_file_aio_write - Perform a write to a file
+ * @iocb: The io context
+ * @iov: The data to write
+ * @nr_segs: Number of @iov segments
+ * @pos: The file position
+ *
+ * We have to do a lock/unlock here to refresh the inode size for
+ * O_APPEND writes, otherwise we can land up writing at the wrong
+ * offset. There is still a race, but provided the app is using its
+ * own file locking, this will make O_APPEND work as expected.
+ *
+ */
+
+static ssize_t gfs2_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
+				   unsigned long nr_segs, loff_t pos)
+{
+	struct file *file = iocb->ki_filp;
+
+	if (file->f_flags & O_APPEND) {
+		struct dentry *dentry = file->f_dentry;
+		struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
+		struct gfs2_holder gh;
+		int ret;
+
+		ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh);
+		if (ret)
+			return ret;
+		gfs2_glock_dq_uninit(&gh);
+	}
+
+	return generic_file_aio_write(iocb, iov, nr_segs, pos);
+}
+
 #ifdef CONFIG_GFS2_FS_LOCKING_DLM
 
 /**
@@ -711,7 +745,7 @@ const struct file_operations gfs2_file_fops = {
 	.read		= do_sync_read,
 	.aio_read	= generic_file_aio_read,
 	.write		= do_sync_write,
-	.aio_write	= generic_file_aio_write,
+	.aio_write	= gfs2_file_aio_write,
 	.unlocked_ioctl	= gfs2_ioctl,
 	.mmap		= gfs2_mmap,
 	.open		= gfs2_open,
@@ -741,7 +775,7 @@ const struct file_operations gfs2_file_fops_nolock = {
 	.read		= do_sync_read,
 	.aio_read	= generic_file_aio_read,
 	.write		= do_sync_write,
-	.aio_write	= generic_file_aio_write,
+	.aio_write	= gfs2_file_aio_write,
 	.unlocked_ioctl	= gfs2_ioctl,
 	.mmap		= gfs2_mmap,
 	.open		= gfs2_open,
diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 26ba2a4c4a2d..6e220f4eee7d 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -125,7 +125,7 @@ static struct inode *gfs2_iget_skip(struct super_block *sb,
  * directory entry when gfs2_inode_lookup() is invoked. Part of the code
  * segment inside gfs2_inode_lookup code needs to get moved around.
  *
- * Clean up I_LOCK and I_NEW as well.
+ * Clears I_NEW as well.
  **/
 
 void gfs2_set_iop(struct inode *inode)
@@ -801,7 +801,8 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip)
 		return err;
 	}
 
-	err = gfs2_xattr_set(&ip->i_inode, GFS2_EATYPE_SECURITY, name, value, len, 0);
+	err = __gfs2_xattr_set(&ip->i_inode, name, value, len, 0,
+			       GFS2_EATYPE_SECURITY);
 	kfree(value);
 	kfree(name);
 
diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c
index cb8d7a93d5ec..6f68a5f18eb8 100644
--- a/fs/gfs2/meta_io.c
+++ b/fs/gfs2/meta_io.c
@@ -121,7 +121,7 @@ struct inode *gfs2_aspace_get(struct gfs2_sbd *sdp)
 	if (aspace) {
 		mapping_set_gfp_mask(aspace->i_mapping, GFP_NOFS);
 		aspace->i_mapping->a_ops = &aspace_aops;
-		aspace->i_size = ~0ULL;
+		aspace->i_size = MAX_LFS_FILESIZE;
 		ip = GFS2_I(aspace);
 		clear_bit(GIF_USER, &ip->i_flags);
 		insert_inode_hash(aspace);
diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index 247436c10deb..84350e1be66d 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -748,7 +748,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
 	struct gfs2_rgrpd *nrgd;
 	unsigned int num_gh;
 	int dir_rename = 0;
-	int alloc_required;
+	int alloc_required = 0;
 	unsigned int x;
 	int error;
 
@@ -867,7 +867,9 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry,
 			goto out_gunlock;
 	}
 
-	alloc_required = error = gfs2_diradd_alloc_required(ndir, &ndentry->d_name);
+	if (nip == NULL)
+		alloc_required = gfs2_diradd_alloc_required(ndir, &ndentry->d_name);
+	error = alloc_required;
 	if (error < 0)
 		goto out_gunlock;
 	error = 0;
@@ -1086,7 +1088,8 @@ static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd)
 		error = vfs_follow_link(nd, buf);
 		if (buf != array)
 			kfree(buf);
-	}
+	} else
+		path_put(&nd->path);
 
 	return ERR_PTR(error);
 }
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 0608f490c295..503b842f3ba2 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -591,11 +591,7 @@ static int gfs2_ri_update(struct gfs2_inode *ip)
 	u64 rgrp_count = ip->i_disksize;
 	int error;
 
-	if (do_div(rgrp_count, sizeof(struct gfs2_rindex))) {
-		gfs2_consist_inode(ip);
-		return -EIO;
-	}
-
+	do_div(rgrp_count, sizeof(struct gfs2_rindex));
 	clear_rgrpdi(sdp);
 
 	file_ra_state_init(&ra_state, inode->i_mapping);
@@ -915,7 +911,7 @@ void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd)
 struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip)
 {
 	BUG_ON(ip->i_alloc != NULL);
-	ip->i_alloc = kzalloc(sizeof(struct gfs2_alloc), GFP_KERNEL);
+	ip->i_alloc = kzalloc(sizeof(struct gfs2_alloc), GFP_NOFS);
 	return ip->i_alloc;
 }
 
diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c
index c5dad1eb7b91..0dc34621f6a6 100644
--- a/fs/gfs2/sys.c
+++ b/fs/gfs2/sys.c
@@ -85,11 +85,7 @@ static ssize_t uuid_show(struct gfs2_sbd *sdp, char *buf)
 	buf[0] = '\0';
 	if (!gfs2_uuid_valid(uuid))
 		return 0;
-	return snprintf(buf, PAGE_SIZE, "%02X%02X%02X%02X-%02X%02X-"
-			"%02X%02X-%02X%02X-%02X%02X%02X%02X%02X%02X\n",
-			uuid[0], uuid[1], uuid[2], uuid[3], uuid[4], uuid[5],
-			uuid[6], uuid[7], uuid[8], uuid[9], uuid[10], uuid[11],
-			uuid[12], uuid[13], uuid[14], uuid[15]);
+	return snprintf(buf, PAGE_SIZE, "%pUB\n", uuid);
 }
 
 static ssize_t freeze_show(struct gfs2_sbd *sdp, char *buf)
@@ -575,14 +571,8 @@ static int gfs2_uevent(struct kset *kset, struct kobject *kobj,
 	add_uevent_var(env, "LOCKPROTO=%s", sdp->sd_proto_name);
 	if (!sdp->sd_args.ar_spectator)
 		add_uevent_var(env, "JOURNALID=%u", sdp->sd_lockstruct.ls_jid);
-	if (gfs2_uuid_valid(uuid)) {
-		add_uevent_var(env, "UUID=%02X%02X%02X%02X-%02X%02X-%02X%02X-"
-			       "%02X%02X-%02X%02X%02X%02X%02X%02X",
-			       uuid[0], uuid[1], uuid[2], uuid[3], uuid[4],
-			       uuid[5], uuid[6], uuid[7], uuid[8], uuid[9],
-			       uuid[10], uuid[11], uuid[12], uuid[13],
-			       uuid[14], uuid[15]);
-	}
+	if (gfs2_uuid_valid(uuid))
+		add_uevent_var(env, "UUID=%pUB", uuid);
 	return 0;
 }
 
diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c
index 912f5cbc4740..c2ebdf2c01d4 100644
--- a/fs/gfs2/xattr.c
+++ b/fs/gfs2/xattr.c
@@ -567,18 +567,17 @@ out:
 /**
  * gfs2_xattr_get - Get a GFS2 extended attribute
  * @inode: The inode
- * @type: The type of extended attribute
  * @name: The name of the extended attribute
  * @buffer: The buffer to write the result into
  * @size: The size of the buffer
+ * @type: The type of extended attribute
  *
  * Returns: actual size of data on success, -errno on error
  */
-
-int gfs2_xattr_get(struct inode *inode, int type, const char *name,
-		   void *buffer, size_t size)
+static int gfs2_xattr_get(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
-	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_inode *ip = GFS2_I(dentry->d_inode);
 	struct gfs2_ea_location el;
 	int error;
 
@@ -1119,7 +1118,7 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)
 
 /**
  * gfs2_xattr_remove - Remove a GFS2 extended attribute
- * @inode: The inode
+ * @ip: The inode
  * @type: The type of the extended attribute
  * @name: The name of the extended attribute
  *
@@ -1130,9 +1129,8 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el)
  * Returns: 0, or errno on failure
  */
 
-static int gfs2_xattr_remove(struct inode *inode, int type, const char *name)
+static int gfs2_xattr_remove(struct gfs2_inode *ip, int type, const char *name)
 {
-	struct gfs2_inode *ip = GFS2_I(inode);
 	struct gfs2_ea_location el;
 	int error;
 
@@ -1156,24 +1154,24 @@ static int gfs2_xattr_remove(struct inode *inode, int type, const char *name)
 }
 
 /**
- * gfs2_xattr_set - Set (or remove) a GFS2 extended attribute
- * @inode: The inode
- * @type: The type of the extended attribute
+ * __gfs2_xattr_set - Set (or remove) a GFS2 extended attribute
+ * @ip: The inode
  * @name: The name of the extended attribute
  * @value: The value of the extended attribute (NULL for remove)
  * @size: The size of the @value argument
  * @flags: Create or Replace
+ * @type: The type of the extended attribute
  *
  * See gfs2_xattr_remove() for details of the removal of xattrs.
  *
  * Returns: 0 or errno on failure
  */
 
-int gfs2_xattr_set(struct inode *inode, int type, const char *name,
-		   const void *value, size_t size, int flags)
+int __gfs2_xattr_set(struct inode *inode, const char *name,
+		   const void *value, size_t size, int flags, int type)
 {
-	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct gfs2_inode *ip = GFS2_I(inode);
+	struct gfs2_sbd *sdp = GFS2_SB(inode);
 	struct gfs2_ea_location el;
 	unsigned int namel = strlen(name);
 	int error;
@@ -1184,7 +1182,7 @@ int gfs2_xattr_set(struct inode *inode, int type, const char *name,
 		return -ERANGE;
 
 	if (value == NULL)
-		return gfs2_xattr_remove(inode, type, name);
+		return gfs2_xattr_remove(ip, type, name);
 
 	if (ea_check_size(sdp, namel, size))
 		return -ERANGE;
@@ -1224,6 +1222,13 @@ int gfs2_xattr_set(struct inode *inode, int type, const char *name,
 	return error;
 }
 
+static int gfs2_xattr_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
+{
+	return __gfs2_xattr_set(dentry->d_inode, name, value,
+				size, flags, type);
+}
+
 static int ea_acl_chmod_unstuffed(struct gfs2_inode *ip,
 				  struct gfs2_ea_header *ea, char *data)
 {
@@ -1291,6 +1296,7 @@ fail:
 
 int gfs2_xattr_acl_chmod(struct gfs2_inode *ip, struct iattr *attr, char *data)
 {
+	struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode);
 	struct gfs2_ea_location el;
 	struct buffer_head *dibh;
 	int error;
@@ -1300,16 +1306,17 @@ int gfs2_xattr_acl_chmod(struct gfs2_inode *ip, struct iattr *attr, char *data)
 		return error;
 
 	if (GFS2_EA_IS_STUFFED(el.el_ea)) {
-		error = gfs2_trans_begin(GFS2_SB(&ip->i_inode), RES_DINODE + RES_EATTR, 0);
-		if (error)
-			return error;
-
-		gfs2_trans_add_bh(ip->i_gl, el.el_bh, 1);
-		memcpy(GFS2_EA2DATA(el.el_ea), data,
-		       GFS2_EA_DATA_LEN(el.el_ea));
-	} else
+		error = gfs2_trans_begin(sdp, RES_DINODE + RES_EATTR, 0);
+		if (error == 0) {
+			gfs2_trans_add_bh(ip->i_gl, el.el_bh, 1);
+			memcpy(GFS2_EA2DATA(el.el_ea), data,
+			       GFS2_EA_DATA_LEN(el.el_ea));
+		}
+	} else {
 		error = ea_acl_chmod_unstuffed(ip, el.el_ea, data);
+	}
 
+	brelse(el.el_bh);
 	if (error)
 		return error;
 
@@ -1322,8 +1329,7 @@ int gfs2_xattr_acl_chmod(struct gfs2_inode *ip, struct iattr *attr, char *data)
 		brelse(dibh);
 	}
 
-	gfs2_trans_end(GFS2_SB(&ip->i_inode));
-
+	gfs2_trans_end(sdp);
 	return error;
 }
 
@@ -1529,40 +1535,18 @@ out_alloc:
 	return error;
 }
 
-static int gfs2_xattr_user_get(struct inode *inode, const char *name,
-			       void *buffer, size_t size)
-{
-	return gfs2_xattr_get(inode, GFS2_EATYPE_USR, name, buffer, size);
-}
-
-static int gfs2_xattr_user_set(struct inode *inode, const char *name,
-			       const void *value, size_t size, int flags)
-{
-	return gfs2_xattr_set(inode, GFS2_EATYPE_USR, name, value, size, flags);
-}
-
-static int gfs2_xattr_security_get(struct inode *inode, const char *name,
-				   void *buffer, size_t size)
-{
-	return gfs2_xattr_get(inode, GFS2_EATYPE_SECURITY, name, buffer, size);
-}
-
-static int gfs2_xattr_security_set(struct inode *inode, const char *name,
-				   const void *value, size_t size, int flags)
-{
-	return gfs2_xattr_set(inode, GFS2_EATYPE_SECURITY, name, value, size, flags);
-}
-
 static struct xattr_handler gfs2_xattr_user_handler = {
 	.prefix = XATTR_USER_PREFIX,
-	.get    = gfs2_xattr_user_get,
-	.set    = gfs2_xattr_user_set,
+	.flags  = GFS2_EATYPE_USR,
+	.get    = gfs2_xattr_get,
+	.set    = gfs2_xattr_set,
 };
 
 static struct xattr_handler gfs2_xattr_security_handler = {
 	.prefix = XATTR_SECURITY_PREFIX,
-	.get    = gfs2_xattr_security_get,
-	.set    = gfs2_xattr_security_set,
+	.flags  = GFS2_EATYPE_SECURITY,
+	.get    = gfs2_xattr_get,
+	.set    = gfs2_xattr_set,
 };
 
 struct xattr_handler *gfs2_xattr_handlers[] = {
diff --git a/fs/gfs2/xattr.h b/fs/gfs2/xattr.h
index 8d6ae5813c4d..d392f8358f2f 100644
--- a/fs/gfs2/xattr.h
+++ b/fs/gfs2/xattr.h
@@ -53,10 +53,9 @@ struct gfs2_ea_location {
 	struct gfs2_ea_header *el_prev;
 };
 
-extern int gfs2_xattr_get(struct inode *inode, int type, const char *name,
-			  void *buffer, size_t size);
-extern int gfs2_xattr_set(struct inode *inode, int type, const char *name,
-			  const void *value, size_t size, int flags);
+extern int __gfs2_xattr_set(struct inode *inode, const char *name,
+			    const void *value, size_t size,
+			    int flags, int type);
 extern ssize_t gfs2_listxattr(struct dentry *dentry, char *buffer, size_t size);
 extern int gfs2_ea_dealloc(struct gfs2_inode *ip);
 
diff --git a/fs/hfs/catalog.c b/fs/hfs/catalog.c
index 6d98f116ca03..424b0337f524 100644
--- a/fs/hfs/catalog.c
+++ b/fs/hfs/catalog.c
@@ -289,6 +289,10 @@ int hfs_cat_move(u32 cnid, struct inode *src_dir, struct qstr *src_name,
 	err = hfs_brec_find(&src_fd);
 	if (err)
 		goto out;
+	if (src_fd.entrylength > sizeof(entry) || src_fd.entrylength < 0) {
+		err = -EIO;
+		goto out;
+	}
 
 	hfs_bnode_read(src_fd.bnode, &entry, src_fd.entryoffset,
 			    src_fd.entrylength);
diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c
index 7c69b98a2e45..2b3b8611b41b 100644
--- a/fs/hfs/dir.c
+++ b/fs/hfs/dir.c
@@ -79,6 +79,11 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		filp->f_pos++;
 		/* fall through */
 	case 1:
+		if (fd.entrylength > sizeof(entry) || fd.entrylength < 0) {
+			err = -EIO;
+			goto out;
+		}
+
 		hfs_bnode_read(fd.bnode, &entry, fd.entryoffset, fd.entrylength);
 		if (entry.type != HFS_CDR_THD) {
 			printk(KERN_ERR "hfs: bad catalog folder thread\n");
@@ -109,6 +114,12 @@ static int hfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 			err = -EIO;
 			goto out;
 		}
+
+		if (fd.entrylength > sizeof(entry) || fd.entrylength < 0) {
+			err = -EIO;
+			goto out;
+		}
+
 		hfs_bnode_read(fd.bnode, &entry, fd.entryoffset, fd.entrylength);
 		type = entry.type;
 		len = hfs_mac2asc(sb, strbuf, &fd.key->cat.CName);
diff --git a/fs/hfs/super.c b/fs/hfs/super.c
index f7fcbe49da72..5ed7252b7b23 100644
--- a/fs/hfs/super.c
+++ b/fs/hfs/super.c
@@ -409,8 +409,13 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent)
 	/* try to get the root inode */
 	hfs_find_init(HFS_SB(sb)->cat_tree, &fd);
 	res = hfs_cat_find_brec(sb, HFS_ROOT_CNID, &fd);
-	if (!res)
+	if (!res) {
+		if (fd.entrylength > sizeof(rec) || fd.entrylength < 0) {
+			res =  -EIO;
+			goto bail;
+		}
 		hfs_bnode_read(fd.bnode, &rec, fd.entryoffset, fd.entrylength);
+	}
 	if (res) {
 		hfs_find_exit(&fd);
 		goto bail_no_root;
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index f2feaa06bf26..cadc4ce48656 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -14,6 +14,7 @@
 #include <linux/magic.h>
 #include <linux/sched.h>
 #include <linux/smp_lock.h>
+#include <linux/bitmap.h>
 
 /* Mark the filesystem dirty, so that chkdsk checks it when os/2 booted */
 
@@ -115,15 +116,13 @@ static void hpfs_put_super(struct super_block *s)
 unsigned hpfs_count_one_bitmap(struct super_block *s, secno secno)
 {
 	struct quad_buffer_head qbh;
-	unsigned *bits;
-	unsigned i, count;
-	if (!(bits = hpfs_map_4sectors(s, secno, &qbh, 4))) return 0;
-	count = 0;
-	for (i = 0; i < 2048 / sizeof(unsigned); i++) {
-		unsigned b; 
-		if (!bits[i]) continue;
-		for (b = bits[i]; b; b>>=1) count += b & 1;
-	}
+	unsigned long *bits;
+	unsigned count;
+
+	bits = hpfs_map_4sectors(s, secno, &qbh, 4);
+	if (!bits)
+		return 0;
+	count = bitmap_weight(bits, 2048 * BITS_PER_BYTE);
 	hpfs_brelse4(&qbh);
 	return count;
 }
diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c
index a5089a6dd67a..7239efc690d8 100644
--- a/fs/hppfs/hppfs.c
+++ b/fs/hppfs/hppfs.c
@@ -646,22 +646,27 @@ static const struct super_operations hppfs_sbops = {
 static int hppfs_readlink(struct dentry *dentry, char __user *buffer,
 			  int buflen)
 {
-	struct dentry *proc_dentry;
-
-	proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+	struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
 	return proc_dentry->d_inode->i_op->readlink(proc_dentry, buffer,
 						    buflen);
 }
 
 static void *hppfs_follow_link(struct dentry *dentry, struct nameidata *nd)
 {
-	struct dentry *proc_dentry;
-
-	proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+	struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
 
 	return proc_dentry->d_inode->i_op->follow_link(proc_dentry, nd);
 }
 
+static void hppfs_put_link(struct dentry *dentry, struct nameidata *nd,
+			   void *cookie)
+{
+	struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+
+	if (proc_dentry->d_inode->i_op->put_link)
+		proc_dentry->d_inode->i_op->put_link(proc_dentry, nd, cookie);
+}
+
 static const struct inode_operations hppfs_dir_iops = {
 	.lookup		= hppfs_lookup,
 };
@@ -669,6 +674,7 @@ static const struct inode_operations hppfs_dir_iops = {
 static const struct inode_operations hppfs_link_iops = {
 	.readlink	= hppfs_readlink,
 	.follow_link	= hppfs_follow_link,
+	.put_link	= hppfs_put_link,
 };
 
 static struct inode *get_inode(struct super_block *sb, struct dentry *dentry)
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 87a1258953b8..a0bbd3d1b41a 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -30,7 +30,6 @@
 #include <linux/dnotify.h>
 #include <linux/statfs.h>
 #include <linux/security.h>
-#include <linux/ima.h>
 #include <linux/magic.h>
 
 #include <asm/uaccess.h>
@@ -922,7 +921,8 @@ struct file *hugetlb_file_setup(const char *name, size_t size, int acctflag,
 	int error = -ENOMEM;
 	struct file *file;
 	struct inode *inode;
-	struct dentry *dentry, *root;
+	struct path path;
+	struct dentry *root;
 	struct qstr quick_string;
 
 	*user = NULL;
@@ -944,10 +944,11 @@ struct file *hugetlb_file_setup(const char *name, size_t size, int acctflag,
 	quick_string.name = name;
 	quick_string.len = strlen(quick_string.name);
 	quick_string.hash = 0;
-	dentry = d_alloc(root, &quick_string);
-	if (!dentry)
+	path.dentry = d_alloc(root, &quick_string);
+	if (!path.dentry)
 		goto out_shm_unlock;
 
+	path.mnt = mntget(hugetlbfs_vfsmount);
 	error = -ENOSPC;
 	inode = hugetlbfs_get_inode(root->d_sb, current_fsuid(),
 				current_fsgid(), S_IFREG | S_IRWXUGO, 0);
@@ -960,24 +961,22 @@ struct file *hugetlb_file_setup(const char *name, size_t size, int acctflag,
 			acctflag))
 		goto out_inode;
 
-	d_instantiate(dentry, inode);
+	d_instantiate(path.dentry, inode);
 	inode->i_size = size;
 	inode->i_nlink = 0;
 
 	error = -ENFILE;
-	file = alloc_file(hugetlbfs_vfsmount, dentry,
-			FMODE_WRITE | FMODE_READ,
+	file = alloc_file(&path, FMODE_WRITE | FMODE_READ,
 			&hugetlbfs_file_operations);
 	if (!file)
 		goto out_dentry; /* inode is already attached */
-	ima_counts_get(file);
 
 	return file;
 
 out_inode:
 	iput(inode);
 out_dentry:
-	dput(dentry);
+	path_put(&path);
 out_shm_unlock:
 	if (*user) {
 		user_shm_unlock(size, *user);
diff --git a/fs/inode.c b/fs/inode.c
index 06c1f02de611..03dfeb2e3928 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -113,7 +113,7 @@ static void wake_up_inode(struct inode *inode)
 	 * Prevent speculative execution through spin_unlock(&inode_lock);
 	 */
 	smp_mb();
-	wake_up_bit(&inode->i_state, __I_LOCK);
+	wake_up_bit(&inode->i_state, __I_NEW);
 }
 
 /**
@@ -690,17 +690,17 @@ void unlock_new_inode(struct inode *inode)
 	}
 #endif
 	/*
-	 * This is special!  We do not need the spinlock when clearing I_LOCK,
+	 * This is special!  We do not need the spinlock when clearing I_NEW,
 	 * because we're guaranteed that nobody else tries to do anything about
 	 * the state of the inode when it is locked, as we just created it (so
-	 * there can be no old holders that haven't tested I_LOCK).
+	 * there can be no old holders that haven't tested I_NEW).
 	 * However we must emit the memory barrier so that other CPUs reliably
-	 * see the clearing of I_LOCK after the other inode initialisation has
+	 * see the clearing of I_NEW after the other inode initialisation has
 	 * completed.
 	 */
 	smp_mb();
-	WARN_ON((inode->i_state & (I_LOCK|I_NEW)) != (I_LOCK|I_NEW));
-	inode->i_state &= ~(I_LOCK|I_NEW);
+	WARN_ON(!(inode->i_state & I_NEW));
+	inode->i_state &= ~I_NEW;
 	wake_up_inode(inode);
 }
 EXPORT_SYMBOL(unlock_new_inode);
@@ -731,7 +731,7 @@ static struct inode *get_new_inode(struct super_block *sb,
 				goto set_failed;
 
 			__inode_add_to_lists(sb, head, inode);
-			inode->i_state = I_LOCK|I_NEW;
+			inode->i_state = I_NEW;
 			spin_unlock(&inode_lock);
 
 			/* Return the locked inode with I_NEW set, the
@@ -778,7 +778,7 @@ static struct inode *get_new_inode_fast(struct super_block *sb,
 		if (!old) {
 			inode->i_ino = ino;
 			__inode_add_to_lists(sb, head, inode);
-			inode->i_state = I_LOCK|I_NEW;
+			inode->i_state = I_NEW;
 			spin_unlock(&inode_lock);
 
 			/* Return the locked inode with I_NEW set, the
@@ -1083,7 +1083,7 @@ int insert_inode_locked(struct inode *inode)
 	ino_t ino = inode->i_ino;
 	struct hlist_head *head = inode_hashtable + hash(sb, ino);
 
-	inode->i_state |= I_LOCK|I_NEW;
+	inode->i_state |= I_NEW;
 	while (1) {
 		struct hlist_node *node;
 		struct inode *old = NULL;
@@ -1120,7 +1120,7 @@ int insert_inode_locked4(struct inode *inode, unsigned long hashval,
 	struct super_block *sb = inode->i_sb;
 	struct hlist_head *head = inode_hashtable + hash(sb, hashval);
 
-	inode->i_state |= I_LOCK|I_NEW;
+	inode->i_state |= I_NEW;
 
 	while (1) {
 		struct hlist_node *node;
@@ -1510,7 +1510,7 @@ EXPORT_SYMBOL(inode_wait);
  * until the deletion _might_ have completed.  Callers are responsible
  * to recheck inode state.
  *
- * It doesn't matter if I_LOCK is not set initially, a call to
+ * It doesn't matter if I_NEW is not set initially, a call to
  * wake_up_inode() after removing from the hash list will DTRT.
  *
  * This is called with inode_lock held.
@@ -1518,8 +1518,8 @@ EXPORT_SYMBOL(inode_wait);
 static void __wait_on_freeing_inode(struct inode *inode)
 {
 	wait_queue_head_t *wq;
-	DEFINE_WAIT_BIT(wait, &inode->i_state, __I_LOCK);
-	wq = bit_waitqueue(&inode->i_state, __I_LOCK);
+	DEFINE_WAIT_BIT(wait, &inode->i_state, __I_NEW);
+	wq = bit_waitqueue(&inode->i_state, __I_NEW);
 	prepare_to_wait(wq, &wait.wait, TASK_UNINTERRUPTIBLE);
 	spin_unlock(&inode_lock);
 	schedule();
diff --git a/fs/internal.h b/fs/internal.h
index 515175b8b72e..e96a1667d749 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -79,8 +79,16 @@ extern void chroot_fs_refs(struct path *, struct path *);
  * file_table.c
  */
 extern void mark_files_ro(struct super_block *);
+extern struct file *get_empty_filp(void);
 
 /*
  * super.c
  */
 extern int do_remount_sb(struct super_block *, int, void *, int);
+
+/*
+ * open.c
+ */
+struct nameidata;
+extern struct file *nameidata_to_filp(struct nameidata *);
+extern void release_open_intent(struct nameidata *);
diff --git a/fs/isofs/compress.c b/fs/isofs/compress.c
index defb932eee9a..0b3fa7974fa8 100644
--- a/fs/isofs/compress.c
+++ b/fs/isofs/compress.c
@@ -36,286 +36,323 @@ static void *zisofs_zlib_workspace;
 static DEFINE_MUTEX(zisofs_zlib_lock);
 
 /*
- * When decompressing, we typically obtain more than one page
- * per reference.  We inject the additional pages into the page
- * cache as a form of readahead.
+ * Read data of @inode from @block_start to @block_end and uncompress
+ * to one zisofs block. Store the data in the @pages array with @pcount
+ * entries. Start storing at offset @poffset of the first page.
  */
-static int zisofs_readpage(struct file *file, struct page *page)
+static loff_t zisofs_uncompress_block(struct inode *inode, loff_t block_start,
+				      loff_t block_end, int pcount,
+				      struct page **pages, unsigned poffset,
+				      int *errp)
 {
-	struct inode *inode = file->f_path.dentry->d_inode;
-	struct address_space *mapping = inode->i_mapping;
-	unsigned int maxpage, xpage, fpage, blockindex;
-	unsigned long offset;
-	unsigned long blockptr, blockendptr, cstart, cend, csize;
-	struct buffer_head *bh, *ptrbh[2];
-	unsigned long bufsize = ISOFS_BUFFER_SIZE(inode);
-	unsigned int bufshift = ISOFS_BUFFER_BITS(inode);
-	unsigned long bufmask  = bufsize - 1;
-	int err = -EIO;
-	int i;
-	unsigned int header_size = ISOFS_I(inode)->i_format_parm[0];
 	unsigned int zisofs_block_shift = ISOFS_I(inode)->i_format_parm[1];
-	/* unsigned long zisofs_block_size = 1UL << zisofs_block_shift; */
-	unsigned int zisofs_block_page_shift = zisofs_block_shift-PAGE_CACHE_SHIFT;
-	unsigned long zisofs_block_pages = 1UL << zisofs_block_page_shift;
-	unsigned long zisofs_block_page_mask = zisofs_block_pages-1;
-	struct page *pages[zisofs_block_pages];
-	unsigned long index = page->index;
-	int indexblocks;
-
-	/* We have already been given one page, this is the one
-	   we must do. */
-	xpage = index & zisofs_block_page_mask;
-	pages[xpage] = page;
- 
-	/* The remaining pages need to be allocated and inserted */
-	offset = index & ~zisofs_block_page_mask;
-	blockindex = offset >> zisofs_block_page_shift;
-	maxpage = (inode->i_size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
-
-	/*
-	 * If this page is wholly outside i_size we just return zero;
-	 * do_generic_file_read() will handle this for us
-	 */
-	if (page->index >= maxpage) {
-		SetPageUptodate(page);
-		unlock_page(page);
+	unsigned int bufsize = ISOFS_BUFFER_SIZE(inode);
+	unsigned int bufshift = ISOFS_BUFFER_BITS(inode);
+	unsigned int bufmask = bufsize - 1;
+	int i, block_size = block_end - block_start;
+	z_stream stream = { .total_out = 0,
+			    .avail_in = 0,
+			    .avail_out = 0, };
+	int zerr;
+	int needblocks = (block_size + (block_start & bufmask) + bufmask)
+				>> bufshift;
+	int haveblocks;
+	blkcnt_t blocknum;
+	struct buffer_head *bhs[needblocks + 1];
+	int curbh, curpage;
+
+	if (block_size > deflateBound(1UL << zisofs_block_shift)) {
+		*errp = -EIO;
 		return 0;
 	}
-
-	maxpage = min(zisofs_block_pages, maxpage-offset);
-
-	for ( i = 0 ; i < maxpage ; i++, offset++ ) {
-		if ( i != xpage ) {
-			pages[i] = grab_cache_page_nowait(mapping, offset);
-		}
-		page = pages[i];
-		if ( page ) {
-			ClearPageError(page);
-			kmap(page);
+	/* Empty block? */
+	if (block_size == 0) {
+		for ( i = 0 ; i < pcount ; i++ ) {
+			if (!pages[i])
+				continue;
+			memset(page_address(pages[i]), 0, PAGE_CACHE_SIZE);
+			flush_dcache_page(pages[i]);
+			SetPageUptodate(pages[i]);
 		}
+		return ((loff_t)pcount) << PAGE_CACHE_SHIFT;
 	}
 
-	/* This is the last page filled, plus one; used in case of abort. */
-	fpage = 0;
+	/* Because zlib is not thread-safe, do all the I/O at the top. */
+	blocknum = block_start >> bufshift;
+	memset(bhs, 0, (needblocks + 1) * sizeof(struct buffer_head *));
+	haveblocks = isofs_get_blocks(inode, blocknum, bhs, needblocks);
+	ll_rw_block(READ, haveblocks, bhs);
 
-	/* Find the pointer to this specific chunk */
-	/* Note: we're not using isonum_731() here because the data is known aligned */
-	/* Note: header_size is in 32-bit words (4 bytes) */
-	blockptr = (header_size + blockindex) << 2;
-	blockendptr = blockptr + 4;
+	curbh = 0;
+	curpage = 0;
+	/*
+	 * First block is special since it may be fractional.  We also wait for
+	 * it before grabbing the zlib mutex; odds are that the subsequent
+	 * blocks are going to come in in short order so we don't hold the zlib
+	 * mutex longer than necessary.
+	 */
 
-	indexblocks = ((blockptr^blockendptr) >> bufshift) ? 2 : 1;
-	ptrbh[0] = ptrbh[1] = NULL;
+	if (!bhs[0])
+		goto b_eio;
 
-	if ( isofs_get_blocks(inode, blockptr >> bufshift, ptrbh, indexblocks) != indexblocks ) {
-		if ( ptrbh[0] ) brelse(ptrbh[0]);
-		printk(KERN_DEBUG "zisofs: Null buffer on reading block table, inode = %lu, block = %lu\n",
-		       inode->i_ino, blockptr >> bufshift);
-		goto eio;
-	}
-	ll_rw_block(READ, indexblocks, ptrbh);
-
-	bh = ptrbh[0];
-	if ( !bh || (wait_on_buffer(bh), !buffer_uptodate(bh)) ) {
-		printk(KERN_DEBUG "zisofs: Failed to read block table, inode = %lu, block = %lu\n",
-		       inode->i_ino, blockptr >> bufshift);
-		if ( ptrbh[1] )
-			brelse(ptrbh[1]);
-		goto eio;
-	}
-	cstart = le32_to_cpu(*(__le32 *)(bh->b_data + (blockptr & bufmask)));
-
-	if ( indexblocks == 2 ) {
-		/* We just crossed a block boundary.  Switch to the next block */
-		brelse(bh);
-		bh = ptrbh[1];
-		if ( !bh || (wait_on_buffer(bh), !buffer_uptodate(bh)) ) {
-			printk(KERN_DEBUG "zisofs: Failed to read block table, inode = %lu, block = %lu\n",
-			       inode->i_ino, blockendptr >> bufshift);
-			goto eio;
-		}
+	wait_on_buffer(bhs[0]);
+	if (!buffer_uptodate(bhs[0])) {
+		*errp = -EIO;
+		goto b_eio;
 	}
-	cend = le32_to_cpu(*(__le32 *)(bh->b_data + (blockendptr & bufmask)));
-	brelse(bh);
 
-	if (cstart > cend)
-		goto eio;
+	stream.workspace = zisofs_zlib_workspace;
+	mutex_lock(&zisofs_zlib_lock);
 		
-	csize = cend-cstart;
-
-	if (csize > deflateBound(1UL << zisofs_block_shift))
-		goto eio;
-
-	/* Now page[] contains an array of pages, any of which can be NULL,
-	   and the locks on which we hold.  We should now read the data and
-	   release the pages.  If the pages are NULL the decompressed data
-	   for that particular page should be discarded. */
-	
-	if ( csize == 0 ) {
-		/* This data block is empty. */
-
-		for ( fpage = 0 ; fpage < maxpage ; fpage++ ) {
-			if ( (page = pages[fpage]) != NULL ) {
-				memset(page_address(page), 0, PAGE_CACHE_SIZE);
-				
-				flush_dcache_page(page);
-				SetPageUptodate(page);
-				kunmap(page);
-				unlock_page(page);
-				if ( fpage == xpage )
-					err = 0; /* The critical page */
-				else
-					page_cache_release(page);
+	zerr = zlib_inflateInit(&stream);
+	if (zerr != Z_OK) {
+		if (zerr == Z_MEM_ERROR)
+			*errp = -ENOMEM;
+		else
+			*errp = -EIO;
+		printk(KERN_DEBUG "zisofs: zisofs_inflateInit returned %d\n",
+			       zerr);
+		goto z_eio;
+	}
+
+	while (curpage < pcount && curbh < haveblocks &&
+	       zerr != Z_STREAM_END) {
+		if (!stream.avail_out) {
+			if (pages[curpage]) {
+				stream.next_out = page_address(pages[curpage])
+						+ poffset;
+				stream.avail_out = PAGE_CACHE_SIZE - poffset;
+				poffset = 0;
+			} else {
+				stream.next_out = (void *)&zisofs_sink_page;
+				stream.avail_out = PAGE_CACHE_SIZE;
 			}
 		}
-	} else {
-		/* This data block is compressed. */
-		z_stream stream;
-		int bail = 0, left_out = -1;
-		int zerr;
-		int needblocks = (csize + (cstart & bufmask) + bufmask) >> bufshift;
-		int haveblocks;
-		struct buffer_head *bhs[needblocks+1];
-		struct buffer_head **bhptr;
-
-		/* Because zlib is not thread-safe, do all the I/O at the top. */
-
-		blockptr = cstart >> bufshift;
-		memset(bhs, 0, (needblocks+1)*sizeof(struct buffer_head *));
-		haveblocks = isofs_get_blocks(inode, blockptr, bhs, needblocks);
-		ll_rw_block(READ, haveblocks, bhs);
-
-		bhptr = &bhs[0];
-		bh = *bhptr++;
-
-		/* First block is special since it may be fractional.
-		   We also wait for it before grabbing the zlib
-		   mutex; odds are that the subsequent blocks are
-		   going to come in in short order so we don't hold
-		   the zlib mutex longer than necessary. */
-
-		if ( !bh || (wait_on_buffer(bh), !buffer_uptodate(bh)) ) {
-			printk(KERN_DEBUG "zisofs: Hit null buffer, fpage = %d, xpage = %d, csize = %ld\n",
-			       fpage, xpage, csize);
-			goto b_eio;
-		}
-		stream.next_in  = bh->b_data + (cstart & bufmask);
-		stream.avail_in = min(bufsize-(cstart & bufmask), csize);
-		csize -= stream.avail_in;
-
-		stream.workspace = zisofs_zlib_workspace;
-		mutex_lock(&zisofs_zlib_lock);
-		
-		zerr = zlib_inflateInit(&stream);
-		if ( zerr != Z_OK ) {
-			if ( err && zerr == Z_MEM_ERROR )
-				err = -ENOMEM;
-			printk(KERN_DEBUG "zisofs: zisofs_inflateInit returned %d\n",
-			       zerr);
-			goto z_eio;
+		if (!stream.avail_in) {
+			wait_on_buffer(bhs[curbh]);
+			if (!buffer_uptodate(bhs[curbh])) {
+				*errp = -EIO;
+				break;
+			}
+			stream.next_in  = bhs[curbh]->b_data +
+						(block_start & bufmask);
+			stream.avail_in = min_t(unsigned, bufsize -
+						(block_start & bufmask),
+						block_size);
+			block_size -= stream.avail_in;
+			block_start = 0;
 		}
 
-		while ( !bail && fpage < maxpage ) {
-			page = pages[fpage];
-			if ( page )
-				stream.next_out = page_address(page);
-			else
-				stream.next_out = (void *)&zisofs_sink_page;
-			stream.avail_out = PAGE_CACHE_SIZE;
-
-			while ( stream.avail_out ) {
-				int ao, ai;
-				if ( stream.avail_in == 0 && left_out ) {
-					if ( !csize ) {
-						printk(KERN_WARNING "zisofs: ZF read beyond end of input\n");
-						bail = 1;
-						break;
-					} else {
-						bh = *bhptr++;
-						if ( !bh ||
-						     (wait_on_buffer(bh), !buffer_uptodate(bh)) ) {
-							/* Reached an EIO */
- 							printk(KERN_DEBUG "zisofs: Hit null buffer, fpage = %d, xpage = %d, csize = %ld\n",
-							       fpage, xpage, csize);
-							       
-							bail = 1;
-							break;
-						}
-						stream.next_in = bh->b_data;
-						stream.avail_in = min(csize,bufsize);
-						csize -= stream.avail_in;
-					}
-				}
-				ao = stream.avail_out;  ai = stream.avail_in;
-				zerr = zlib_inflate(&stream, Z_SYNC_FLUSH);
-				left_out = stream.avail_out;
-				if ( zerr == Z_BUF_ERROR && stream.avail_in == 0 )
-					continue;
-				if ( zerr != Z_OK ) {
-					/* EOF, error, or trying to read beyond end of input */
-					if ( err && zerr == Z_MEM_ERROR )
-						err = -ENOMEM;
-					if ( zerr != Z_STREAM_END )
-						printk(KERN_DEBUG "zisofs: zisofs_inflate returned %d, inode = %lu, index = %lu, fpage = %d, xpage = %d, avail_in = %d, avail_out = %d, ai = %d, ao = %d\n",
-						       zerr, inode->i_ino, index,
-						       fpage, xpage,
-						       stream.avail_in, stream.avail_out,
-						       ai, ao);
-					bail = 1;
-					break;
+		while (stream.avail_out && stream.avail_in) {
+			zerr = zlib_inflate(&stream, Z_SYNC_FLUSH);
+			if (zerr == Z_BUF_ERROR && stream.avail_in == 0)
+				break;
+			if (zerr == Z_STREAM_END)
+				break;
+			if (zerr != Z_OK) {
+				/* EOF, error, or trying to read beyond end of input */
+				if (zerr == Z_MEM_ERROR)
+					*errp = -ENOMEM;
+				else {
+					printk(KERN_DEBUG
+					       "zisofs: zisofs_inflate returned"
+					       " %d, inode = %lu,"
+					       " page idx = %d, bh idx = %d,"
+					       " avail_in = %d,"
+					       " avail_out = %d\n",
+					       zerr, inode->i_ino, curpage,
+					       curbh, stream.avail_in,
+					       stream.avail_out);
+					*errp = -EIO;
 				}
+				goto inflate_out;
 			}
+		}
 
-			if ( stream.avail_out && zerr == Z_STREAM_END ) {
-				/* Fractional page written before EOF.  This may
-				   be the last page in the file. */
-				memset(stream.next_out, 0, stream.avail_out);
-				stream.avail_out = 0;
+		if (!stream.avail_out) {
+			/* This page completed */
+			if (pages[curpage]) {
+				flush_dcache_page(pages[curpage]);
+				SetPageUptodate(pages[curpage]);
 			}
+			curpage++;
+		}
+		if (!stream.avail_in)
+			curbh++;
+	}
+inflate_out:
+	zlib_inflateEnd(&stream);
 
-			if ( !stream.avail_out ) {
-				/* This page completed */
-				if ( page ) {
-					flush_dcache_page(page);
-					SetPageUptodate(page);
-					kunmap(page);
-					unlock_page(page);
-					if ( fpage == xpage )
-						err = 0; /* The critical page */
-					else
-						page_cache_release(page);
-				}
-				fpage++;
-			}
+z_eio:
+	mutex_unlock(&zisofs_zlib_lock);
+
+b_eio:
+	for (i = 0; i < haveblocks; i++)
+		brelse(bhs[i]);
+	return stream.total_out;
+}
+
+/*
+ * Uncompress data so that pages[full_page] is fully uptodate and possibly
+ * fills in other pages if we have data for them.
+ */
+static int zisofs_fill_pages(struct inode *inode, int full_page, int pcount,
+			     struct page **pages)
+{
+	loff_t start_off, end_off;
+	loff_t block_start, block_end;
+	unsigned int header_size = ISOFS_I(inode)->i_format_parm[0];
+	unsigned int zisofs_block_shift = ISOFS_I(inode)->i_format_parm[1];
+	unsigned int blockptr;
+	loff_t poffset = 0;
+	blkcnt_t cstart_block, cend_block;
+	struct buffer_head *bh;
+	unsigned int blkbits = ISOFS_BUFFER_BITS(inode);
+	unsigned int blksize = 1 << blkbits;
+	int err;
+	loff_t ret;
+
+	BUG_ON(!pages[full_page]);
+
+	/*
+	 * We want to read at least 'full_page' page. Because we have to
+	 * uncompress the whole compression block anyway, fill the surrounding
+	 * pages with the data we have anyway...
+	 */
+	start_off = page_offset(pages[full_page]);
+	end_off = min_t(loff_t, start_off + PAGE_CACHE_SIZE, inode->i_size);
+
+	cstart_block = start_off >> zisofs_block_shift;
+	cend_block = (end_off + (1 << zisofs_block_shift) - 1)
+			>> zisofs_block_shift;
+
+	WARN_ON(start_off - (full_page << PAGE_CACHE_SHIFT) !=
+		((cstart_block << zisofs_block_shift) & PAGE_CACHE_MASK));
+
+	/* Find the pointer to this specific chunk */
+	/* Note: we're not using isonum_731() here because the data is known aligned */
+	/* Note: header_size is in 32-bit words (4 bytes) */
+	blockptr = (header_size + cstart_block) << 2;
+	bh = isofs_bread(inode, blockptr >> blkbits);
+	if (!bh)
+		return -EIO;
+	block_start = le32_to_cpu(*(__le32 *)
+				(bh->b_data + (blockptr & (blksize - 1))));
+
+	while (cstart_block < cend_block && pcount > 0) {
+		/* Load end of the compressed block in the file */
+		blockptr += 4;
+		/* Traversed to next block? */
+		if (!(blockptr & (blksize - 1))) {
+			brelse(bh);
+
+			bh = isofs_bread(inode, blockptr >> blkbits);
+			if (!bh)
+				return -EIO;
+		}
+		block_end = le32_to_cpu(*(__le32 *)
+				(bh->b_data + (blockptr & (blksize - 1))));
+		if (block_start > block_end) {
+			brelse(bh);
+			return -EIO;
+		}
+		err = 0;
+		ret = zisofs_uncompress_block(inode, block_start, block_end,
+					      pcount, pages, poffset, &err);
+		poffset += ret;
+		pages += poffset >> PAGE_CACHE_SHIFT;
+		pcount -= poffset >> PAGE_CACHE_SHIFT;
+		full_page -= poffset >> PAGE_CACHE_SHIFT;
+		poffset &= ~PAGE_CACHE_MASK;
+
+		if (err) {
+			brelse(bh);
+			/*
+			 * Did we finish reading the page we really wanted
+			 * to read?
+			 */
+			if (full_page < 0)
+				return 0;
+			return err;
 		}
-		zlib_inflateEnd(&stream);
 
-	z_eio:
-		mutex_unlock(&zisofs_zlib_lock);
+		block_start = block_end;
+		cstart_block++;
+	}
+
+	if (poffset && *pages) {
+		memset(page_address(*pages) + poffset, 0,
+		       PAGE_CACHE_SIZE - poffset);
+		flush_dcache_page(*pages);
+		SetPageUptodate(*pages);
+	}
+	return 0;
+}
 
-	b_eio:
-		for ( i = 0 ; i < haveblocks ; i++ ) {
-			if ( bhs[i] )
-				brelse(bhs[i]);
+/*
+ * When decompressing, we typically obtain more than one page
+ * per reference.  We inject the additional pages into the page
+ * cache as a form of readahead.
+ */
+static int zisofs_readpage(struct file *file, struct page *page)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct address_space *mapping = inode->i_mapping;
+	int err;
+	int i, pcount, full_page;
+	unsigned int zisofs_block_shift = ISOFS_I(inode)->i_format_parm[1];
+	unsigned int zisofs_pages_per_cblock =
+		PAGE_CACHE_SHIFT <= zisofs_block_shift ?
+		(1 << (zisofs_block_shift - PAGE_CACHE_SHIFT)) : 0;
+	struct page *pages[max_t(unsigned, zisofs_pages_per_cblock, 1)];
+	pgoff_t index = page->index, end_index;
+
+	end_index = (inode->i_size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+	/*
+	 * If this page is wholly outside i_size we just return zero;
+	 * do_generic_file_read() will handle this for us
+	 */
+	if (index >= end_index) {
+		SetPageUptodate(page);
+		unlock_page(page);
+		return 0;
+	}
+
+	if (PAGE_CACHE_SHIFT <= zisofs_block_shift) {
+		/* We have already been given one page, this is the one
+		   we must do. */
+		full_page = index & (zisofs_pages_per_cblock - 1);
+		pcount = min_t(int, zisofs_pages_per_cblock,
+			end_index - (index & ~(zisofs_pages_per_cblock - 1)));
+		index -= full_page;
+	} else {
+		full_page = 0;
+		pcount = 1;
+	}
+	pages[full_page] = page;
+
+	for (i = 0; i < pcount; i++, index++) {
+		if (i != full_page)
+			pages[i] = grab_cache_page_nowait(mapping, index);
+		if (pages[i]) {
+			ClearPageError(pages[i]);
+			kmap(pages[i]);
 		}
 	}
 
-eio:
+	err = zisofs_fill_pages(inode, full_page, pcount, pages);
 
 	/* Release any residual pages, do not SetPageUptodate */
-	while ( fpage < maxpage ) {
-		page = pages[fpage];
-		if ( page ) {
-			flush_dcache_page(page);
-			if ( fpage == xpage )
-				SetPageError(page);
-			kunmap(page);
-			unlock_page(page);
-			if ( fpage != xpage )
-				page_cache_release(page);
+	for (i = 0; i < pcount; i++) {
+		if (pages[i]) {
+			flush_dcache_page(pages[i]);
+			if (i == full_page && err)
+				SetPageError(pages[i]);
+			kunmap(pages[i]);
+			unlock_page(pages[i]);
+			if (i != full_page)
+				page_cache_release(pages[i]);
 		}
-		fpage++;
 	}			
 
 	/* At this point, err contains 0 or -EIO depending on the "critical" page */
diff --git a/fs/isofs/export.c b/fs/isofs/export.c
index e81a30593ba9..ed752cb38474 100644
--- a/fs/isofs/export.c
+++ b/fs/isofs/export.c
@@ -9,7 +9,7 @@
  *
  * The following files are helpful:
  *
- *     Documentation/filesystems/Exporting
+ *     Documentation/filesystems/nfs/Exporting
  *     fs/exportfs/expfs.c.
  */
 
diff --git a/fs/isofs/rock.c b/fs/isofs/rock.c
index c2fb2dd0131f..96a685c550fd 100644
--- a/fs/isofs/rock.c
+++ b/fs/isofs/rock.c
@@ -518,8 +518,7 @@ repeat:
 			if (algo == SIG('p', 'z')) {
 				int block_shift =
 					isonum_711(&rr->u.ZF.parms[1]);
-				if (block_shift < PAGE_CACHE_SHIFT
-						|| block_shift > 17) {
+				if (block_shift > 17) {
 					printk(KERN_WARNING "isofs: "
 						"Can't handle ZF block "
 						"size of 2^%d\n",
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 4160afad6d00..bd224eec9b07 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -1913,7 +1913,7 @@ static void __init jbd_create_debugfs_entry(void)
 {
 	jbd_debugfs_dir = debugfs_create_dir("jbd", NULL);
 	if (jbd_debugfs_dir)
-		jbd_debug = debugfs_create_u8("jbd-debug", S_IRUGO,
+		jbd_debug = debugfs_create_u8("jbd-debug", S_IRUGO | S_IWUSR,
 					       jbd_debugfs_dir,
 					       &journal_enable_debug);
 }
diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index ca0f5eb62b20..886849370950 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -22,6 +22,7 @@
 #include <linux/jbd2.h>
 #include <linux/errno.h>
 #include <linux/slab.h>
+#include <linux/blkdev.h>
 #include <trace/events/jbd2.h>
 
 /*
@@ -515,6 +516,20 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	journal->j_tail_sequence = first_tid;
 	journal->j_tail = blocknr;
 	spin_unlock(&journal->j_state_lock);
+
+	/*
+	 * If there is an external journal, we need to make sure that
+	 * any data blocks that were recently written out --- perhaps
+	 * by jbd2_log_do_checkpoint() --- are flushed out before we
+	 * drop the transactions from the external journal.  It's
+	 * unlikely this will be necessary, especially with a
+	 * appropriately sized journal, but we need this to guarantee
+	 * correctness.  Fortunately jbd2_cleanup_journal_tail()
+	 * doesn't get called all that often.
+	 */
+	if ((journal->j_fs_dev != journal->j_dev) &&
+	    (journal->j_flags & JBD2_BARRIER))
+		blkdev_issue_flush(journal->j_fs_dev, NULL);
 	if (!(journal->j_flags & JBD2_ABORT))
 		jbd2_journal_update_superblock(journal, 1);
 	return 0;
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index d4cfd6d2779e..1bc74b6f26d2 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -259,6 +259,7 @@ static int journal_submit_data_buffers(journal_t *journal,
 			ret = err;
 		spin_lock(&journal->j_list_lock);
 		J_ASSERT(jinode->i_transaction == commit_transaction);
+		commit_transaction->t_flushed_data_blocks = 1;
 		jinode->i_flags &= ~JI_COMMIT_RUNNING;
 		wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING);
 	}
@@ -286,7 +287,7 @@ static int journal_finish_inode_data_buffers(journal_t *journal,
 		if (err) {
 			/*
 			 * Because AS_EIO is cleared by
-			 * wait_on_page_writeback_range(), set it again so
+			 * filemap_fdatawait_range(), set it again so
 			 * that user process can get -EIO from fsync().
 			 */
 			set_bit(AS_EIO,
@@ -636,6 +637,10 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		JBUFFER_TRACE(jh, "ph3: write metadata");
 		flags = jbd2_journal_write_metadata_buffer(commit_transaction,
 						      jh, &new_jh, blocknr);
+		if (flags < 0) {
+			jbd2_journal_abort(journal, flags);
+			continue;
+		}
 		set_bit(BH_JWrite, &jh2bh(new_jh)->b_state);
 		wbuf[bufs++] = jh2bh(new_jh);
 
@@ -704,8 +709,17 @@ start_journal_io:
 		}
 	}
 
-	/* Done it all: now write the commit record asynchronously. */
+	/* 
+	 * If the journal is not located on the file system device,
+	 * then we must flush the file system device before we issue
+	 * the commit record
+	 */
+	if (commit_transaction->t_flushed_data_blocks &&
+	    (journal->j_fs_dev != journal->j_dev) &&
+	    (journal->j_flags & JBD2_BARRIER))
+		blkdev_issue_flush(journal->j_fs_dev, NULL);
 
+	/* Done it all: now write the commit record asynchronously. */
 	if (JBD2_HAS_INCOMPAT_FEATURE(journal,
 				      JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) {
 		err = journal_submit_commit_record(journal, commit_transaction,
@@ -716,13 +730,6 @@ start_journal_io:
 			blkdev_issue_flush(journal->j_dev, NULL);
 	}
 
-	/*
-	 * This is the right place to wait for data buffers both for ASYNC
-	 * and !ASYNC commit. If commit is ASYNC, we need to wait only after
-	 * the commit block went to disk (which happens above). If commit is
-	 * SYNC, we need to wait for data buffers before we start writing
-	 * commit block, which happens below in such setting.
-	 */
 	err = journal_finish_inode_data_buffers(journal, commit_transaction);
 	if (err) {
 		printk(KERN_WARNING
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index fed85388ee86..ac0d027595d0 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -78,6 +78,7 @@ EXPORT_SYMBOL(jbd2_journal_errno);
 EXPORT_SYMBOL(jbd2_journal_ack_err);
 EXPORT_SYMBOL(jbd2_journal_clear_err);
 EXPORT_SYMBOL(jbd2_log_wait_commit);
+EXPORT_SYMBOL(jbd2_log_start_commit);
 EXPORT_SYMBOL(jbd2_journal_start_commit);
 EXPORT_SYMBOL(jbd2_journal_force_commit_nested);
 EXPORT_SYMBOL(jbd2_journal_wipe);
@@ -358,6 +359,10 @@ repeat:
 
 		jbd_unlock_bh_state(bh_in);
 		tmp = jbd2_alloc(bh_in->b_size, GFP_NOFS);
+		if (!tmp) {
+			jbd2_journal_put_journal_head(new_jh);
+			return -ENOMEM;
+		}
 		jbd_lock_bh_state(bh_in);
 		if (jh_in->b_frozen_data) {
 			jbd2_free(tmp, bh_in->b_size);
@@ -809,7 +814,7 @@ static journal_t * journal_init_common (void)
 	journal_t *journal;
 	int err;
 
-	journal = kzalloc(sizeof(*journal), GFP_KERNEL|__GFP_NOFAIL);
+	journal = kzalloc(sizeof(*journal), GFP_KERNEL);
 	if (!journal)
 		goto fail;
 
@@ -1248,6 +1253,13 @@ int jbd2_journal_load(journal_t *journal)
 	if (jbd2_journal_recover(journal))
 		goto recovery_error;
 
+	if (journal->j_failed_commit) {
+		printk(KERN_ERR "JBD2: journal transaction %u on %s "
+		       "is corrupt.\n", journal->j_failed_commit,
+		       journal->j_devname);
+		return -EIO;
+	}
+
 	/* OK, we've finished with the dynamic journal bits:
 	 * reinitialise the dynamic contents of the superblock in memory
 	 * and reset them on disk. */
@@ -2103,7 +2115,8 @@ static void __init jbd2_create_debugfs_entry(void)
 {
 	jbd2_debugfs_dir = debugfs_create_dir("jbd2", NULL);
 	if (jbd2_debugfs_dir)
-		jbd2_debug = debugfs_create_u8(JBD2_DEBUG_NAME, S_IRUGO,
+		jbd2_debug = debugfs_create_u8(JBD2_DEBUG_NAME,
+					       S_IRUGO | S_IWUSR,
 					       jbd2_debugfs_dir,
 					       &jbd2_journal_enable_debug);
 }
diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c
index 7edb62e97419..7cdc3196476a 100644
--- a/fs/jffs2/acl.c
+++ b/fs/jffs2/acl.c
@@ -350,8 +350,8 @@ int jffs2_acl_chmod(struct inode *inode)
 	return rc;
 }
 
-static size_t jffs2_acl_access_listxattr(struct inode *inode, char *list, size_t list_size,
-					 const char *name, size_t name_len)
+static size_t jffs2_acl_access_listxattr(struct dentry *dentry, char *list,
+		size_t list_size, const char *name, size_t name_len, int type)
 {
 	const int retlen = sizeof(POSIX_ACL_XATTR_ACCESS);
 
@@ -360,8 +360,8 @@ static size_t jffs2_acl_access_listxattr(struct inode *inode, char *list, size_t
 	return retlen;
 }
 
-static size_t jffs2_acl_default_listxattr(struct inode *inode, char *list, size_t list_size,
-					  const char *name, size_t name_len)
+static size_t jffs2_acl_default_listxattr(struct dentry *dentry, char *list,
+		size_t list_size, const char *name, size_t name_len, int type)
 {
 	const int retlen = sizeof(POSIX_ACL_XATTR_DEFAULT);
 
@@ -370,12 +370,16 @@ static size_t jffs2_acl_default_listxattr(struct inode *inode, char *list, size_
 	return retlen;
 }
 
-static int jffs2_acl_getxattr(struct inode *inode, int type, void *buffer, size_t size)
+static int jffs2_acl_getxattr(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
 	struct posix_acl *acl;
 	int rc;
 
-	acl = jffs2_get_acl(inode, type);
+	if (name[0] != '\0')
+		return -EINVAL;
+
+	acl = jffs2_get_acl(dentry->d_inode, type);
 	if (IS_ERR(acl))
 		return PTR_ERR(acl);
 	if (!acl)
@@ -386,26 +390,15 @@ static int jffs2_acl_getxattr(struct inode *inode, int type, void *buffer, size_
 	return rc;
 }
 
-static int jffs2_acl_access_getxattr(struct inode *inode, const char *name, void *buffer, size_t size)
-{
-	if (name[0] != '\0')
-		return -EINVAL;
-	return jffs2_acl_getxattr(inode, ACL_TYPE_ACCESS, buffer, size);
-}
-
-static int jffs2_acl_default_getxattr(struct inode *inode, const char *name, void *buffer, size_t size)
-{
-	if (name[0] != '\0')
-		return -EINVAL;
-	return jffs2_acl_getxattr(inode, ACL_TYPE_DEFAULT, buffer, size);
-}
-
-static int jffs2_acl_setxattr(struct inode *inode, int type, const void *value, size_t size)
+static int jffs2_acl_setxattr(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	struct posix_acl *acl;
 	int rc;
 
-	if (!is_owner_or_cap(inode))
+	if (name[0] != '\0')
+		return -EINVAL;
+	if (!is_owner_or_cap(dentry->d_inode))
 		return -EPERM;
 
 	if (value) {
@@ -420,38 +413,24 @@ static int jffs2_acl_setxattr(struct inode *inode, int type, const void *value,
 	} else {
 		acl = NULL;
 	}
-	rc = jffs2_set_acl(inode, type, acl);
+	rc = jffs2_set_acl(dentry->d_inode, type, acl);
  out:
 	posix_acl_release(acl);
 	return rc;
 }
 
-static int jffs2_acl_access_setxattr(struct inode *inode, const char *name,
-				     const void *buffer, size_t size, int flags)
-{
-	if (name[0] != '\0')
-		return -EINVAL;
-	return jffs2_acl_setxattr(inode, ACL_TYPE_ACCESS, buffer, size);
-}
-
-static int jffs2_acl_default_setxattr(struct inode *inode, const char *name,
-				      const void *buffer, size_t size, int flags)
-{
-	if (name[0] != '\0')
-		return -EINVAL;
-	return jffs2_acl_setxattr(inode, ACL_TYPE_DEFAULT, buffer, size);
-}
-
 struct xattr_handler jffs2_acl_access_xattr_handler = {
 	.prefix	= POSIX_ACL_XATTR_ACCESS,
+	.flags	= ACL_TYPE_DEFAULT,
 	.list	= jffs2_acl_access_listxattr,
-	.get	= jffs2_acl_access_getxattr,
-	.set	= jffs2_acl_access_setxattr,
+	.get	= jffs2_acl_getxattr,
+	.set	= jffs2_acl_setxattr,
 };
 
 struct xattr_handler jffs2_acl_default_xattr_handler = {
 	.prefix	= POSIX_ACL_XATTR_DEFAULT,
+	.flags	= ACL_TYPE_DEFAULT,
 	.list	= jffs2_acl_default_listxattr,
-	.get	= jffs2_acl_default_getxattr,
-	.set	= jffs2_acl_default_setxattr,
+	.get	= jffs2_acl_getxattr,
+	.set	= jffs2_acl_setxattr,
 };
diff --git a/fs/jffs2/gc.c b/fs/jffs2/gc.c
index 090c556ffed2..3b6f2fa12cff 100644
--- a/fs/jffs2/gc.c
+++ b/fs/jffs2/gc.c
@@ -700,7 +700,8 @@ static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_
 	struct jffs2_raw_inode ri;
 	struct jffs2_node_frag *last_frag;
 	union jffs2_device_node dev;
-	char *mdata = NULL, mdatalen = 0;
+	char *mdata = NULL;
+	int mdatalen = 0;
 	uint32_t alloclen, ilen;
 	int ret;
 
diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c
index 378991cfe40f..e22de8397b74 100644
--- a/fs/jffs2/readinode.c
+++ b/fs/jffs2/readinode.c
@@ -1284,7 +1284,7 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c,
 				f->target = NULL;
 				mutex_unlock(&f->sem);
 				jffs2_do_clear_inode(c, f);
-				return -ret;
+				return ret;
 			}
 
 			f->target[je32_to_cpu(latest_node->csize)] = '\0';
diff --git a/fs/jffs2/security.c b/fs/jffs2/security.c
index 02c39c64ecb3..eaccee058583 100644
--- a/fs/jffs2/security.c
+++ b/fs/jffs2/security.c
@@ -44,26 +44,28 @@ int jffs2_init_security(struct inode *inode, struct inode *dir)
 }
 
 /* ---- XATTR Handler for "security.*" ----------------- */
-static int jffs2_security_getxattr(struct inode *inode, const char *name,
-				   void *buffer, size_t size)
+static int jffs2_security_getxattr(struct dentry *dentry, const char *name,
+				   void *buffer, size_t size, int type)
 {
 	if (!strcmp(name, ""))
 		return -EINVAL;
 
-	return do_jffs2_getxattr(inode, JFFS2_XPREFIX_SECURITY, name, buffer, size);
+	return do_jffs2_getxattr(dentry->d_inode, JFFS2_XPREFIX_SECURITY,
+				 name, buffer, size);
 }
 
-static int jffs2_security_setxattr(struct inode *inode, const char *name, const void *buffer,
-				   size_t size, int flags)
+static int jffs2_security_setxattr(struct dentry *dentry, const char *name,
+		const void *buffer, size_t size, int flags, int type)
 {
 	if (!strcmp(name, ""))
 		return -EINVAL;
 
-	return do_jffs2_setxattr(inode, JFFS2_XPREFIX_SECURITY, name, buffer, size, flags);
+	return do_jffs2_setxattr(dentry->d_inode, JFFS2_XPREFIX_SECURITY,
+				 name, buffer, size, flags);
 }
 
-static size_t jffs2_security_listxattr(struct inode *inode, char *list, size_t list_size,
-				       const char *name, size_t name_len)
+static size_t jffs2_security_listxattr(struct dentry *dentry, char *list,
+		size_t list_size, const char *name, size_t name_len, int type)
 {
 	size_t retlen = XATTR_SECURITY_PREFIX_LEN + name_len + 1;
 
diff --git a/fs/jffs2/summary.c b/fs/jffs2/summary.c
index 6caf1e1ee26d..800171dca53b 100644
--- a/fs/jffs2/summary.c
+++ b/fs/jffs2/summary.c
@@ -23,7 +23,7 @@
 
 int jffs2_sum_init(struct jffs2_sb_info *c)
 {
-	uint32_t sum_size = max_t(uint32_t, c->sector_size, MAX_SUMMARY_SIZE);
+	uint32_t sum_size = min_t(uint32_t, c->sector_size, MAX_SUMMARY_SIZE);
 
 	c->summary = kzalloc(sizeof(struct jffs2_summary), GFP_KERNEL);
 
diff --git a/fs/jffs2/xattr.c b/fs/jffs2/xattr.c
index 4b107881acd5..9e75c62c85d6 100644
--- a/fs/jffs2/xattr.c
+++ b/fs/jffs2/xattr.c
@@ -990,9 +990,11 @@ ssize_t jffs2_listxattr(struct dentry *dentry, char *buffer, size_t size)
 		if (!xhandle)
 			continue;
 		if (buffer) {
-			rc = xhandle->list(inode, buffer+len, size-len, xd->xname, xd->name_len);
+			rc = xhandle->list(dentry, buffer+len, size-len,
+					   xd->xname, xd->name_len, xd->flags);
 		} else {
-			rc = xhandle->list(inode, NULL, 0, xd->xname, xd->name_len);
+			rc = xhandle->list(dentry, NULL, 0, xd->xname,
+					   xd->name_len, xd->flags);
 		}
 		if (rc < 0)
 			goto out;
diff --git a/fs/jffs2/xattr_trusted.c b/fs/jffs2/xattr_trusted.c
index 8ec5765ef348..3e5a5e356e05 100644
--- a/fs/jffs2/xattr_trusted.c
+++ b/fs/jffs2/xattr_trusted.c
@@ -16,24 +16,26 @@
 #include <linux/mtd/mtd.h>
 #include "nodelist.h"
 
-static int jffs2_trusted_getxattr(struct inode *inode, const char *name,
-				  void *buffer, size_t size)
+static int jffs2_trusted_getxattr(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
 	if (!strcmp(name, ""))
 		return -EINVAL;
-	return do_jffs2_getxattr(inode, JFFS2_XPREFIX_TRUSTED, name, buffer, size);
+	return do_jffs2_getxattr(dentry->d_inode, JFFS2_XPREFIX_TRUSTED,
+				 name, buffer, size);
 }
 
-static int jffs2_trusted_setxattr(struct inode *inode, const char *name, const void *buffer,
-				  size_t size, int flags)
+static int jffs2_trusted_setxattr(struct dentry *dentry, const char *name,
+		const void *buffer, size_t size, int flags, int type)
 {
 	if (!strcmp(name, ""))
 		return -EINVAL;
-	return do_jffs2_setxattr(inode, JFFS2_XPREFIX_TRUSTED, name, buffer, size, flags);
+	return do_jffs2_setxattr(dentry->d_inode, JFFS2_XPREFIX_TRUSTED,
+				 name, buffer, size, flags);
 }
 
-static size_t jffs2_trusted_listxattr(struct inode *inode, char *list, size_t list_size,
-				      const char *name, size_t name_len)
+static size_t jffs2_trusted_listxattr(struct dentry *dentry, char *list,
+		size_t list_size, const char *name, size_t name_len, int type)
 {
 	size_t retlen = XATTR_TRUSTED_PREFIX_LEN + name_len + 1;
 
diff --git a/fs/jffs2/xattr_user.c b/fs/jffs2/xattr_user.c
index 8bbeab90ada1..8544af67dffe 100644
--- a/fs/jffs2/xattr_user.c
+++ b/fs/jffs2/xattr_user.c
@@ -16,24 +16,26 @@
 #include <linux/mtd/mtd.h>
 #include "nodelist.h"
 
-static int jffs2_user_getxattr(struct inode *inode, const char *name,
-			       void *buffer, size_t size)
+static int jffs2_user_getxattr(struct dentry *dentry, const char *name,
+			       void *buffer, size_t size, int type)
 {
 	if (!strcmp(name, ""))
 		return -EINVAL;
-	return do_jffs2_getxattr(inode, JFFS2_XPREFIX_USER, name, buffer, size);
+	return do_jffs2_getxattr(dentry->d_inode, JFFS2_XPREFIX_USER,
+				 name, buffer, size);
 }
 
-static int jffs2_user_setxattr(struct inode *inode, const char *name, const void *buffer,
-			       size_t size, int flags)
+static int jffs2_user_setxattr(struct dentry *dentry, const char *name,
+		const void *buffer, size_t size, int flags, int type)
 {
 	if (!strcmp(name, ""))
 		return -EINVAL;
-	return do_jffs2_setxattr(inode, JFFS2_XPREFIX_USER, name, buffer, size, flags);
+	return do_jffs2_setxattr(dentry->d_inode, JFFS2_XPREFIX_USER,
+				 name, buffer, size, flags);
 }
 
-static size_t jffs2_user_listxattr(struct inode *inode, char *list, size_t list_size,
-				   const char *name, size_t name_len)
+static size_t jffs2_user_listxattr(struct dentry *dentry, char *list,
+		size_t list_size, const char *name, size_t name_len, int type)
 {
 	size_t retlen = XATTR_USER_PREFIX_LEN + name_len + 1;
 
diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c
index f26e4d03ada5..d945ea76b445 100644
--- a/fs/jfs/jfs_txnmgr.c
+++ b/fs/jfs/jfs_txnmgr.c
@@ -1292,7 +1292,7 @@ int txCommit(tid_t tid,		/* transaction identifier */
 		 */
 		/*
 		 * I believe this code is no longer needed.  Splitting I_LOCK
-		 * into two bits, I_LOCK and I_SYNC should prevent this
+		 * into two bits, I_NEW and I_SYNC should prevent this
 		 * deadlock as well.  But since I don't have a JFS testload
 		 * to verify this, only a trivial s/I_LOCK/I_SYNC/ was done.
 		 * Joern
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 2234c73fc577..d929a822a74e 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -524,7 +524,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
 	 * Page cache is indexed by long.
 	 * I would use MAX_LFS_FILESIZE, but it's only half as big
 	 */
-	sb->s_maxbytes = min(((u64) PAGE_CACHE_SIZE << 32) - 1, sb->s_maxbytes);
+	sb->s_maxbytes = min(((u64) PAGE_CACHE_SIZE << 32) - 1, (u64)sb->s_maxbytes);
 #endif
 	sb->s_time_gran = 1;
 	return 0;
diff --git a/fs/libfs.c b/fs/libfs.c
index 219576c52d80..6e8d17e1dc4c 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -848,7 +848,6 @@ EXPORT_SYMBOL(simple_write_end);
 EXPORT_SYMBOL(simple_dir_inode_operations);
 EXPORT_SYMBOL(simple_dir_operations);
 EXPORT_SYMBOL(simple_empty);
-EXPORT_SYMBOL(d_alloc_name);
 EXPORT_SYMBOL(simple_fill_super);
 EXPORT_SYMBOL(simple_getattr);
 EXPORT_SYMBOL(simple_link);
diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c
index bd173a6ca3b1..a7966eed3c17 100644
--- a/fs/lockd/svc4proc.c
+++ b/fs/lockd/svc4proc.c
@@ -11,10 +11,6 @@
 #include <linux/time.h>
 #include <linux/slab.h>
 #include <linux/smp_lock.h>
-#include <linux/in.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/sunrpc/clnt.h>
-#include <linux/nfsd/nfsd.h>
 #include <linux/lockd/lockd.h>
 #include <linux/lockd/share.h>
 
diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c
index e1d28ddd2169..56c9519d900a 100644
--- a/fs/lockd/svcproc.c
+++ b/fs/lockd/svcproc.c
@@ -11,10 +11,6 @@
 #include <linux/time.h>
 #include <linux/slab.h>
 #include <linux/smp_lock.h>
-#include <linux/in.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/sunrpc/clnt.h>
-#include <linux/nfsd/nfsd.h>
 #include <linux/lockd/lockd.h>
 #include <linux/lockd/share.h>
 
diff --git a/fs/namei.c b/fs/namei.c
index d11f404667e9..94a5e60779f9 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -35,7 +35,7 @@
 #include <linux/fs_struct.h>
 #include <asm/uaccess.h>
 
-#define ACC_MODE(x) ("\000\004\002\006"[(x)&O_ACCMODE])
+#include "internal.h"
 
 /* [Feb-1997 T. Schoebel-Theuer]
  * Fundamental changes in the pathname lookup mechanisms (namei)
@@ -108,8 +108,6 @@
  * any extra contention...
  */
 
-static int __link_path_walk(const char *name, struct nameidata *nd);
-
 /* In order to reduce some races, while at the same time doing additional
  * checking and hopefully speeding things up, we copy filenames to the
  * kernel data space before using them..
@@ -234,6 +232,7 @@ int generic_permission(struct inode *inode, int mask,
 	/*
 	 * Searching includes executable on directories, else just read.
 	 */
+	mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
 	if (mask == MAY_READ || (S_ISDIR(inode->i_mode) && !(mask & MAY_WRITE)))
 		if (capable(CAP_DAC_READ_SEARCH))
 			return 0;
@@ -414,36 +413,55 @@ do_revalidate(struct dentry *dentry, struct nameidata *nd)
 }
 
 /*
- * Internal lookup() using the new generic dcache.
- * SMP-safe
+ * force_reval_path - force revalidation of a dentry
+ *
+ * In some situations the path walking code will trust dentries without
+ * revalidating them. This causes problems for filesystems that depend on
+ * d_revalidate to handle file opens (e.g. NFSv4). When FS_REVAL_DOT is set
+ * (which indicates that it's possible for the dentry to go stale), force
+ * a d_revalidate call before proceeding.
+ *
+ * Returns 0 if the revalidation was successful. If the revalidation fails,
+ * either return the error returned by d_revalidate or -ESTALE if the
+ * revalidation it just returned 0. If d_revalidate returns 0, we attempt to
+ * invalidate the dentry. It's up to the caller to handle putting references
+ * to the path if necessary.
  */
-static struct dentry * cached_lookup(struct dentry * parent, struct qstr * name, struct nameidata *nd)
+static int
+force_reval_path(struct path *path, struct nameidata *nd)
 {
-	struct dentry * dentry = __d_lookup(parent, name);
+	int status;
+	struct dentry *dentry = path->dentry;
 
-	/* lockess __d_lookup may fail due to concurrent d_move() 
-	 * in some unrelated directory, so try with d_lookup
+	/*
+	 * only check on filesystems where it's possible for the dentry to
+	 * become stale. It's assumed that if this flag is set then the
+	 * d_revalidate op will also be defined.
 	 */
-	if (!dentry)
-		dentry = d_lookup(parent, name);
+	if (!(dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT))
+		return 0;
 
-	if (dentry && dentry->d_op && dentry->d_op->d_revalidate)
-		dentry = do_revalidate(dentry, nd);
+	status = dentry->d_op->d_revalidate(dentry, nd);
+	if (status > 0)
+		return 0;
 
-	return dentry;
+	if (!status) {
+		d_invalidate(dentry);
+		status = -ESTALE;
+	}
+	return status;
 }
 
 /*
- * Short-cut version of permission(), for calling by
- * path_walk(), when dcache lock is held.  Combines parts
- * of permission() and generic_permission(), and tests ONLY for
- * MAY_EXEC permission.
+ * Short-cut version of permission(), for calling on directories
+ * during pathname resolution.  Combines parts of permission()
+ * and generic_permission(), and tests ONLY for MAY_EXEC permission.
  *
  * If appropriate, check DAC only.  If not appropriate, or
- * short-cut DAC fails, then call permission() to do more
+ * short-cut DAC fails, then call ->permission() to do more
  * complete permission check.
  */
-static int exec_permission_lite(struct inode *inode)
+static int exec_permission(struct inode *inode)
 {
 	int ret;
 
@@ -465,99 +483,6 @@ ok:
 	return security_inode_permission(inode, MAY_EXEC);
 }
 
-/*
- * This is called when everything else fails, and we actually have
- * to go to the low-level filesystem to find out what we should do..
- *
- * We get the directory semaphore, and after getting that we also
- * make sure that nobody added the entry to the dcache in the meantime..
- * SMP-safe
- */
-static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, struct nameidata *nd)
-{
-	struct dentry * result;
-	struct inode *dir = parent->d_inode;
-
-	mutex_lock(&dir->i_mutex);
-	/*
-	 * First re-do the cached lookup just in case it was created
-	 * while we waited for the directory semaphore..
-	 *
-	 * FIXME! This could use version numbering or similar to
-	 * avoid unnecessary cache lookups.
-	 *
-	 * The "dcache_lock" is purely to protect the RCU list walker
-	 * from concurrent renames at this point (we mustn't get false
-	 * negatives from the RCU list walk here, unlike the optimistic
-	 * fast walk).
-	 *
-	 * so doing d_lookup() (with seqlock), instead of lockfree __d_lookup
-	 */
-	result = d_lookup(parent, name);
-	if (!result) {
-		struct dentry *dentry;
-
-		/* Don't create child dentry for a dead directory. */
-		result = ERR_PTR(-ENOENT);
-		if (IS_DEADDIR(dir))
-			goto out_unlock;
-
-		dentry = d_alloc(parent, name);
-		result = ERR_PTR(-ENOMEM);
-		if (dentry) {
-			result = dir->i_op->lookup(dir, dentry, nd);
-			if (result)
-				dput(dentry);
-			else
-				result = dentry;
-		}
-out_unlock:
-		mutex_unlock(&dir->i_mutex);
-		return result;
-	}
-
-	/*
-	 * Uhhuh! Nasty case: the cache was re-populated while
-	 * we waited on the semaphore. Need to revalidate.
-	 */
-	mutex_unlock(&dir->i_mutex);
-	if (result->d_op && result->d_op->d_revalidate) {
-		result = do_revalidate(result, nd);
-		if (!result)
-			result = ERR_PTR(-ENOENT);
-	}
-	return result;
-}
-
-/*
- * Wrapper to retry pathname resolution whenever the underlying
- * file system returns an ESTALE.
- *
- * Retry the whole path once, forcing real lookup requests
- * instead of relying on the dcache.
- */
-static __always_inline int link_path_walk(const char *name, struct nameidata *nd)
-{
-	struct path save = nd->path;
-	int result;
-
-	/* make sure the stuff we saved doesn't go away */
-	path_get(&save);
-
-	result = __link_path_walk(name, nd);
-	if (result == -ESTALE) {
-		/* nd->path had been dropped */
-		nd->path = save;
-		path_get(&nd->path);
-		nd->flags |= LOOKUP_REVAL;
-		result = __link_path_walk(name, nd);
-	}
-
-	path_put(&save);
-
-	return result;
-}
-
 static __always_inline void set_root(struct nameidata *nd)
 {
 	if (!nd->root.mnt) {
@@ -569,6 +494,8 @@ static __always_inline void set_root(struct nameidata *nd)
 	}
 }
 
+static int link_path_walk(const char *, struct nameidata *);
+
 static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *link)
 {
 	int res = 0;
@@ -634,6 +561,7 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata
 		dget(dentry);
 	}
 	mntget(path->mnt);
+	nd->last_type = LAST_BIND;
 	cookie = dentry->d_inode->i_op->follow_link(dentry, nd);
 	error = PTR_ERR(cookie);
 	if (!IS_ERR(cookie)) {
@@ -641,11 +569,14 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata
 		error = 0;
 		if (s)
 			error = __vfs_follow_link(nd, s);
+		else if (nd->last_type == LAST_BIND) {
+			error = force_reval_path(&nd->path, nd);
+			if (error)
+				path_put(&nd->path);
+		}
 		if (dentry->d_inode->i_op->put_link)
 			dentry->d_inode->i_op->put_link(dentry, nd, cookie);
 	}
-	path_put(path);
-
 	return error;
 }
 
@@ -672,6 +603,7 @@ static inline int do_follow_link(struct path *path, struct nameidata *nd)
 	current->total_link_count++;
 	nd->depth++;
 	err = __do_follow_link(path, nd);
+	path_put(path);
 	current->link_count--;
 	nd->depth--;
 	return err;
@@ -797,8 +729,19 @@ static int do_lookup(struct nameidata *nd, struct qstr *name,
 		     struct path *path)
 {
 	struct vfsmount *mnt = nd->path.mnt;
-	struct dentry *dentry = __d_lookup(nd->path.dentry, name);
+	struct dentry *dentry, *parent;
+	struct inode *dir;
+	/*
+	 * See if the low-level filesystem might want
+	 * to use its own hash..
+	 */
+	if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) {
+		int err = nd->path.dentry->d_op->d_hash(nd->path.dentry, name);
+		if (err < 0)
+			return err;
+	}
 
+	dentry = __d_lookup(nd->path.dentry, name);
 	if (!dentry)
 		goto need_lookup;
 	if (dentry->d_op && dentry->d_op->d_revalidate)
@@ -810,7 +753,59 @@ done:
 	return 0;
 
 need_lookup:
-	dentry = real_lookup(nd->path.dentry, name, nd);
+	parent = nd->path.dentry;
+	dir = parent->d_inode;
+
+	mutex_lock(&dir->i_mutex);
+	/*
+	 * First re-do the cached lookup just in case it was created
+	 * while we waited for the directory semaphore..
+	 *
+	 * FIXME! This could use version numbering or similar to
+	 * avoid unnecessary cache lookups.
+	 *
+	 * The "dcache_lock" is purely to protect the RCU list walker
+	 * from concurrent renames at this point (we mustn't get false
+	 * negatives from the RCU list walk here, unlike the optimistic
+	 * fast walk).
+	 *
+	 * so doing d_lookup() (with seqlock), instead of lockfree __d_lookup
+	 */
+	dentry = d_lookup(parent, name);
+	if (!dentry) {
+		struct dentry *new;
+
+		/* Don't create child dentry for a dead directory. */
+		dentry = ERR_PTR(-ENOENT);
+		if (IS_DEADDIR(dir))
+			goto out_unlock;
+
+		new = d_alloc(parent, name);
+		dentry = ERR_PTR(-ENOMEM);
+		if (new) {
+			dentry = dir->i_op->lookup(dir, new, nd);
+			if (dentry)
+				dput(new);
+			else
+				dentry = new;
+		}
+out_unlock:
+		mutex_unlock(&dir->i_mutex);
+		if (IS_ERR(dentry))
+			goto fail;
+		goto done;
+	}
+
+	/*
+	 * Uhhuh! Nasty case: the cache was re-populated while
+	 * we waited on the semaphore. Need to revalidate.
+	 */
+	mutex_unlock(&dir->i_mutex);
+	if (dentry->d_op && dentry->d_op->d_revalidate) {
+		dentry = do_revalidate(dentry, nd);
+		if (!dentry)
+			dentry = ERR_PTR(-ENOENT);
+	}
 	if (IS_ERR(dentry))
 		goto fail;
 	goto done;
@@ -835,7 +830,7 @@ fail:
  * Returns 0 and nd will have valid dentry and mnt on success.
  * Returns error and drops reference to input namei data on failure.
  */
-static int __link_path_walk(const char *name, struct nameidata *nd)
+static int link_path_walk(const char *name, struct nameidata *nd)
 {
 	struct path next;
 	struct inode *inode;
@@ -858,7 +853,7 @@ static int __link_path_walk(const char *name, struct nameidata *nd)
 		unsigned int c;
 
 		nd->flags |= LOOKUP_CONTINUE;
-		err = exec_permission_lite(inode);
+		err = exec_permission(inode);
  		if (err)
 			break;
 
@@ -898,16 +893,6 @@ static int __link_path_walk(const char *name, struct nameidata *nd)
 			case 1:
 				continue;
 		}
-		/*
-		 * See if the low-level filesystem might want
-		 * to use its own hash..
-		 */
-		if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) {
-			err = nd->path.dentry->d_op->d_hash(nd->path.dentry,
-							    &this);
-			if (err < 0)
-				break;
-		}
 		/* This does the actual lookups.. */
 		err = do_lookup(nd, &this, &next);
 		if (err)
@@ -953,12 +938,6 @@ last_component:
 			case 1:
 				goto return_reval;
 		}
-		if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) {
-			err = nd->path.dentry->d_op->d_hash(nd->path.dentry,
-							    &this);
-			if (err < 0)
-				break;
-		}
 		err = do_lookup(nd, &this, &next);
 		if (err)
 			break;
@@ -1017,8 +996,27 @@ return_err:
 
 static int path_walk(const char *name, struct nameidata *nd)
 {
+	struct path save = nd->path;
+	int result;
+
 	current->total_link_count = 0;
-	return link_path_walk(name, nd);
+
+	/* make sure the stuff we saved doesn't go away */
+	path_get(&save);
+
+	result = link_path_walk(name, nd);
+	if (result == -ESTALE) {
+		/* nd->path had been dropped */
+		current->total_link_count = 0;
+		nd->path = save;
+		path_get(&nd->path);
+		nd->flags |= LOOKUP_REVAL;
+		result = link_path_walk(name, nd);
+	}
+
+	path_put(&save);
+
+	return result;
 }
 
 static int path_init(int dfd, const char *name, unsigned int flags, struct nameidata *nd)
@@ -1141,36 +1139,6 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt,
 	return retval;
 }
 
-/**
- * path_lookup_open - lookup a file path with open intent
- * @dfd: the directory to use as base, or AT_FDCWD
- * @name: pointer to file name
- * @lookup_flags: lookup intent flags
- * @nd: pointer to nameidata
- * @open_flags: open intent flags
- */
-static int path_lookup_open(int dfd, const char *name,
-		unsigned int lookup_flags, struct nameidata *nd, int open_flags)
-{
-	struct file *filp = get_empty_filp();
-	int err;
-
-	if (filp == NULL)
-		return -ENFILE;
-	nd->intent.open.file = filp;
-	nd->intent.open.flags = open_flags;
-	nd->intent.open.create_mode = 0;
-	err = do_path_lookup(dfd, name, lookup_flags|LOOKUP_OPEN, nd);
-	if (IS_ERR(nd->intent.open.file)) {
-		if (err == 0) {
-			err = PTR_ERR(nd->intent.open.file);
-			path_put(&nd->path);
-		}
-	} else if (err != 0)
-		release_open_intent(nd);
-	return err;
-}
-
 static struct dentry *__lookup_hash(struct qstr *name,
 		struct dentry *base, struct nameidata *nd)
 {
@@ -1191,7 +1159,17 @@ static struct dentry *__lookup_hash(struct qstr *name,
 			goto out;
 	}
 
-	dentry = cached_lookup(base, name, nd);
+	dentry = __d_lookup(base, name);
+
+	/* lockess __d_lookup may fail due to concurrent d_move()
+	 * in some unrelated directory, so try with d_lookup
+	 */
+	if (!dentry)
+		dentry = d_lookup(base, name);
+
+	if (dentry && dentry->d_op && dentry->d_op->d_revalidate)
+		dentry = do_revalidate(dentry, nd);
+
 	if (!dentry) {
 		struct dentry *new;
 
@@ -1223,7 +1201,7 @@ static struct dentry *lookup_hash(struct nameidata *nd)
 {
 	int err;
 
-	err = inode_permission(nd->path.dentry->d_inode, MAY_EXEC);
+	err = exec_permission(nd->path.dentry->d_inode);
 	if (err)
 		return ERR_PTR(err);
 	return __lookup_hash(&nd->last, nd->path.dentry, nd);
@@ -1273,29 +1251,7 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len)
 	if (err)
 		return ERR_PTR(err);
 
-	err = inode_permission(base->d_inode, MAY_EXEC);
-	if (err)
-		return ERR_PTR(err);
-	return __lookup_hash(&this, base, NULL);
-}
-
-/**
- * lookup_one_noperm - bad hack for sysfs
- * @name:	pathname component to lookup
- * @base:	base directory to lookup from
- *
- * This is a variant of lookup_one_len that doesn't perform any permission
- * checks.   It's a horrible hack to work around the braindead sysfs
- * architecture and should not be used anywhere else.
- *
- * DON'T USE THIS FUNCTION EVER, thanks.
- */
-struct dentry *lookup_one_noperm(const char *name, struct dentry *base)
-{
-	int err;
-	struct qstr this;
-
-	err = __lookup_one_len(name, &this, base, strlen(name));
+	err = exec_permission(base->d_inode);
 	if (err)
 		return ERR_PTR(err);
 	return __lookup_hash(&this, base, NULL);
@@ -1533,69 +1489,45 @@ int may_open(struct path *path, int acc_mode, int flag)
 	if (error)
 		return error;
 
-	error = ima_path_check(path, acc_mode ?
-			       acc_mode & (MAY_READ | MAY_WRITE | MAY_EXEC) :
-			       ACC_MODE(flag) & (MAY_READ | MAY_WRITE),
-			       IMA_COUNT_UPDATE);
-
-	if (error)
-		return error;
 	/*
 	 * An append-only file must be opened in append mode for writing.
 	 */
 	if (IS_APPEND(inode)) {
-		error = -EPERM;
 		if  ((flag & FMODE_WRITE) && !(flag & O_APPEND))
-			goto err_out;
+			return -EPERM;
 		if (flag & O_TRUNC)
-			goto err_out;
+			return -EPERM;
 	}
 
 	/* O_NOATIME can only be set by the owner or superuser */
-	if (flag & O_NOATIME)
-		if (!is_owner_or_cap(inode)) {
-			error = -EPERM;
-			goto err_out;
-		}
+	if (flag & O_NOATIME && !is_owner_or_cap(inode))
+		return -EPERM;
 
 	/*
 	 * Ensure there are no outstanding leases on the file.
 	 */
-	error = break_lease(inode, flag);
-	if (error)
-		goto err_out;
-
-	if (flag & O_TRUNC) {
-		error = get_write_access(inode);
-		if (error)
-			goto err_out;
-
-		/*
-		 * Refuse to truncate files with mandatory locks held on them.
-		 */
-		error = locks_verify_locked(inode);
-		if (!error)
-			error = security_path_truncate(path, 0,
-					       ATTR_MTIME|ATTR_CTIME|ATTR_OPEN);
-		if (!error) {
-			vfs_dq_init(inode);
-
-			error = do_truncate(dentry, 0,
-					    ATTR_MTIME|ATTR_CTIME|ATTR_OPEN,
-					    NULL);
-		}
-		put_write_access(inode);
-		if (error)
-			goto err_out;
-	} else
-		if (flag & FMODE_WRITE)
-			vfs_dq_init(inode);
+	return break_lease(inode, flag);
+}
 
-	return 0;
-err_out:
-	ima_counts_put(path, acc_mode ?
-		       acc_mode & (MAY_READ | MAY_WRITE | MAY_EXEC) :
-		       ACC_MODE(flag) & (MAY_READ | MAY_WRITE));
+static int handle_truncate(struct path *path)
+{
+	struct inode *inode = path->dentry->d_inode;
+	int error = get_write_access(inode);
+	if (error)
+		return error;
+	/*
+	 * Refuse to truncate files with mandatory locks held on them.
+	 */
+	error = locks_verify_locked(inode);
+	if (!error)
+		error = security_path_truncate(path, 0,
+				       ATTR_MTIME|ATTR_CTIME|ATTR_OPEN);
+	if (!error) {
+		error = do_truncate(path->dentry, 0,
+				    ATTR_MTIME|ATTR_CTIME|ATTR_OPEN,
+				    NULL);
+	}
+	put_write_access(inode);
 	return error;
 }
 
@@ -1650,7 +1582,7 @@ static inline int open_to_namei_flags(int flag)
 	return flag;
 }
 
-static int open_will_write_to_fs(int flag, struct inode *inode)
+static int open_will_truncate(int flag, struct inode *inode)
 {
 	/*
 	 * We'll never write to the fs underlying
@@ -1675,11 +1607,21 @@ struct file *do_filp_open(int dfd, const char *pathname,
 	struct path path;
 	struct dentry *dir;
 	int count = 0;
-	int will_write;
+	int will_truncate;
 	int flag = open_to_namei_flags(open_flag);
+	int force_reval = 0;
+
+	/*
+	 * O_SYNC is implemented as __O_SYNC|O_DSYNC.  As many places only
+	 * check for O_DSYNC if the need any syncing at all we enforce it's
+	 * always set instead of having to deal with possibly weird behaviour
+	 * for malicious applications setting only __O_SYNC.
+	 */
+	if (open_flag & __O_SYNC)
+		open_flag |= O_DSYNC;
 
 	if (!acc_mode)
-		acc_mode = MAY_OPEN | ACC_MODE(flag);
+		acc_mode = MAY_OPEN | ACC_MODE(open_flag);
 
 	/* O_TRUNC implies we need access checks for write permissions */
 	if (flag & O_TRUNC)
@@ -1694,8 +1636,23 @@ struct file *do_filp_open(int dfd, const char *pathname,
 	 * The simplest case - just a plain lookup.
 	 */
 	if (!(flag & O_CREAT)) {
-		error = path_lookup_open(dfd, pathname, lookup_flags(flag),
-					 &nd, flag);
+		filp = get_empty_filp();
+
+		if (filp == NULL)
+			return ERR_PTR(-ENFILE);
+		nd.intent.open.file = filp;
+		filp->f_flags = open_flag;
+		nd.intent.open.flags = flag;
+		nd.intent.open.create_mode = 0;
+		error = do_path_lookup(dfd, pathname,
+					lookup_flags(flag)|LOOKUP_OPEN, &nd);
+		if (IS_ERR(nd.intent.open.file)) {
+			if (error == 0) {
+				error = PTR_ERR(nd.intent.open.file);
+				path_put(&nd.path);
+			}
+		} else if (error)
+			release_open_intent(&nd);
 		if (error)
 			return ERR_PTR(error);
 		goto ok;
@@ -1704,9 +1661,12 @@ struct file *do_filp_open(int dfd, const char *pathname,
 	/*
 	 * Create - we need to know the parent.
 	 */
+reval:
 	error = path_init(dfd, pathname, LOOKUP_PARENT, &nd);
 	if (error)
 		return ERR_PTR(error);
+	if (force_reval)
+		nd.flags |= LOOKUP_REVAL;
 	error = path_walk(pathname, &nd);
 	if (error) {
 		if (nd.root.mnt)
@@ -1730,6 +1690,7 @@ struct file *do_filp_open(int dfd, const char *pathname,
 	if (filp == NULL)
 		goto exit_parent;
 	nd.intent.open.file = filp;
+	filp->f_flags = open_flag;
 	nd.intent.open.flags = flag;
 	nd.intent.open.create_mode = mode;
 	dir = nd.path.dentry;
@@ -1770,14 +1731,18 @@ do_last:
 			mnt_drop_write(nd.path.mnt);
 			goto exit;
 		}
-		filp = nameidata_to_filp(&nd, open_flag);
-		if (IS_ERR(filp))
-			ima_counts_put(&nd.path,
-				       acc_mode & (MAY_READ | MAY_WRITE |
-						   MAY_EXEC));
+		filp = nameidata_to_filp(&nd);
 		mnt_drop_write(nd.path.mnt);
 		if (nd.root.mnt)
 			path_put(&nd.root);
+		if (!IS_ERR(filp)) {
+			error = ima_path_check(&filp->f_path, filp->f_mode &
+				       (MAY_READ | MAY_WRITE | MAY_EXEC));
+			if (error) {
+				fput(filp);
+				filp = ERR_PTR(error);
+			}
+		}
 		return filp;
 	}
 
@@ -1805,7 +1770,7 @@ do_last:
 
 	path_to_nameidata(&path, &nd);
 	error = -EISDIR;
-	if (path.dentry->d_inode && S_ISDIR(path.dentry->d_inode->i_mode))
+	if (S_ISDIR(path.dentry->d_inode->i_mode))
 		goto exit;
 ok:
 	/*
@@ -1818,28 +1783,45 @@ ok:
 	 * be avoided. Taking this mnt write here
 	 * ensures that (2) can not occur.
 	 */
-	will_write = open_will_write_to_fs(flag, nd.path.dentry->d_inode);
-	if (will_write) {
+	will_truncate = open_will_truncate(flag, nd.path.dentry->d_inode);
+	if (will_truncate) {
 		error = mnt_want_write(nd.path.mnt);
 		if (error)
 			goto exit;
 	}
 	error = may_open(&nd.path, acc_mode, flag);
 	if (error) {
-		if (will_write)
+		if (will_truncate)
 			mnt_drop_write(nd.path.mnt);
 		goto exit;
 	}
-	filp = nameidata_to_filp(&nd, open_flag);
-	if (IS_ERR(filp))
-		ima_counts_put(&nd.path,
-			       acc_mode & (MAY_READ | MAY_WRITE | MAY_EXEC));
+	filp = nameidata_to_filp(&nd);
+	if (!IS_ERR(filp)) {
+		error = ima_path_check(&filp->f_path, filp->f_mode &
+			       (MAY_READ | MAY_WRITE | MAY_EXEC));
+		if (error) {
+			fput(filp);
+			filp = ERR_PTR(error);
+		}
+	}
+	if (!IS_ERR(filp)) {
+		if (acc_mode & MAY_WRITE)
+			vfs_dq_init(nd.path.dentry->d_inode);
+
+		if (will_truncate) {
+			error = handle_truncate(&nd.path);
+			if (error) {
+				fput(filp);
+				filp = ERR_PTR(error);
+			}
+		}
+	}
 	/*
 	 * It is now safe to drop the mnt write
 	 * because the filp has had a write taken
 	 * on its behalf.
 	 */
-	if (will_write)
+	if (will_truncate)
 		mnt_drop_write(nd.path.mnt);
 	if (nd.root.mnt)
 		path_put(&nd.root);
@@ -1877,6 +1859,7 @@ do_link:
 	if (error)
 		goto exit_dput;
 	error = __do_follow_link(&path, &nd);
+	path_put(&path);
 	if (error) {
 		/* Does someone understand code flow here? Or it is only
 		 * me so stupid? Anathema to whoever designed this non-sense
@@ -1885,6 +1868,10 @@ do_link:
 		release_open_intent(&nd);
 		if (nd.root.mnt)
 			path_put(&nd.root);
+		if (error == -ESTALE && !force_reval) {
+			force_reval = 1;
+			goto reval;
+		}
 		return ERR_PTR(error);
 	}
 	nd.flags &= ~LOOKUP_PARENT;
diff --git a/fs/namespace.c b/fs/namespace.c
index 7d70d63ceb29..c768f733c8d6 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -965,10 +965,12 @@ EXPORT_SYMBOL(may_umount_tree);
 int may_umount(struct vfsmount *mnt)
 {
 	int ret = 1;
+	down_read(&namespace_sem);
 	spin_lock(&vfsmount_lock);
 	if (propagate_mount_busy(mnt, 2))
 		ret = 0;
 	spin_unlock(&vfsmount_lock);
+	up_read(&namespace_sem);
 	return ret;
 }
 
@@ -1352,12 +1354,12 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt,
 	if (err)
 		goto out_cleanup_ids;
 
+	spin_lock(&vfsmount_lock);
+
 	if (IS_MNT_SHARED(dest_mnt)) {
 		for (p = source_mnt; p; p = next_mnt(p, source_mnt))
 			set_mnt_shared(p);
 	}
-
-	spin_lock(&vfsmount_lock);
 	if (parent_path) {
 		detach_mnt(source_mnt, parent_path);
 		attach_mnt(source_mnt, path);
@@ -1534,8 +1536,12 @@ static int do_remount(struct path *path, int flags, int mnt_flags,
 		err = change_mount_flags(path->mnt, flags);
 	else
 		err = do_remount_sb(sb, flags, data, 0);
-	if (!err)
+	if (!err) {
+		spin_lock(&vfsmount_lock);
+		mnt_flags |= path->mnt->mnt_flags & MNT_PNODE_MASK;
 		path->mnt->mnt_flags = mnt_flags;
+		spin_unlock(&vfsmount_lock);
+	}
 	up_write(&sb->s_umount);
 	if (!err) {
 		security_sb_post_remount(path->mnt, flags, data);
@@ -1665,6 +1671,8 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path,
 {
 	int err;
 
+	mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD);
+
 	down_write(&namespace_sem);
 	/* Something was mounted here while we slept */
 	while (d_mountpoint(path->dentry) &&
diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig
index 2a77bc25d5af..59e5673b4597 100644
--- a/fs/nfs/Kconfig
+++ b/fs/nfs/Kconfig
@@ -90,7 +90,7 @@ config ROOT_NFS
 	  If you want your system to mount its root file system via NFS,
 	  choose Y here.  This is common practice for managing systems
 	  without local permanent storage.  For details, read
-	  <file:Documentation/filesystems/nfsroot.txt>.
+	  <file:Documentation/filesystems/nfs/nfsroot.txt>.
 
 	  Most people say N here.
 
diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c
index 293fa0528a6e..73ab220354df 100644
--- a/fs/nfs/callback.c
+++ b/fs/nfs/callback.c
@@ -78,11 +78,6 @@ nfs4_callback_svc(void *vrqstp)
 
 	set_freezable();
 
-	/*
-	 * FIXME: do we really need to run this under the BKL? If so, please
-	 * add a comment about what it's intended to protect.
-	 */
-	lock_kernel();
 	while (!kthread_should_stop()) {
 		/*
 		 * Listen for a request on the socket
@@ -104,7 +99,6 @@ nfs4_callback_svc(void *vrqstp)
 		preverr = err;
 		svc_process(rqstp);
 	}
-	unlock_kernel();
 	return 0;
 }
 
@@ -160,11 +154,6 @@ nfs41_callback_svc(void *vrqstp)
 
 	set_freezable();
 
-	/*
-	 * FIXME: do we really need to run this under the BKL? If so, please
-	 * add a comment about what it's intended to protect.
-	 */
-	lock_kernel();
 	while (!kthread_should_stop()) {
 		prepare_to_wait(&serv->sv_cb_waitq, &wq, TASK_INTERRUPTIBLE);
 		spin_lock_bh(&serv->sv_cb_lock);
@@ -183,7 +172,6 @@ nfs41_callback_svc(void *vrqstp)
 		}
 		finish_wait(&serv->sv_cb_waitq, &wq);
 	}
-	unlock_kernel();
 	return 0;
 }
 
@@ -397,6 +385,7 @@ static int nfs_callback_authenticate(struct svc_rqst *rqstp)
  */
 static struct svc_version *nfs4_callback_version[] = {
 	[1] = &nfs4_callback_version1,
+	[4] = &nfs4_callback_version4,
 };
 
 static struct svc_stat nfs4_callback_stats;
diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h
index 07baa8254ca1..d4036be0b589 100644
--- a/fs/nfs/callback.h
+++ b/fs/nfs/callback.h
@@ -106,6 +106,19 @@ struct cb_sequenceres {
 extern unsigned nfs4_callback_sequence(struct cb_sequenceargs *args,
 				       struct cb_sequenceres *res);
 
+extern int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation,
+					     const nfs4_stateid *stateid);
+
+#define RCA4_TYPE_MASK_RDATA_DLG	0
+#define RCA4_TYPE_MASK_WDATA_DLG	1
+
+struct cb_recallanyargs {
+	struct sockaddr	*craa_addr;
+	uint32_t	craa_objs_to_keep;
+	uint32_t	craa_type_mask;
+};
+
+extern unsigned nfs4_callback_recallany(struct cb_recallanyargs *args, void *dummy);
 #endif /* CONFIG_NFS_V4_1 */
 
 extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res);
@@ -114,8 +127,9 @@ extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy);
 #ifdef CONFIG_NFS_V4
 extern int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt);
 extern void nfs_callback_down(int minorversion);
+extern int nfs4_validate_delegation_stateid(struct nfs_delegation *delegation,
+					    const nfs4_stateid *stateid);
 #endif /* CONFIG_NFS_V4 */
-
 /*
  * nfs41: Callbacks are expected to not cause substantial latency,
  * so we limit their concurrency to 1 by setting up the maximum number
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index b7da1f54da68..defa9b4c470e 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -61,6 +61,16 @@ out:
 	return res->status;
 }
 
+static int (*nfs_validate_delegation_stateid(struct nfs_client *clp))(struct nfs_delegation *, const nfs4_stateid *)
+{
+#if defined(CONFIG_NFS_V4_1)
+	if (clp->cl_minorversion > 0)
+		return nfs41_validate_delegation_stateid;
+#endif
+	return nfs4_validate_delegation_stateid;
+}
+
+
 __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy)
 {
 	struct nfs_client *clp;
@@ -81,7 +91,8 @@ __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy)
 		inode = nfs_delegation_find_inode(clp, &args->fh);
 		if (inode != NULL) {
 			/* Set up a helper thread to actually return the delegation */
-			switch(nfs_async_inode_return_delegation(inode, &args->stateid)) {
+			switch (nfs_async_inode_return_delegation(inode, &args->stateid,
+								  nfs_validate_delegation_stateid(clp))) {
 				case 0:
 					res = 0;
 					break;
@@ -102,8 +113,31 @@ out:
 	return res;
 }
 
+int nfs4_validate_delegation_stateid(struct nfs_delegation *delegation, const nfs4_stateid *stateid)
+{
+	if (delegation == NULL || memcmp(delegation->stateid.data, stateid->data,
+					 sizeof(delegation->stateid.data)) != 0)
+		return 0;
+	return 1;
+}
+
 #if defined(CONFIG_NFS_V4_1)
 
+int nfs41_validate_delegation_stateid(struct nfs_delegation *delegation, const nfs4_stateid *stateid)
+{
+	if (delegation == NULL)
+		return 0;
+
+	/* seqid is 4-bytes long */
+	if (((u32 *) &stateid->data)[0] != 0)
+		return 0;
+	if (memcmp(&delegation->stateid.data[4], &stateid->data[4],
+		   sizeof(stateid->data)-4))
+		return 0;
+
+	return 1;
+}
+
 /*
  * Validate the sequenceID sent by the server.
  * Return success if the sequenceID is one more than what we last saw on
@@ -227,4 +261,32 @@ out:
 	return res->csr_status;
 }
 
+unsigned nfs4_callback_recallany(struct cb_recallanyargs *args, void *dummy)
+{
+	struct nfs_client *clp;
+	int status;
+	fmode_t flags = 0;
+
+	status = htonl(NFS4ERR_OP_NOT_IN_SESSION);
+	clp = nfs_find_client(args->craa_addr, 4);
+	if (clp == NULL)
+		goto out;
+
+	dprintk("NFS: RECALL_ANY callback request from %s\n",
+		rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR));
+
+	if (test_bit(RCA4_TYPE_MASK_RDATA_DLG, (const unsigned long *)
+		     &args->craa_type_mask))
+		flags = FMODE_READ;
+	if (test_bit(RCA4_TYPE_MASK_WDATA_DLG, (const unsigned long *)
+		     &args->craa_type_mask))
+		flags |= FMODE_WRITE;
+
+	if (flags)
+		nfs_expire_all_delegation_types(clp, flags);
+	status = htonl(NFS4_OK);
+out:
+	dprintk("%s: exit with status = %d\n", __func__, ntohl(status));
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c
index 76b0aa0f73bf..8e1a2511c8be 100644
--- a/fs/nfs/callback_xdr.c
+++ b/fs/nfs/callback_xdr.c
@@ -23,6 +23,7 @@
 #if defined(CONFIG_NFS_V4_1)
 #define CB_OP_SEQUENCE_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ + \
 					4 + 1 + 3)
+#define CB_OP_RECALLANY_RES_MAXSZ	(CB_OP_HDR_RES_MAXSZ)
 #endif /* CONFIG_NFS_V4_1 */
 
 #define NFSDBG_FACILITY NFSDBG_CALLBACK
@@ -326,6 +327,25 @@ out_free:
 	goto out;
 }
 
+static unsigned decode_recallany_args(struct svc_rqst *rqstp,
+				      struct xdr_stream *xdr,
+				      struct cb_recallanyargs *args)
+{
+	uint32_t *p;
+
+	args->craa_addr = svc_addr(rqstp);
+	p = read_buf(xdr, 4);
+	if (unlikely(p == NULL))
+		return htonl(NFS4ERR_BADXDR);
+	args->craa_objs_to_keep = ntohl(*p++);
+	p = read_buf(xdr, 4);
+	if (unlikely(p == NULL))
+		return htonl(NFS4ERR_BADXDR);
+	args->craa_type_mask = ntohl(*p);
+
+	return 0;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str)
@@ -533,6 +553,7 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 	case OP_CB_GETATTR:
 	case OP_CB_RECALL:
 	case OP_CB_SEQUENCE:
+	case OP_CB_RECALL_ANY:
 		*op = &callback_ops[op_nr];
 		break;
 
@@ -540,7 +561,6 @@ preprocess_nfs41_op(int nop, unsigned int op_nr, struct callback_op **op)
 	case OP_CB_NOTIFY_DEVICEID:
 	case OP_CB_NOTIFY:
 	case OP_CB_PUSH_DELEG:
-	case OP_CB_RECALL_ANY:
 	case OP_CB_RECALLABLE_OBJ_AVAIL:
 	case OP_CB_RECALL_SLOT:
 	case OP_CB_WANTS_CANCELLED:
@@ -688,6 +708,11 @@ static struct callback_op callback_ops[] = {
 		.encode_res = (callback_encode_res_t)encode_cb_sequence_res,
 		.res_maxsize = CB_OP_SEQUENCE_RES_MAXSZ,
 	},
+	[OP_CB_RECALL_ANY] = {
+		.process_op = (callback_process_op_t)nfs4_callback_recallany,
+		.decode_args = (callback_decode_arg_t)decode_recallany_args,
+		.res_maxsize = CB_OP_RECALLANY_RES_MAXSZ,
+	},
 #endif /* CONFIG_NFS_V4_1 */
 };
 
@@ -718,3 +743,10 @@ struct svc_version nfs4_callback_version1 = {
 	.vs_dispatch = NULL,
 };
 
+struct svc_version nfs4_callback_version4 = {
+	.vs_vers = 4,
+	.vs_nproc = ARRAY_SIZE(nfs4_callback_procedures1),
+	.vs_proc = nfs4_callback_procedures1,
+	.vs_xdrsize = NFS4_CALLBACK_XDRSIZE,
+	.vs_dispatch = NULL,
+};
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 99ea196f071f..ee77713ce68b 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -1260,10 +1260,20 @@ error:
 static void nfs4_session_set_rwsize(struct nfs_server *server)
 {
 #ifdef CONFIG_NFS_V4_1
+	struct nfs4_session *sess;
+	u32 server_resp_sz;
+	u32 server_rqst_sz;
+
 	if (!nfs4_has_session(server->nfs_client))
 		return;
-	server->rsize = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
-	server->wsize = server->nfs_client->cl_session->fc_attrs.max_rqst_sz;
+	sess = server->nfs_client->cl_session;
+	server_resp_sz = sess->fc_attrs.max_resp_sz - nfs41_maxread_overhead;
+	server_rqst_sz = sess->fc_attrs.max_rqst_sz - nfs41_maxwrite_overhead;
+
+	if (server->rsize > server_resp_sz)
+		server->rsize = server_resp_sz;
+	if (server->wsize > server_rqst_sz)
+		server->wsize = server_rqst_sz;
 #endif /* CONFIG_NFS_V4_1 */
 }
 
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 6dd48a4405b4..2563bebc4c67 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -92,7 +92,7 @@ out:
 	return status;
 }
 
-static void nfs_delegation_claim_opens(struct inode *inode, const nfs4_stateid *stateid)
+static int nfs_delegation_claim_opens(struct inode *inode, const nfs4_stateid *stateid)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	struct nfs_open_context *ctx;
@@ -116,10 +116,11 @@ again:
 			err = nfs_delegation_claim_locks(ctx, state);
 		put_nfs_open_context(ctx);
 		if (err != 0)
-			return;
+			return err;
 		goto again;
 	}
 	spin_unlock(&inode->i_lock);
+	return 0;
 }
 
 /*
@@ -261,30 +262,34 @@ static void nfs_msync_inode(struct inode *inode)
 /*
  * Basic procedure for returning a delegation to the server
  */
-static int __nfs_inode_return_delegation(struct inode *inode, struct nfs_delegation *delegation)
+static int __nfs_inode_return_delegation(struct inode *inode, struct nfs_delegation *delegation, int issync)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
+	int err;
 
-	nfs_msync_inode(inode);
 	/*
 	 * Guard against new delegated open/lock/unlock calls and against
 	 * state recovery
 	 */
 	down_write(&nfsi->rwsem);
-	nfs_delegation_claim_opens(inode, &delegation->stateid);
+	err = nfs_delegation_claim_opens(inode, &delegation->stateid);
 	up_write(&nfsi->rwsem);
-	nfs_msync_inode(inode);
+	if (err)
+		goto out;
 
-	return nfs_do_return_delegation(inode, delegation, 1);
+	err = nfs_do_return_delegation(inode, delegation, issync);
+out:
+	return err;
 }
 
 /*
  * Return all delegations that have been marked for return
  */
-void nfs_client_return_marked_delegations(struct nfs_client *clp)
+int nfs_client_return_marked_delegations(struct nfs_client *clp)
 {
 	struct nfs_delegation *delegation;
 	struct inode *inode;
+	int err = 0;
 
 restart:
 	rcu_read_lock();
@@ -298,12 +303,18 @@ restart:
 		delegation = nfs_detach_delegation_locked(NFS_I(inode), NULL);
 		spin_unlock(&clp->cl_lock);
 		rcu_read_unlock();
-		if (delegation != NULL)
-			__nfs_inode_return_delegation(inode, delegation);
+		if (delegation != NULL) {
+			filemap_flush(inode->i_mapping);
+			err = __nfs_inode_return_delegation(inode, delegation, 0);
+		}
 		iput(inode);
-		goto restart;
+		if (!err)
+			goto restart;
+		set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
+		return err;
 	}
 	rcu_read_unlock();
+	return 0;
 }
 
 /*
@@ -338,8 +349,10 @@ int nfs_inode_return_delegation(struct inode *inode)
 		spin_lock(&clp->cl_lock);
 		delegation = nfs_detach_delegation_locked(nfsi, NULL);
 		spin_unlock(&clp->cl_lock);
-		if (delegation != NULL)
-			err = __nfs_inode_return_delegation(inode, delegation);
+		if (delegation != NULL) {
+			nfs_msync_inode(inode);
+			err = __nfs_inode_return_delegation(inode, delegation, 1);
+		}
 	}
 	return err;
 }
@@ -368,33 +381,47 @@ void nfs_super_return_all_delegations(struct super_block *sb)
 		spin_unlock(&delegation->lock);
 	}
 	rcu_read_unlock();
-	nfs_client_return_marked_delegations(clp);
+	if (nfs_client_return_marked_delegations(clp) != 0)
+		nfs4_schedule_state_manager(clp);
 }
 
-static void nfs_client_mark_return_all_delegations(struct nfs_client *clp)
+static
+void nfs_client_mark_return_all_delegation_types(struct nfs_client *clp, fmode_t flags)
 {
 	struct nfs_delegation *delegation;
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(delegation, &clp->cl_delegations, super_list) {
-		set_bit(NFS_DELEGATION_RETURN, &delegation->flags);
-		set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
+		if ((delegation->type == (FMODE_READ|FMODE_WRITE)) && !(flags & FMODE_WRITE))
+			continue;
+		if (delegation->type & flags)
+			nfs_mark_return_delegation(clp, delegation);
 	}
 	rcu_read_unlock();
 }
 
+static void nfs_client_mark_return_all_delegations(struct nfs_client *clp)
+{
+	nfs_client_mark_return_all_delegation_types(clp, FMODE_READ|FMODE_WRITE);
+}
+
 static void nfs_delegation_run_state_manager(struct nfs_client *clp)
 {
 	if (test_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state))
 		nfs4_schedule_state_manager(clp);
 }
 
-void nfs_expire_all_delegations(struct nfs_client *clp)
+void nfs_expire_all_delegation_types(struct nfs_client *clp, fmode_t flags)
 {
-	nfs_client_mark_return_all_delegations(clp);
+	nfs_client_mark_return_all_delegation_types(clp, flags);
 	nfs_delegation_run_state_manager(clp);
 }
 
+void nfs_expire_all_delegations(struct nfs_client *clp)
+{
+	nfs_expire_all_delegation_types(clp, FMODE_READ|FMODE_WRITE);
+}
+
 /*
  * Return all delegations following an NFS4ERR_CB_PATH_DOWN error.
  */
@@ -413,8 +440,7 @@ static void nfs_client_mark_return_unreferenced_delegations(struct nfs_client *c
 	list_for_each_entry_rcu(delegation, &clp->cl_delegations, super_list) {
 		if (test_and_clear_bit(NFS_DELEGATION_REFERENCED, &delegation->flags))
 			continue;
-		set_bit(NFS_DELEGATION_RETURN, &delegation->flags);
-		set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state);
+		nfs_mark_return_delegation(clp, delegation);
 	}
 	rcu_read_unlock();
 }
@@ -428,18 +454,21 @@ void nfs_expire_unreferenced_delegations(struct nfs_client *clp)
 /*
  * Asynchronous delegation recall!
  */
-int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid)
+int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid,
+				      int (*validate_stateid)(struct nfs_delegation *delegation,
+							      const nfs4_stateid *stateid))
 {
 	struct nfs_client *clp = NFS_SERVER(inode)->nfs_client;
 	struct nfs_delegation *delegation;
 
 	rcu_read_lock();
 	delegation = rcu_dereference(NFS_I(inode)->delegation);
-	if (delegation == NULL || memcmp(delegation->stateid.data, stateid->data,
-				sizeof(delegation->stateid.data)) != 0) {
+
+	if (!validate_stateid(delegation, stateid)) {
 		rcu_read_unlock();
 		return -ENOENT;
 	}
+
 	nfs_mark_return_delegation(clp, delegation);
 	rcu_read_unlock();
 	nfs_delegation_run_state_manager(clp);
diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h
index 09f383795174..944b627ec6e1 100644
--- a/fs/nfs/delegation.h
+++ b/fs/nfs/delegation.h
@@ -34,15 +34,18 @@ enum {
 int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res);
 void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res);
 int nfs_inode_return_delegation(struct inode *inode);
-int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid);
+int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid,
+				      int (*validate_stateid)(struct nfs_delegation *delegation,
+							      const nfs4_stateid *stateid));
 void nfs_inode_return_delegation_noreclaim(struct inode *inode);
 
 struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle);
 void nfs_super_return_all_delegations(struct super_block *sb);
 void nfs_expire_all_delegations(struct nfs_client *clp);
+void nfs_expire_all_delegation_types(struct nfs_client *clp, fmode_t flags);
 void nfs_expire_unreferenced_delegations(struct nfs_client *clp);
 void nfs_handle_cb_pathdown(struct nfs_client *clp);
-void nfs_client_return_marked_delegations(struct nfs_client *clp);
+int nfs_client_return_marked_delegations(struct nfs_client *clp);
 
 void nfs_delegation_mark_reclaim(struct nfs_client *clp);
 void nfs_delegation_reap_unclaimed(struct nfs_client *clp);
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 7cb298525eef..3c7f03b669fb 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1579,55 +1579,47 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct dentry *dentry = NULL, *rehash = NULL;
 	int error = -EBUSY;
 
-	/*
-	 * To prevent any new references to the target during the rename,
-	 * we unhash the dentry and free the inode in advance.
-	 */
-	if (!d_unhashed(new_dentry)) {
-		d_drop(new_dentry);
-		rehash = new_dentry;
-	}
-
 	dfprintk(VFS, "NFS: rename(%s/%s -> %s/%s, ct=%d)\n",
 		 old_dentry->d_parent->d_name.name, old_dentry->d_name.name,
 		 new_dentry->d_parent->d_name.name, new_dentry->d_name.name,
 		 atomic_read(&new_dentry->d_count));
 
 	/*
-	 * First check whether the target is busy ... we can't
-	 * safely do _any_ rename if the target is in use.
-	 *
-	 * For files, make a copy of the dentry and then do a 
-	 * silly-rename. If the silly-rename succeeds, the
-	 * copied dentry is hashed and becomes the new target.
+	 * For non-directories, check whether the target is busy and if so,
+	 * make a copy of the dentry and then do a silly-rename. If the
+	 * silly-rename succeeds, the copied dentry is hashed and becomes
+	 * the new target.
 	 */
-	if (!new_inode)
-		goto go_ahead;
-	if (S_ISDIR(new_inode->i_mode)) {
-		error = -EISDIR;
-		if (!S_ISDIR(old_inode->i_mode))
-			goto out;
-	} else if (atomic_read(&new_dentry->d_count) > 2) {
-		int err;
-		/* copy the target dentry's name */
-		dentry = d_alloc(new_dentry->d_parent,
-				 &new_dentry->d_name);
-		if (!dentry)
-			goto out;
+	if (new_inode && !S_ISDIR(new_inode->i_mode)) {
+		/*
+		 * To prevent any new references to the target during the
+		 * rename, we unhash the dentry in advance.
+		 */
+		if (!d_unhashed(new_dentry)) {
+			d_drop(new_dentry);
+			rehash = new_dentry;
+		}
+
+		if (atomic_read(&new_dentry->d_count) > 2) {
+			int err;
+
+			/* copy the target dentry's name */
+			dentry = d_alloc(new_dentry->d_parent,
+					 &new_dentry->d_name);
+			if (!dentry)
+				goto out;
 
-		/* silly-rename the existing target ... */
-		err = nfs_sillyrename(new_dir, new_dentry);
-		if (!err) {
-			new_dentry = rehash = dentry;
+			/* silly-rename the existing target ... */
+			err = nfs_sillyrename(new_dir, new_dentry);
+			if (err)
+				goto out;
+
+			new_dentry = dentry;
+			rehash = NULL;
 			new_inode = NULL;
-			/* instantiate the replacement target */
-			d_instantiate(new_dentry, NULL);
-		} else if (atomic_read(&new_dentry->d_count) > 1)
-			/* dentry still busy? */
-			goto out;
+		}
 	}
 
-go_ahead:
 	/*
 	 * ... prune child dentries and writebacks if needed.
 	 */
diff --git a/fs/nfs/dns_resolve.c b/fs/nfs/dns_resolve.c
index f4d54ba97cc6..95e1ca765d47 100644
--- a/fs/nfs/dns_resolve.c
+++ b/fs/nfs/dns_resolve.c
@@ -146,7 +146,7 @@ static int nfs_dns_show(struct seq_file *m, struct cache_detail *cd,
 	return 0;
 }
 
-struct nfs_dns_ent *nfs_dns_lookup(struct cache_detail *cd,
+static struct nfs_dns_ent *nfs_dns_lookup(struct cache_detail *cd,
 		struct nfs_dns_ent *key)
 {
 	struct cache_head *ch;
@@ -159,7 +159,7 @@ struct nfs_dns_ent *nfs_dns_lookup(struct cache_detail *cd,
 	return container_of(ch, struct nfs_dns_ent, h);
 }
 
-struct nfs_dns_ent *nfs_dns_update(struct cache_detail *cd,
+static struct nfs_dns_ent *nfs_dns_update(struct cache_detail *cd,
 		struct nfs_dns_ent *new,
 		struct nfs_dns_ent *key)
 {
diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index f5fdd39e037a..6b891328f332 100644
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -581,7 +581,7 @@ static int nfs_need_sync_write(struct file *filp, struct inode *inode)
 {
 	struct nfs_open_context *ctx;
 
-	if (IS_SYNC(inode) || (filp->f_flags & O_SYNC))
+	if (IS_SYNC(inode) || (filp->f_flags & O_DSYNC))
 		return 1;
 	ctx = nfs_file_open_context(filp);
 	if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags))
@@ -622,7 +622,7 @@ static ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov,
 
 	nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, count);
 	result = generic_file_aio_write(iocb, iov, nr_segs, pos);
-	/* Return error values for O_SYNC and IS_SYNC() */
+	/* Return error values for O_DSYNC and IS_SYNC() */
 	if (result >= 0 && nfs_need_sync_write(iocb->ki_filp, inode)) {
 		int err = nfs_do_fsync(nfs_file_open_context(iocb->ki_filp), inode);
 		if (err < 0)
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index e21b1bb9972f..29e464d23b32 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -30,6 +30,15 @@ static inline int nfs4_has_session(const struct nfs_client *clp)
 	return 0;
 }
 
+static inline int nfs4_has_persistent_session(const struct nfs_client *clp)
+{
+#ifdef CONFIG_NFS_V4_1
+	if (nfs4_has_session(clp))
+		return (clp->cl_session->flags & SESSION4_PERSIST);
+#endif /* CONFIG_NFS_V4_1 */
+	return 0;
+}
+
 struct nfs_clone_mount {
 	const struct super_block *sb;
 	const struct dentry *dentry;
@@ -156,6 +165,7 @@ struct vfsmount *nfs_do_refmount(const struct vfsmount *mnt_parent, struct dentr
 
 /* callback_xdr.c */
 extern struct svc_version nfs4_callback_version1;
+extern struct svc_version nfs4_callback_version4;
 
 /* pagelist.c */
 extern int __init nfs_init_nfspagecache(void);
@@ -177,24 +187,14 @@ extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int);
 extern struct rpc_procinfo nfs3_procedures[];
 extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int);
 
-/* nfs4proc.c */
-static inline void nfs4_restart_rpc(struct rpc_task *task,
-				    const struct nfs_client *clp)
-{
-#ifdef CONFIG_NFS_V4_1
-	if (nfs4_has_session(clp) &&
-	    test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)) {
-		rpc_restart_call_prepare(task);
-		return;
-	}
-#endif /* CONFIG_NFS_V4_1 */
-	rpc_restart_call(task);
-}
-
 /* nfs4xdr.c */
 #ifdef CONFIG_NFS_V4
 extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus);
 #endif
+#ifdef CONFIG_NFS_V4_1
+extern const u32 nfs41_maxread_overhead;
+extern const u32 nfs41_maxwrite_overhead;
+#endif
 
 /* nfs4proc.c */
 #ifdef CONFIG_NFS_V4
@@ -273,20 +273,6 @@ extern int _nfs4_call_sync_session(struct nfs_server *server,
 				   struct nfs4_sequence_res *res,
 				   int cache_reply);
 
-#ifdef CONFIG_NFS_V4_1
-extern void nfs41_sequence_free_slot(const struct nfs_client *,
-				     struct nfs4_sequence_res *res);
-#endif /* CONFIG_NFS_V4_1 */
-
-static inline void nfs4_sequence_free_slot(const struct nfs_client *clp,
-					   struct nfs4_sequence_res *res)
-{
-#ifdef CONFIG_NFS_V4_1
-	if (nfs4_has_session(clp))
-		nfs41_sequence_free_slot(clp, res);
-#endif /* CONFIG_NFS_V4_1 */
-}
-
 /*
  * Determine the device name as a string
  */
@@ -380,3 +366,15 @@ unsigned int nfs_page_array_len(unsigned int base, size_t len)
 	return ((unsigned long)len + (unsigned long)base +
 		PAGE_SIZE - 1) >> PAGE_SHIFT;
 }
+
+/*
+ * Helper for restarting RPC calls in the possible presence of NFSv4.1
+ * sessions.
+ */
+static inline void nfs_restart_rpc(struct rpc_task *task, const struct nfs_client *clp)
+{
+	if (nfs4_has_session(clp))
+		rpc_restart_call_prepare(task);
+	else
+		rpc_restart_call(task);
+}
diff --git a/fs/nfs/iostat.h b/fs/nfs/iostat.h
index ceda50aad73c..46d779abafd3 100644
--- a/fs/nfs/iostat.h
+++ b/fs/nfs/iostat.h
@@ -25,13 +25,7 @@ struct nfs_iostats {
 static inline void nfs_inc_server_stats(const struct nfs_server *server,
 					enum nfs_stat_eventcounters stat)
 {
-	struct nfs_iostats *iostats;
-	int cpu;
-
-	cpu = get_cpu();
-	iostats = per_cpu_ptr(server->io_stats, cpu);
-	iostats->events[stat]++;
-	put_cpu();
+	this_cpu_inc(server->io_stats->events[stat]);
 }
 
 static inline void nfs_inc_stats(const struct inode *inode,
@@ -44,13 +38,7 @@ static inline void nfs_add_server_stats(const struct nfs_server *server,
 					enum nfs_stat_bytecounters stat,
 					unsigned long addend)
 {
-	struct nfs_iostats *iostats;
-	int cpu;
-
-	cpu = get_cpu();
-	iostats = per_cpu_ptr(server->io_stats, cpu);
-	iostats->bytes[stat] += addend;
-	put_cpu();
+	this_cpu_add(server->io_stats->bytes[stat], addend);
 }
 
 static inline void nfs_add_stats(const struct inode *inode,
@@ -65,13 +53,7 @@ static inline void nfs_add_fscache_stats(struct inode *inode,
 					 enum nfs_stat_fscachecounters stat,
 					 unsigned long addend)
 {
-	struct nfs_iostats *iostats;
-	int cpu;
-
-	cpu = get_cpu();
-	iostats = per_cpu_ptr(NFS_SERVER(inode)->io_stats, cpu);
-	iostats->fscache[stat] += addend;
-	put_cpu();
+	this_cpu_add(NFS_SERVER(inode)->io_stats->fscache[stat], addend);
 }
 #endif
 
diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
index 6ea07a3c75d4..865265bdca03 100644
--- a/fs/nfs/nfs4_fs.h
+++ b/fs/nfs/nfs4_fs.h
@@ -44,7 +44,8 @@ enum nfs4_client_state {
 	NFS4CLNT_RECLAIM_REBOOT,
 	NFS4CLNT_RECLAIM_NOGRACE,
 	NFS4CLNT_DELEGRETURN,
-	NFS4CLNT_SESSION_SETUP,
+	NFS4CLNT_SESSION_RESET,
+	NFS4CLNT_SESSION_DRAINING,
 };
 
 /*
@@ -107,6 +108,10 @@ enum {
 	NFS_OWNER_RECLAIM_NOGRACE
 };
 
+#define NFS_LOCK_NEW		0
+#define NFS_LOCK_RECLAIM	1
+#define NFS_LOCK_EXPIRED	2
+
 /*
  * struct nfs4_state maintains the client-side state for a given
  * (state_owner,inode) tuple (OPEN) or state_owner (LOCK).
@@ -180,6 +185,7 @@ struct nfs4_state_recovery_ops {
 	int (*recover_lock)(struct nfs4_state *, struct file_lock *);
 	int (*establish_clid)(struct nfs_client *, struct rpc_cred *);
 	struct rpc_cred * (*get_clid_cred)(struct nfs_client *);
+	int (*reclaim_complete)(struct nfs_client *);
 };
 
 struct nfs4_state_maintenance_ops {
@@ -200,9 +206,11 @@ extern ssize_t nfs4_listxattr(struct dentry *, char *, size_t);
 /* nfs4proc.c */
 extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *);
 extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct rpc_cred *);
+extern int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred);
 extern int nfs4_proc_async_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_proc_renew(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_init_clientid(struct nfs_client *, struct rpc_cred *);
+extern int nfs41_init_clientid(struct nfs_client *, struct rpc_cred *);
 extern int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait);
 extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *);
 extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *);
@@ -218,9 +226,11 @@ extern int nfs4_setup_sequence(struct nfs_client *clp,
 		int cache_reply, struct rpc_task *task);
 extern void nfs4_destroy_session(struct nfs4_session *session);
 extern struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp);
-extern int nfs4_proc_create_session(struct nfs_client *, int reset);
+extern int nfs4_proc_create_session(struct nfs_client *);
 extern int nfs4_proc_destroy_session(struct nfs4_session *);
 extern int nfs4_init_session(struct nfs_server *server);
+extern int nfs4_proc_get_lease_time(struct nfs_client *clp,
+		struct nfs_fsinfo *fsinfo);
 #else /* CONFIG_NFS_v4_1 */
 static inline int nfs4_setup_sequence(struct nfs_client *clp,
 		struct nfs4_sequence_args *args, struct nfs4_sequence_res *res,
@@ -267,6 +277,7 @@ extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t);
 extern void nfs4_schedule_state_recovery(struct nfs_client *);
 extern void nfs4_schedule_state_manager(struct nfs_client *);
 extern int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state);
+extern void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags);
 extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp);
 extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl);
 extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t);
@@ -275,6 +286,7 @@ extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter);
 extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task);
 extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid);
 extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid);
+extern void nfs_release_seqid(struct nfs_seqid *seqid);
 extern void nfs_free_seqid(struct nfs_seqid *seqid);
 
 extern const nfs4_stateid zero_stateid;
@@ -287,6 +299,7 @@ struct nfs4_mount_data;
 
 /* callback_xdr.c */
 extern struct svc_version nfs4_callback_version1;
+extern struct svc_version nfs4_callback_version4;
 
 #else
 
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 741a562177fc..198d51d17c13 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -64,6 +64,7 @@
 
 struct nfs4_opendata;
 static int _nfs4_proc_open(struct nfs4_opendata *data);
+static int _nfs4_recover_proc_open(struct nfs4_opendata *data);
 static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *);
 static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *, struct nfs4_state *);
 static int _nfs4_proc_lookup(struct inode *dir, const struct qstr *name, struct nfs_fh *fhandle, struct nfs_fattr *fattr);
@@ -270,11 +271,18 @@ static int nfs4_handle_exception(const struct nfs_server *server, int errorcode,
 		case -NFS4ERR_SEQ_MISORDERED:
 			dprintk("%s ERROR: %d Reset session\n", __func__,
 				errorcode);
-			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			nfs4_schedule_state_recovery(clp);
 			exception->retry = 1;
-			/* FALLTHROUGH */
+			break;
 #endif /* !defined(CONFIG_NFS_V4_1) */
 		case -NFS4ERR_FILE_OPEN:
+			if (exception->timeout > HZ) {
+				/* We have retried a decent amount, time to
+				 * fail
+				 */
+				ret = -EBUSY;
+				break;
+			}
 		case -NFS4ERR_GRACE:
 		case -NFS4ERR_DELAY:
 			ret = nfs4_delay(server->client, &exception->timeout);
@@ -311,48 +319,67 @@ static void renew_lease(const struct nfs_server *server, unsigned long timestamp
  * so we need to scan down from highest_used_slotid to 0 looking for the now
  * highest slotid in use.
  * If none found, highest_used_slotid is set to -1.
+ *
+ * Must be called while holding tbl->slot_tbl_lock
  */
 static void
 nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid)
 {
 	int slotid = free_slotid;
 
-	spin_lock(&tbl->slot_tbl_lock);
 	/* clear used bit in bitmap */
 	__clear_bit(slotid, tbl->used_slots);
 
 	/* update highest_used_slotid when it is freed */
 	if (slotid == tbl->highest_used_slotid) {
 		slotid = find_last_bit(tbl->used_slots, tbl->max_slots);
-		if (slotid >= 0 && slotid < tbl->max_slots)
+		if (slotid < tbl->max_slots)
 			tbl->highest_used_slotid = slotid;
 		else
 			tbl->highest_used_slotid = -1;
 	}
-	rpc_wake_up_next(&tbl->slot_tbl_waitq);
-	spin_unlock(&tbl->slot_tbl_lock);
 	dprintk("%s: free_slotid %u highest_used_slotid %d\n", __func__,
 		free_slotid, tbl->highest_used_slotid);
 }
 
-void nfs41_sequence_free_slot(const struct nfs_client *clp,
-			      struct nfs4_sequence_res *res)
+/*
+ * Signal state manager thread if session is drained
+ */
+static void nfs41_check_drain_session_complete(struct nfs4_session *ses)
 {
-	struct nfs4_slot_table *tbl;
+	struct rpc_task *task;
 
-	if (!nfs4_has_session(clp)) {
-		dprintk("%s: No session\n", __func__);
+	if (!test_bit(NFS4CLNT_SESSION_DRAINING, &ses->clp->cl_state)) {
+		task = rpc_wake_up_next(&ses->fc_slot_table.slot_tbl_waitq);
+		if (task)
+			rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
 		return;
 	}
+
+	if (ses->fc_slot_table.highest_used_slotid != -1)
+		return;
+
+	dprintk("%s COMPLETE: Session Drained\n", __func__);
+	complete(&ses->complete);
+}
+
+static void nfs41_sequence_free_slot(const struct nfs_client *clp,
+			      struct nfs4_sequence_res *res)
+{
+	struct nfs4_slot_table *tbl;
+
 	tbl = &clp->cl_session->fc_slot_table;
 	if (res->sr_slotid == NFS4_MAX_SLOT_TABLE) {
-		dprintk("%s: No slot\n", __func__);
 		/* just wake up the next guy waiting since
 		 * we may have not consumed a slot after all */
-		rpc_wake_up_next(&tbl->slot_tbl_waitq);
+		dprintk("%s: No slot\n", __func__);
 		return;
 	}
+
+	spin_lock(&tbl->slot_tbl_lock);
 	nfs4_free_slot(tbl, res->sr_slotid);
+	nfs41_check_drain_session_complete(clp->cl_session);
+	spin_unlock(&tbl->slot_tbl_lock);
 	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
 }
 
@@ -377,10 +404,10 @@ static void nfs41_sequence_done(struct nfs_client *clp,
 	if (res->sr_slotid == NFS4_MAX_SLOT_TABLE)
 		goto out;
 
-	tbl = &clp->cl_session->fc_slot_table;
-	slot = tbl->slots + res->sr_slotid;
-
+	/* Check the SEQUENCE operation status */
 	if (res->sr_status == 0) {
+		tbl = &clp->cl_session->fc_slot_table;
+		slot = tbl->slots + res->sr_slotid;
 		/* Update the slot's sequence and clientid lease timer */
 		++slot->seq_nr;
 		timestamp = res->sr_renewal_time;
@@ -388,7 +415,8 @@ static void nfs41_sequence_done(struct nfs_client *clp,
 		if (time_before(clp->cl_last_renewal, timestamp))
 			clp->cl_last_renewal = timestamp;
 		spin_unlock(&clp->cl_lock);
-		return;
+		/* Check sequence flags */
+		nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags);
 	}
 out:
 	/* The session may be reset by one of the error handlers. */
@@ -407,7 +435,7 @@ out:
  * Note: must be called with under the slot_tbl_lock.
  */
 static u8
-nfs4_find_slot(struct nfs4_slot_table *tbl, struct rpc_task *task)
+nfs4_find_slot(struct nfs4_slot_table *tbl)
 {
 	int slotid;
 	u8 ret_id = NFS4_MAX_SLOT_TABLE;
@@ -429,24 +457,6 @@ out:
 	return ret_id;
 }
 
-static int nfs4_recover_session(struct nfs4_session *session)
-{
-	struct nfs_client *clp = session->clp;
-	unsigned int loop;
-	int ret;
-
-	for (loop = NFS4_MAX_LOOP_ON_RECOVER; loop != 0; loop--) {
-		ret = nfs4_wait_clnt_recover(clp);
-		if (ret != 0)
-			break;
-		if (!test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
-			break;
-		nfs4_schedule_state_manager(clp);
-		ret = -EIO;
-	}
-	return ret;
-}
-
 static int nfs41_setup_sequence(struct nfs4_session *session,
 				struct nfs4_sequence_args *args,
 				struct nfs4_sequence_res *res,
@@ -455,7 +465,6 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 {
 	struct nfs4_slot *slot;
 	struct nfs4_slot_table *tbl;
-	int status = 0;
 	u8 slotid;
 
 	dprintk("--> %s\n", __func__);
@@ -468,24 +477,27 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 	tbl = &session->fc_slot_table;
 
 	spin_lock(&tbl->slot_tbl_lock);
-	if (test_bit(NFS4CLNT_SESSION_SETUP, &session->clp->cl_state)) {
-		if (tbl->highest_used_slotid != -1) {
-			rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
-			spin_unlock(&tbl->slot_tbl_lock);
-			dprintk("<-- %s: Session reset: draining\n", __func__);
-			return -EAGAIN;
-		}
+	if (test_bit(NFS4CLNT_SESSION_DRAINING, &session->clp->cl_state) &&
+	    !rpc_task_has_priority(task, RPC_PRIORITY_PRIVILEGED)) {
+		/*
+		 * The state manager will wait until the slot table is empty.
+		 * Schedule the reset thread
+		 */
+		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
+		spin_unlock(&tbl->slot_tbl_lock);
+		dprintk("%s Schedule Session Reset\n", __func__);
+		return -EAGAIN;
+	}
 
-		/* The slot table is empty; start the reset thread */
-		dprintk("%s Session Reset\n", __func__);
+	if (!rpc_queue_empty(&tbl->slot_tbl_waitq) &&
+	    !rpc_task_has_priority(task, RPC_PRIORITY_PRIVILEGED)) {
+		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
 		spin_unlock(&tbl->slot_tbl_lock);
-		status = nfs4_recover_session(session);
-		if (status)
-			return status;
-		spin_lock(&tbl->slot_tbl_lock);
+		dprintk("%s enforce FIFO order\n", __func__);
+		return -EAGAIN;
 	}
 
-	slotid = nfs4_find_slot(tbl, task);
+	slotid = nfs4_find_slot(tbl);
 	if (slotid == NFS4_MAX_SLOT_TABLE) {
 		rpc_sleep_on(&tbl->slot_tbl_waitq, task, NULL);
 		spin_unlock(&tbl->slot_tbl_lock);
@@ -494,6 +506,7 @@ static int nfs41_setup_sequence(struct nfs4_session *session,
 	}
 	spin_unlock(&tbl->slot_tbl_lock);
 
+	rpc_task_set_priority(task, RPC_PRIORITY_NORMAL);
 	slot = tbl->slots + slotid;
 	args->sa_session = session;
 	args->sa_slotid = slotid;
@@ -527,7 +540,7 @@ int nfs4_setup_sequence(struct nfs_client *clp,
 		goto out;
 	ret = nfs41_setup_sequence(clp->cl_session, args, res, cache_reply,
 				   task);
-	if (ret != -EAGAIN) {
+	if (ret && ret != -EAGAIN) {
 		/* terminate rpc task */
 		task->tk_status = ret;
 		task->tk_action = NULL;
@@ -556,12 +569,17 @@ static void nfs41_call_sync_prepare(struct rpc_task *task, void *calldata)
 	rpc_call_start(task);
 }
 
+static void nfs41_call_priv_sync_prepare(struct rpc_task *task, void *calldata)
+{
+	rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
+	nfs41_call_sync_prepare(task, calldata);
+}
+
 static void nfs41_call_sync_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs41_call_sync_data *data = calldata;
 
 	nfs41_sequence_done(data->clp, data->seq_res, task->tk_status);
-	nfs41_sequence_free_slot(data->clp, data->seq_res);
 }
 
 struct rpc_call_ops nfs41_call_sync_ops = {
@@ -569,12 +587,18 @@ struct rpc_call_ops nfs41_call_sync_ops = {
 	.rpc_call_done = nfs41_call_sync_done,
 };
 
+struct rpc_call_ops nfs41_call_priv_sync_ops = {
+	.rpc_call_prepare = nfs41_call_priv_sync_prepare,
+	.rpc_call_done = nfs41_call_sync_done,
+};
+
 static int nfs4_call_sync_sequence(struct nfs_client *clp,
 				   struct rpc_clnt *clnt,
 				   struct rpc_message *msg,
 				   struct nfs4_sequence_args *args,
 				   struct nfs4_sequence_res *res,
-				   int cache_reply)
+				   int cache_reply,
+				   int privileged)
 {
 	int ret;
 	struct rpc_task *task;
@@ -592,6 +616,8 @@ static int nfs4_call_sync_sequence(struct nfs_client *clp,
 	};
 
 	res->sr_slotid = NFS4_MAX_SLOT_TABLE;
+	if (privileged)
+		task_setup.callback_ops = &nfs41_call_priv_sync_ops;
 	task = rpc_run_task(&task_setup);
 	if (IS_ERR(task))
 		ret = PTR_ERR(task);
@@ -609,7 +635,7 @@ int _nfs4_call_sync_session(struct nfs_server *server,
 			    int cache_reply)
 {
 	return nfs4_call_sync_sequence(server->nfs_client, server->client,
-				       msg, args, res, cache_reply);
+				       msg, args, res, cache_reply, 0);
 }
 
 #endif /* CONFIG_NFS_V4_1 */
@@ -637,15 +663,6 @@ static void nfs4_sequence_done(const struct nfs_server *server,
 #endif /* CONFIG_NFS_V4_1 */
 }
 
-/* no restart, therefore free slot here */
-static void nfs4_sequence_done_free_slot(const struct nfs_server *server,
-					 struct nfs4_sequence_res *res,
-					 int rpc_status)
-{
-	nfs4_sequence_done(server, res, rpc_status);
-	nfs4_sequence_free_slot(server->nfs_client, res);
-}
-
 static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo)
 {
 	struct nfs_inode *nfsi = NFS_I(dir);
@@ -720,9 +737,15 @@ static struct nfs4_opendata *nfs4_opendata_alloc(struct path *path,
 	p->o_arg.bitmask = server->attr_bitmask;
 	p->o_arg.claim = NFS4_OPEN_CLAIM_NULL;
 	if (flags & O_EXCL) {
-		u32 *s = (u32 *) p->o_arg.u.verifier.data;
-		s[0] = jiffies;
-		s[1] = current->pid;
+		if (nfs4_has_persistent_session(server->nfs_client)) {
+			/* GUARDED */
+			p->o_arg.u.attrs = &p->attrs;
+			memcpy(&p->attrs, attrs, sizeof(p->attrs));
+		} else { /* EXCLUSIVE4_1 */
+			u32 *s = (u32 *) p->o_arg.u.verifier.data;
+			s[0] = jiffies;
+			s[1] = current->pid;
+		}
 	} else if (flags & O_CREAT) {
 		p->o_arg.u.attrs = &p->attrs;
 		memcpy(&p->attrs, attrs, sizeof(p->attrs));
@@ -776,13 +799,16 @@ static int can_open_cached(struct nfs4_state *state, fmode_t mode, int open_mode
 		goto out;
 	switch (mode & (FMODE_READ|FMODE_WRITE)) {
 		case FMODE_READ:
-			ret |= test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0;
+			ret |= test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0
+				&& state->n_rdonly != 0;
 			break;
 		case FMODE_WRITE:
-			ret |= test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0;
+			ret |= test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0
+				&& state->n_wronly != 0;
 			break;
 		case FMODE_READ|FMODE_WRITE:
-			ret |= test_bit(NFS_O_RDWR_STATE, &state->flags) != 0;
+			ret |= test_bit(NFS_O_RDWR_STATE, &state->flags) != 0
+				&& state->n_rdwr != 0;
 	}
 out:
 	return ret;
@@ -1047,7 +1073,7 @@ static int nfs4_open_recover_helper(struct nfs4_opendata *opendata, fmode_t fmod
 	memset(&opendata->o_res, 0, sizeof(opendata->o_res));
 	memset(&opendata->c_res, 0, sizeof(opendata->c_res));
 	nfs4_init_opendata_res(opendata);
-	ret = _nfs4_proc_open(opendata);
+	ret = _nfs4_recover_proc_open(opendata);
 	if (ret != 0)
 		return ret; 
 	newstate = nfs4_opendata_to_nfs4_state(opendata);
@@ -1183,6 +1209,14 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, struct nfs4_state
 			case -ENOENT:
 			case -ESTALE:
 				goto out;
+			case -NFS4ERR_BADSESSION:
+			case -NFS4ERR_BADSLOT:
+			case -NFS4ERR_BAD_HIGH_SLOT:
+			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+			case -NFS4ERR_DEADSESSION:
+				nfs4_schedule_state_recovery(
+					server->nfs_client);
+				goto out;
 			case -NFS4ERR_STALE_CLIENTID:
 			case -NFS4ERR_STALE_STATEID:
 			case -NFS4ERR_EXPIRED:
@@ -1330,14 +1364,20 @@ out_no_action:
 
 }
 
+static void nfs4_recover_open_prepare(struct rpc_task *task, void *calldata)
+{
+	rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
+	nfs4_open_prepare(task, calldata);
+}
+
 static void nfs4_open_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_opendata *data = calldata;
 
 	data->rpc_status = task->tk_status;
 
-	nfs4_sequence_done_free_slot(data->o_arg.server, &data->o_res.seq_res,
-				     task->tk_status);
+	nfs4_sequence_done(data->o_arg.server, &data->o_res.seq_res,
+			task->tk_status);
 
 	if (RPC_ASSASSINATED(task))
 		return;
@@ -1388,10 +1428,13 @@ static const struct rpc_call_ops nfs4_open_ops = {
 	.rpc_release = nfs4_open_release,
 };
 
-/*
- * Note: On error, nfs4_proc_open will free the struct nfs4_opendata
- */
-static int _nfs4_proc_open(struct nfs4_opendata *data)
+static const struct rpc_call_ops nfs4_recover_open_ops = {
+	.rpc_call_prepare = nfs4_recover_open_prepare,
+	.rpc_call_done = nfs4_open_done,
+	.rpc_release = nfs4_open_release,
+};
+
+static int nfs4_run_open_task(struct nfs4_opendata *data, int isrecover)
 {
 	struct inode *dir = data->dir->d_inode;
 	struct nfs_server *server = NFS_SERVER(dir);
@@ -1418,21 +1461,57 @@ static int _nfs4_proc_open(struct nfs4_opendata *data)
 	data->rpc_done = 0;
 	data->rpc_status = 0;
 	data->cancelled = 0;
+	if (isrecover)
+		task_setup_data.callback_ops = &nfs4_recover_open_ops;
 	task = rpc_run_task(&task_setup_data);
-	if (IS_ERR(task))
-		return PTR_ERR(task);
-	status = nfs4_wait_for_completion_rpc_task(task);
-	if (status != 0) {
-		data->cancelled = 1;
-		smp_wmb();
-	} else
-		status = data->rpc_status;
-	rpc_put_task(task);
+        if (IS_ERR(task))
+                return PTR_ERR(task);
+        status = nfs4_wait_for_completion_rpc_task(task);
+        if (status != 0) {
+                data->cancelled = 1;
+                smp_wmb();
+        } else
+                status = data->rpc_status;
+        rpc_put_task(task);
+
+	return status;
+}
+
+static int _nfs4_recover_proc_open(struct nfs4_opendata *data)
+{
+	struct inode *dir = data->dir->d_inode;
+	struct nfs_openres *o_res = &data->o_res;
+        int status;
+
+	status = nfs4_run_open_task(data, 1);
 	if (status != 0 || !data->rpc_done)
 		return status;
 
-	if (o_res->fh.size == 0)
-		_nfs4_proc_lookup(dir, o_arg->name, &o_res->fh, o_res->f_attr);
+	nfs_refresh_inode(dir, o_res->dir_attr);
+
+	if (o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) {
+		status = _nfs4_proc_open_confirm(data);
+		if (status != 0)
+			return status;
+	}
+
+	return status;
+}
+
+/*
+ * Note: On error, nfs4_proc_open will free the struct nfs4_opendata
+ */
+static int _nfs4_proc_open(struct nfs4_opendata *data)
+{
+	struct inode *dir = data->dir->d_inode;
+	struct nfs_server *server = NFS_SERVER(dir);
+	struct nfs_openargs *o_arg = &data->o_arg;
+	struct nfs_openres *o_res = &data->o_res;
+	int status;
+
+	status = nfs4_run_open_task(data, 0);
+	if (status != 0 || !data->rpc_done)
+		return status;
 
 	if (o_arg->open_flags & O_CREAT) {
 		update_changeattr(dir, &o_res->cinfo);
@@ -1488,7 +1567,7 @@ static int _nfs4_open_expired(struct nfs_open_context *ctx, struct nfs4_state *s
 	return ret;
 }
 
-static inline int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state)
+static int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4_state *state)
 {
 	struct nfs_server *server = NFS_SERVER(state->inode);
 	struct nfs4_exception exception = { };
@@ -1496,10 +1575,16 @@ static inline int nfs4_do_open_expired(struct nfs_open_context *ctx, struct nfs4
 
 	do {
 		err = _nfs4_open_expired(ctx, state);
-		if (err != -NFS4ERR_DELAY)
-			break;
-		nfs4_handle_exception(server, err, &exception);
+		switch (err) {
+		default:
+			goto out;
+		case -NFS4ERR_GRACE:
+		case -NFS4ERR_DELAY:
+			nfs4_handle_exception(server, err, &exception);
+			err = 0;
+		}
 	} while (exception.retry);
+out:
 	return err;
 }
 
@@ -1712,6 +1797,18 @@ static void nfs4_free_closedata(void *data)
 	kfree(calldata);
 }
 
+static void nfs4_close_clear_stateid_flags(struct nfs4_state *state,
+		fmode_t fmode)
+{
+	spin_lock(&state->owner->so_lock);
+	if (!(fmode & FMODE_READ))
+		clear_bit(NFS_O_RDONLY_STATE, &state->flags);
+	if (!(fmode & FMODE_WRITE))
+		clear_bit(NFS_O_WRONLY_STATE, &state->flags);
+	clear_bit(NFS_O_RDWR_STATE, &state->flags);
+	spin_unlock(&state->owner->so_lock);
+}
+
 static void nfs4_close_done(struct rpc_task *task, void *data)
 {
 	struct nfs4_closedata *calldata = data;
@@ -1728,6 +1825,8 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 		case 0:
 			nfs_set_open_stateid(state, &calldata->res.stateid, 0);
 			renew_lease(server, calldata->timestamp);
+			nfs4_close_clear_stateid_flags(state,
+					calldata->arg.fmode);
 			break;
 		case -NFS4ERR_STALE_STATEID:
 		case -NFS4ERR_OLD_STATEID:
@@ -1736,12 +1835,10 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
 			if (calldata->arg.fmode == 0)
 				break;
 		default:
-			if (nfs4_async_handle_error(task, server, state) == -EAGAIN) {
-				nfs4_restart_rpc(task, server->nfs_client);
-				return;
-			}
+			if (nfs4_async_handle_error(task, server, state) == -EAGAIN)
+				rpc_restart_call_prepare(task);
 	}
-	nfs4_sequence_free_slot(server->nfs_client, &calldata->res.seq_res);
+	nfs_release_seqid(calldata->arg.seqid);
 	nfs_refresh_inode(calldata->inode, calldata->res.fattr);
 }
 
@@ -1749,38 +1846,39 @@ static void nfs4_close_prepare(struct rpc_task *task, void *data)
 {
 	struct nfs4_closedata *calldata = data;
 	struct nfs4_state *state = calldata->state;
-	int clear_rd, clear_wr, clear_rdwr;
+	int call_close = 0;
 
 	if (nfs_wait_on_sequence(calldata->arg.seqid, task) != 0)
 		return;
 
-	clear_rd = clear_wr = clear_rdwr = 0;
+	task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
+	calldata->arg.fmode = FMODE_READ|FMODE_WRITE;
 	spin_lock(&state->owner->so_lock);
 	/* Calculate the change in open mode */
 	if (state->n_rdwr == 0) {
 		if (state->n_rdonly == 0) {
-			clear_rd |= test_and_clear_bit(NFS_O_RDONLY_STATE, &state->flags);
-			clear_rdwr |= test_and_clear_bit(NFS_O_RDWR_STATE, &state->flags);
+			call_close |= test_bit(NFS_O_RDONLY_STATE, &state->flags);
+			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
+			calldata->arg.fmode &= ~FMODE_READ;
 		}
 		if (state->n_wronly == 0) {
-			clear_wr |= test_and_clear_bit(NFS_O_WRONLY_STATE, &state->flags);
-			clear_rdwr |= test_and_clear_bit(NFS_O_RDWR_STATE, &state->flags);
+			call_close |= test_bit(NFS_O_WRONLY_STATE, &state->flags);
+			call_close |= test_bit(NFS_O_RDWR_STATE, &state->flags);
+			calldata->arg.fmode &= ~FMODE_WRITE;
 		}
 	}
 	spin_unlock(&state->owner->so_lock);
-	if (!clear_rd && !clear_wr && !clear_rdwr) {
+
+	if (!call_close) {
 		/* Note: exit _without_ calling nfs4_close_done */
 		task->tk_action = NULL;
 		return;
 	}
+
+	if (calldata->arg.fmode == 0)
+		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE];
+
 	nfs_fattr_init(calldata->res.fattr);
-	if (test_bit(NFS_O_RDONLY_STATE, &state->flags) != 0) {
-		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
-		calldata->arg.fmode = FMODE_READ;
-	} else if (test_bit(NFS_O_WRONLY_STATE, &state->flags) != 0) {
-		task->tk_msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE];
-		calldata->arg.fmode = FMODE_WRITE;
-	}
 	calldata->timestamp = jiffies;
 	if (nfs4_setup_sequence((NFS_SERVER(calldata->inode))->nfs_client,
 				&calldata->arg.seq_args, &calldata->res.seq_res,
@@ -1832,8 +1930,6 @@ int nfs4_do_close(struct path *path, struct nfs4_state *state, int wait)
 	calldata->state = state;
 	calldata->arg.fh = NFS_FH(state->inode);
 	calldata->arg.stateid = &state->open_stateid;
-	if (nfs4_has_session(server->nfs_client))
-		memset(calldata->arg.stateid->data, 0, 4);    /* clear seqid */
 	/* Serialization for the sequence id */
 	calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid);
 	if (calldata->arg.seqid == NULL)
@@ -1981,7 +2077,7 @@ out_drop:
 	return 0;
 }
 
-void nfs4_close_context(struct nfs_open_context *ctx, int is_sync)
+static void nfs4_close_context(struct nfs_open_context *ctx, int is_sync)
 {
 	if (ctx->state == NULL)
 		return;
@@ -2532,7 +2628,6 @@ static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir)
 	nfs4_sequence_done(res->server, &res->seq_res, task->tk_status);
 	if (nfs4_async_handle_error(task, res->server, NULL) == -EAGAIN)
 		return 0;
-	nfs4_sequence_free_slot(res->server->nfs_client, &res->seq_res);
 	update_changeattr(dir, &res->cinfo);
 	nfs_post_op_update_inode(dir, &res->dir_attr);
 	return 1;
@@ -2971,11 +3066,10 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data)
 
 	dprintk("--> %s\n", __func__);
 
-	/* nfs4_sequence_free_slot called in the read rpc_call_done */
 	nfs4_sequence_done(server, &data->res.seq_res, task->tk_status);
 
 	if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) {
-		nfs4_restart_rpc(task, server->nfs_client);
+		nfs_restart_rpc(task, server->nfs_client);
 		return -EAGAIN;
 	}
 
@@ -2995,12 +3089,11 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data)
 {
 	struct inode *inode = data->inode;
 	
-	/* slot is freed in nfs_writeback_done */
 	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
 			   task->tk_status);
 
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) {
-		nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
+		nfs_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
 		return -EAGAIN;
 	}
 	if (task->tk_status >= 0) {
@@ -3028,11 +3121,9 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data)
 	nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res,
 			   task->tk_status);
 	if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) {
-		nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
+		nfs_restart_rpc(task, NFS_SERVER(inode)->nfs_client);
 		return -EAGAIN;
 	}
-	nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client,
-				&data->res.seq_res);
 	nfs_refresh_inode(inode, data->res.fattr);
 	return 0;
 }
@@ -3350,7 +3441,7 @@ _nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server,
 		case -NFS4ERR_SEQ_MISORDERED:
 			dprintk("%s ERROR %d, Reset session\n", __func__,
 				task->tk_status);
-			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			nfs4_schedule_state_recovery(clp);
 			task->tk_status = 0;
 			return -EAGAIN;
 #endif /* CONFIG_NFS_V4_1 */
@@ -3483,12 +3574,23 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_delegreturndata *data = calldata;
 
-	nfs4_sequence_done_free_slot(data->res.server, &data->res.seq_res,
-				     task->tk_status);
+	nfs4_sequence_done(data->res.server, &data->res.seq_res,
+			task->tk_status);
 
-	data->rpc_status = task->tk_status;
-	if (data->rpc_status == 0)
+	switch (task->tk_status) {
+	case -NFS4ERR_STALE_STATEID:
+	case -NFS4ERR_EXPIRED:
+	case 0:
 		renew_lease(data->res.server, data->timestamp);
+		break;
+	default:
+		if (nfs4_async_handle_error(task, data->res.server, NULL) ==
+				-EAGAIN) {
+			nfs_restart_rpc(task, data->res.server->nfs_client);
+			return;
+		}
+	}
+	data->rpc_status = task->tk_status;
 }
 
 static void nfs4_delegreturn_release(void *calldata)
@@ -3741,11 +3843,9 @@ static void nfs4_locku_done(struct rpc_task *task, void *data)
 			break;
 		default:
 			if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN)
-				nfs4_restart_rpc(task,
-						calldata->server->nfs_client);
+				nfs_restart_rpc(task,
+						 calldata->server->nfs_client);
 	}
-	nfs4_sequence_free_slot(calldata->server->nfs_client,
-				&calldata->res.seq_res);
 }
 
 static void nfs4_locku_prepare(struct rpc_task *task, void *data)
@@ -3921,14 +4021,20 @@ static void nfs4_lock_prepare(struct rpc_task *task, void *calldata)
 	dprintk("%s: done!, ret = %d\n", __func__, data->rpc_status);
 }
 
+static void nfs4_recover_lock_prepare(struct rpc_task *task, void *calldata)
+{
+	rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
+	nfs4_lock_prepare(task, calldata);
+}
+
 static void nfs4_lock_done(struct rpc_task *task, void *calldata)
 {
 	struct nfs4_lockdata *data = calldata;
 
 	dprintk("%s: begin!\n", __func__);
 
-	nfs4_sequence_done_free_slot(data->server, &data->res.seq_res,
-				     task->tk_status);
+	nfs4_sequence_done(data->server, &data->res.seq_res,
+			task->tk_status);
 
 	data->rpc_status = task->tk_status;
 	if (RPC_ASSASSINATED(task))
@@ -3976,7 +4082,13 @@ static const struct rpc_call_ops nfs4_lock_ops = {
 	.rpc_release = nfs4_lock_release,
 };
 
-static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *fl, int reclaim)
+static const struct rpc_call_ops nfs4_recover_lock_ops = {
+	.rpc_call_prepare = nfs4_recover_lock_prepare,
+	.rpc_call_done = nfs4_lock_done,
+	.rpc_release = nfs4_lock_release,
+};
+
+static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *fl, int recovery_type)
 {
 	struct nfs4_lockdata *data;
 	struct rpc_task *task;
@@ -4000,8 +4112,11 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
 		return -ENOMEM;
 	if (IS_SETLKW(cmd))
 		data->arg.block = 1;
-	if (reclaim != 0)
-		data->arg.reclaim = 1;
+	if (recovery_type > NFS_LOCK_NEW) {
+		if (recovery_type == NFS_LOCK_RECLAIM)
+			data->arg.reclaim = NFS_LOCK_RECLAIM;
+		task_setup_data.callback_ops = &nfs4_recover_lock_ops;
+	}
 	msg.rpc_argp = &data->arg,
 	msg.rpc_resp = &data->res,
 	task_setup_data.callback_data = data;
@@ -4028,7 +4143,7 @@ static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request
 		/* Cache the lock if possible... */
 		if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0)
 			return 0;
-		err = _nfs4_do_setlk(state, F_SETLK, request, 1);
+		err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_RECLAIM);
 		if (err != -NFS4ERR_DELAY)
 			break;
 		nfs4_handle_exception(server, err, &exception);
@@ -4048,11 +4163,17 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request
 	do {
 		if (test_bit(NFS_DELEGATED_STATE, &state->flags) != 0)
 			return 0;
-		err = _nfs4_do_setlk(state, F_SETLK, request, 0);
-		if (err != -NFS4ERR_DELAY)
-			break;
-		nfs4_handle_exception(server, err, &exception);
+		err = _nfs4_do_setlk(state, F_SETLK, request, NFS_LOCK_EXPIRED);
+		switch (err) {
+		default:
+			goto out;
+		case -NFS4ERR_GRACE:
+		case -NFS4ERR_DELAY:
+			nfs4_handle_exception(server, err, &exception);
+			err = 0;
+		}
 	} while (exception.retry);
+out:
 	return err;
 }
 
@@ -4078,7 +4199,7 @@ static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock
 		status = do_vfs_lock(request->fl_file, request);
 		goto out_unlock;
 	}
-	status = _nfs4_do_setlk(state, cmd, request, 0);
+	status = _nfs4_do_setlk(state, cmd, request, NFS_LOCK_NEW);
 	if (status != 0)
 		goto out_unlock;
 	/* Note: we always want to sleep here! */
@@ -4161,7 +4282,7 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
 	if (err != 0)
 		goto out;
 	do {
-		err = _nfs4_do_setlk(state, F_SETLK, fl, 0);
+		err = _nfs4_do_setlk(state, F_SETLK, fl, NFS_LOCK_NEW);
 		switch (err) {
 			default:
 				printk(KERN_ERR "%s: unhandled error %d.\n",
@@ -4172,6 +4293,11 @@ int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl)
 			case -NFS4ERR_EXPIRED:
 			case -NFS4ERR_STALE_CLIENTID:
 			case -NFS4ERR_STALE_STATEID:
+			case -NFS4ERR_BADSESSION:
+			case -NFS4ERR_BADSLOT:
+			case -NFS4ERR_BAD_HIGH_SLOT:
+			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
+			case -NFS4ERR_DEADSESSION:
 				nfs4_schedule_state_recovery(server->nfs_client);
 				goto out;
 			case -ERESTARTSYS:
@@ -4296,7 +4422,7 @@ int nfs4_proc_fs_locations(struct inode *dir, const struct qstr *name,
  * NFS4ERR_BADSESSION in the sequence operation, and will therefore
  * be in some phase of session reset.
  */
-static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
+int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 {
 	nfs4_verifier verifier;
 	struct nfs41_exchange_id_args args = {
@@ -4318,6 +4444,9 @@ static int nfs4_proc_exchange_id(struct nfs_client *clp, struct rpc_cred *cred)
 	dprintk("--> %s\n", __func__);
 	BUG_ON(clp == NULL);
 
+	/* Remove server-only flags */
+	args.flags &= ~EXCHGID4_FLAG_CONFIRMED_R;
+
 	p = (u32 *)verifier.data;
 	*p++ = htonl((u32)clp->cl_boot_time.tv_sec);
 	*p = htonl((u32)clp->cl_boot_time.tv_nsec);
@@ -4361,11 +4490,12 @@ static void nfs4_get_lease_time_prepare(struct rpc_task *task,
 			(struct nfs4_get_lease_time_data *)calldata;
 
 	dprintk("--> %s\n", __func__);
+	rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
 	/* just setup sequence, do not trigger session recovery
 	   since we're invoked within one */
 	ret = nfs41_setup_sequence(data->clp->cl_session,
-					&data->args->la_seq_args,
-					&data->res->lr_seq_res, 0, task);
+				   &data->args->la_seq_args,
+				   &data->res->lr_seq_res, 0, task);
 
 	BUG_ON(ret == -EAGAIN);
 	rpc_call_start(task);
@@ -4389,10 +4519,9 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata)
 		dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status);
 		rpc_delay(task, NFS4_POLL_RETRY_MIN);
 		task->tk_status = 0;
-		nfs4_restart_rpc(task, data->clp);
+		nfs_restart_rpc(task, data->clp);
 		return;
 	}
-	nfs41_sequence_free_slot(data->clp, &data->res->lr_seq_res);
 	dprintk("<-- %s\n", __func__);
 }
 
@@ -4465,7 +4594,6 @@ static int nfs4_reset_slot_table(struct nfs4_slot_table *tbl, int max_slots,
 	spin_lock(&tbl->slot_tbl_lock);
 	for (i = 0; i < max_slots; ++i)
 		tbl->slots[i].seq_nr = ivalue;
-	tbl->highest_used_slotid = -1;
 	spin_unlock(&tbl->slot_tbl_lock);
 	dprintk("%s: tbl=%p slots=%p max_slots=%d\n", __func__,
 		tbl, tbl->slots, tbl->max_slots);
@@ -4515,7 +4643,6 @@ static void nfs4_destroy_slot_tables(struct nfs4_session *session)
 static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
 		int max_slots, int ivalue)
 {
-	int i;
 	struct nfs4_slot *slot;
 	int ret = -ENOMEM;
 
@@ -4526,18 +4653,9 @@ static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
 	slot = kcalloc(max_slots, sizeof(struct nfs4_slot), GFP_KERNEL);
 	if (!slot)
 		goto out;
-	for (i = 0; i < max_slots; ++i)
-		slot[i].seq_nr = ivalue;
 	ret = 0;
 
 	spin_lock(&tbl->slot_tbl_lock);
-	if (tbl->slots != NULL) {
-		spin_unlock(&tbl->slot_tbl_lock);
-		dprintk("%s: slot table already initialized. tbl=%p slots=%p\n",
-			__func__, tbl, tbl->slots);
-		WARN_ON(1);
-		goto out_free;
-	}
 	tbl->max_slots = max_slots;
 	tbl->slots = slot;
 	tbl->highest_used_slotid = -1;  /* no slot is currently used */
@@ -4547,10 +4665,6 @@ static int nfs4_init_slot_table(struct nfs4_slot_table *tbl,
 out:
 	dprintk("<-- %s: return %d\n", __func__, ret);
 	return ret;
-
-out_free:
-	kfree(slot);
-	goto out;
 }
 
 /*
@@ -4558,17 +4672,24 @@ out_free:
  */
 static int nfs4_init_slot_tables(struct nfs4_session *session)
 {
-	int status;
+	struct nfs4_slot_table *tbl;
+	int status = 0;
 
-	status = nfs4_init_slot_table(&session->fc_slot_table,
-			session->fc_attrs.max_reqs, 1);
-	if (status)
-		return status;
+	tbl = &session->fc_slot_table;
+	if (tbl->slots == NULL) {
+		status = nfs4_init_slot_table(tbl,
+				session->fc_attrs.max_reqs, 1);
+		if (status)
+			return status;
+	}
 
-	status = nfs4_init_slot_table(&session->bc_slot_table,
-			session->bc_attrs.max_reqs, 0);
-	if (status)
-		nfs4_destroy_slot_tables(session);
+	tbl = &session->bc_slot_table;
+	if (tbl->slots == NULL) {
+		status = nfs4_init_slot_table(tbl,
+				session->bc_attrs.max_reqs, 0);
+		if (status)
+			nfs4_destroy_slot_tables(session);
+	}
 
 	return status;
 }
@@ -4582,7 +4703,6 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 	if (!session)
 		return NULL;
 
-	set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
 	/*
 	 * The create session reply races with the server back
 	 * channel probe. Mark the client NFS_CS_SESSION_INITING
@@ -4590,12 +4710,15 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp)
 	 * nfs_client struct
 	 */
 	clp->cl_cons_state = NFS_CS_SESSION_INITING;
+	init_completion(&session->complete);
 
 	tbl = &session->fc_slot_table;
+	tbl->highest_used_slotid = -1;
 	spin_lock_init(&tbl->slot_tbl_lock);
-	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "ForeChannel Slot table");
+	rpc_init_priority_wait_queue(&tbl->slot_tbl_waitq, "ForeChannel Slot table");
 
 	tbl = &session->bc_slot_table;
+	tbl->highest_used_slotid = -1;
 	spin_lock_init(&tbl->slot_tbl_lock);
 	rpc_init_wait_queue(&tbl->slot_tbl_waitq, "BackChannel Slot table");
 
@@ -4747,11 +4870,10 @@ static int _nfs4_proc_create_session(struct nfs_client *clp)
  * It is the responsibility of the caller to verify the session is
  * expired before calling this routine.
  */
-int nfs4_proc_create_session(struct nfs_client *clp, int reset)
+int nfs4_proc_create_session(struct nfs_client *clp)
 {
 	int status;
 	unsigned *ptr;
-	struct nfs_fsinfo fsinfo;
 	struct nfs4_session *session = clp->cl_session;
 
 	dprintk("--> %s clp=%p session=%p\n", __func__, clp, session);
@@ -4760,35 +4882,19 @@ int nfs4_proc_create_session(struct nfs_client *clp, int reset)
 	if (status)
 		goto out;
 
-	/* Init or reset the fore channel */
-	if (reset)
-		status = nfs4_reset_slot_tables(session);
-	else
-		status = nfs4_init_slot_tables(session);
-	dprintk("fore channel slot table initialization returned %d\n", status);
+	/* Init and reset the fore channel */
+	status = nfs4_init_slot_tables(session);
+	dprintk("slot table initialization returned %d\n", status);
+	if (status)
+		goto out;
+	status = nfs4_reset_slot_tables(session);
+	dprintk("slot table reset returned %d\n", status);
 	if (status)
 		goto out;
 
 	ptr = (unsigned *)&session->sess_id.data[0];
 	dprintk("%s client>seqid %d sessionid %u:%u:%u:%u\n", __func__,
 		clp->cl_seqid, ptr[0], ptr[1], ptr[2], ptr[3]);
-
-	if (reset)
-		/* Lease time is aleady set */
-		goto out;
-
-	/* Get the lease time */
-	status = nfs4_proc_get_lease_time(clp, &fsinfo);
-	if (status == 0) {
-		/* Update lease time and schedule renewal */
-		spin_lock(&clp->cl_lock);
-		clp->cl_lease_time = fsinfo.lease_time * HZ;
-		clp->cl_last_renewal = jiffies;
-		clear_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
-		spin_unlock(&clp->cl_lock);
-
-		nfs4_schedule_state_renewal(clp);
-	}
 out:
 	dprintk("<-- %s\n", __func__);
 	return status;
@@ -4827,13 +4933,24 @@ int nfs4_proc_destroy_session(struct nfs4_session *session)
 int nfs4_init_session(struct nfs_server *server)
 {
 	struct nfs_client *clp = server->nfs_client;
+	struct nfs4_session *session;
+	unsigned int rsize, wsize;
 	int ret;
 
 	if (!nfs4_has_session(clp))
 		return 0;
 
-	clp->cl_session->fc_attrs.max_rqst_sz = server->wsize;
-	clp->cl_session->fc_attrs.max_resp_sz = server->rsize;
+	rsize = server->rsize;
+	if (rsize == 0)
+		rsize = NFS_MAX_FILE_IO_SIZE;
+	wsize = server->wsize;
+	if (wsize == 0)
+		wsize = NFS_MAX_FILE_IO_SIZE;
+
+	session = clp->cl_session;
+	session->fc_attrs.max_rqst_sz = wsize + nfs41_maxwrite_overhead;
+	session->fc_attrs.max_resp_sz = rsize + nfs41_maxread_overhead;
+
 	ret = nfs4_recover_expired_lease(server);
 	if (!ret)
 		ret = nfs4_check_client_ready(clp);
@@ -4858,7 +4975,7 @@ static int nfs4_proc_sequence(struct nfs_client *clp, struct rpc_cred *cred)
 	args.sa_cache_this = 0;
 
 	return nfs4_call_sync_sequence(clp, clp->cl_rpcclient, &msg, &args,
-				       &res, 0);
+				       &res, args.sa_cache_this, 1);
 }
 
 void nfs41_sequence_call_done(struct rpc_task *task, void *data)
@@ -4872,11 +4989,10 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data)
 
 		if (_nfs4_async_handle_error(task, NULL, clp, NULL)
 								== -EAGAIN) {
-			nfs4_restart_rpc(task, clp);
+			nfs_restart_rpc(task, clp);
 			return;
 		}
 	}
-	nfs41_sequence_free_slot(clp, task->tk_msg.rpc_resp);
 	dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred);
 
 	kfree(task->tk_msg.rpc_argp);
@@ -4931,6 +5047,110 @@ static int nfs41_proc_async_sequence(struct nfs_client *clp,
 			      &nfs41_sequence_ops, (void *)clp);
 }
 
+struct nfs4_reclaim_complete_data {
+	struct nfs_client *clp;
+	struct nfs41_reclaim_complete_args arg;
+	struct nfs41_reclaim_complete_res res;
+};
+
+static void nfs4_reclaim_complete_prepare(struct rpc_task *task, void *data)
+{
+	struct nfs4_reclaim_complete_data *calldata = data;
+
+	rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
+	if (nfs4_setup_sequence(calldata->clp, &calldata->arg.seq_args,
+				&calldata->res.seq_res, 0, task))
+		return;
+
+	rpc_call_start(task);
+}
+
+static void nfs4_reclaim_complete_done(struct rpc_task *task, void *data)
+{
+	struct nfs4_reclaim_complete_data *calldata = data;
+	struct nfs_client *clp = calldata->clp;
+	struct nfs4_sequence_res *res = &calldata->res.seq_res;
+
+	dprintk("--> %s\n", __func__);
+	nfs41_sequence_done(clp, res, task->tk_status);
+	switch (task->tk_status) {
+	case 0:
+	case -NFS4ERR_COMPLETE_ALREADY:
+		break;
+	case -NFS4ERR_BADSESSION:
+	case -NFS4ERR_DEADSESSION:
+		/*
+		 * Handle the session error, but do not retry the operation, as
+		 * we have no way of telling whether the clientid had to be
+		 * reset before we got our reply.  If reset, a new wave of
+		 * reclaim operations will follow, containing their own reclaim
+		 * complete.  We don't want our retry to get on the way of
+		 * recovery by incorrectly indicating to the server that we're
+		 * done reclaiming state since the process had to be restarted.
+		 */
+		_nfs4_async_handle_error(task, NULL, clp, NULL);
+		break;
+	default:
+		if (_nfs4_async_handle_error(
+				task, NULL, clp, NULL) == -EAGAIN) {
+			rpc_restart_call_prepare(task);
+			return;
+		}
+	}
+
+	dprintk("<-- %s\n", __func__);
+}
+
+static void nfs4_free_reclaim_complete_data(void *data)
+{
+	struct nfs4_reclaim_complete_data *calldata = data;
+
+	kfree(calldata);
+}
+
+static const struct rpc_call_ops nfs4_reclaim_complete_call_ops = {
+	.rpc_call_prepare = nfs4_reclaim_complete_prepare,
+	.rpc_call_done = nfs4_reclaim_complete_done,
+	.rpc_release = nfs4_free_reclaim_complete_data,
+};
+
+/*
+ * Issue a global reclaim complete.
+ */
+static int nfs41_proc_reclaim_complete(struct nfs_client *clp)
+{
+	struct nfs4_reclaim_complete_data *calldata;
+	struct rpc_task *task;
+	struct rpc_message msg = {
+		.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RECLAIM_COMPLETE],
+	};
+	struct rpc_task_setup task_setup_data = {
+		.rpc_client = clp->cl_rpcclient,
+		.rpc_message = &msg,
+		.callback_ops = &nfs4_reclaim_complete_call_ops,
+		.flags = RPC_TASK_ASYNC,
+	};
+	int status = -ENOMEM;
+
+	dprintk("--> %s\n", __func__);
+	calldata = kzalloc(sizeof(*calldata), GFP_KERNEL);
+	if (calldata == NULL)
+		goto out;
+	calldata->clp = clp;
+	calldata->arg.one_fs = 0;
+	calldata->res.seq_res.sr_slotid = NFS4_MAX_SLOT_TABLE;
+
+	msg.rpc_argp = &calldata->arg;
+	msg.rpc_resp = &calldata->res;
+	task_setup_data.callback_data = calldata;
+	task = rpc_run_task(&task_setup_data);
+	if (IS_ERR(task))
+		status = PTR_ERR(task);
+	rpc_put_task(task);
+out:
+	dprintk("<-- %s status=%d\n", __func__, status);
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 struct nfs4_state_recovery_ops nfs40_reboot_recovery_ops = {
@@ -4948,8 +5168,9 @@ struct nfs4_state_recovery_ops nfs41_reboot_recovery_ops = {
 	.state_flag_bit	= NFS_STATE_RECLAIM_REBOOT,
 	.recover_open	= nfs4_open_reclaim,
 	.recover_lock	= nfs4_lock_reclaim,
-	.establish_clid = nfs4_proc_exchange_id,
+	.establish_clid = nfs41_init_clientid,
 	.get_clid_cred	= nfs4_get_exchange_id_cred,
+	.reclaim_complete = nfs41_proc_reclaim_complete,
 };
 #endif /* CONFIG_NFS_V4_1 */
 
@@ -4968,7 +5189,7 @@ struct nfs4_state_recovery_ops nfs41_nograce_recovery_ops = {
 	.state_flag_bit	= NFS_STATE_RECLAIM_NOGRACE,
 	.recover_open	= nfs4_open_expired,
 	.recover_lock	= nfs4_lock_expired,
-	.establish_clid = nfs4_proc_exchange_id,
+	.establish_clid = nfs41_init_clientid,
 	.get_clid_cred	= nfs4_get_exchange_id_cred,
 };
 #endif /* CONFIG_NFS_V4_1 */
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 2ef4fecf3984..6d263ed79e92 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -116,6 +116,79 @@ struct rpc_cred *nfs4_get_renew_cred_locked(struct nfs_client *clp)
 
 #if defined(CONFIG_NFS_V4_1)
 
+static int nfs41_setup_state_renewal(struct nfs_client *clp)
+{
+	int status;
+	struct nfs_fsinfo fsinfo;
+
+	status = nfs4_proc_get_lease_time(clp, &fsinfo);
+	if (status == 0) {
+		/* Update lease time and schedule renewal */
+		spin_lock(&clp->cl_lock);
+		clp->cl_lease_time = fsinfo.lease_time * HZ;
+		clp->cl_last_renewal = jiffies;
+		spin_unlock(&clp->cl_lock);
+
+		nfs4_schedule_state_renewal(clp);
+	}
+
+	return status;
+}
+
+static void nfs4_end_drain_session(struct nfs_client *clp)
+{
+	struct nfs4_session *ses = clp->cl_session;
+	int max_slots;
+
+	if (test_and_clear_bit(NFS4CLNT_SESSION_DRAINING, &clp->cl_state)) {
+		spin_lock(&ses->fc_slot_table.slot_tbl_lock);
+		max_slots = ses->fc_slot_table.max_slots;
+		while (max_slots--) {
+			struct rpc_task *task;
+
+			task = rpc_wake_up_next(&ses->fc_slot_table.
+						slot_tbl_waitq);
+			if (!task)
+				break;
+			rpc_task_set_priority(task, RPC_PRIORITY_PRIVILEGED);
+		}
+		spin_unlock(&ses->fc_slot_table.slot_tbl_lock);
+	}
+}
+
+static int nfs4_begin_drain_session(struct nfs_client *clp)
+{
+	struct nfs4_session *ses = clp->cl_session;
+	struct nfs4_slot_table *tbl = &ses->fc_slot_table;
+
+	spin_lock(&tbl->slot_tbl_lock);
+	set_bit(NFS4CLNT_SESSION_DRAINING, &clp->cl_state);
+	if (tbl->highest_used_slotid != -1) {
+		INIT_COMPLETION(ses->complete);
+		spin_unlock(&tbl->slot_tbl_lock);
+		return wait_for_completion_interruptible(&ses->complete);
+	}
+	spin_unlock(&tbl->slot_tbl_lock);
+	return 0;
+}
+
+int nfs41_init_clientid(struct nfs_client *clp, struct rpc_cred *cred)
+{
+	int status;
+
+	nfs4_begin_drain_session(clp);
+	status = nfs4_proc_exchange_id(clp, cred);
+	if (status != 0)
+		goto out;
+	status = nfs4_proc_create_session(clp);
+	if (status != 0)
+		goto out;
+	nfs41_setup_state_renewal(clp);
+	nfs_mark_client_ready(clp, NFS_CS_READY);
+out:
+	return status;
+}
+
 struct rpc_cred *nfs4_get_exchange_id_cred(struct nfs_client *clp)
 {
 	struct rpc_cred *cred;
@@ -693,16 +766,21 @@ struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter)
 	return new;
 }
 
-void nfs_free_seqid(struct nfs_seqid *seqid)
+void nfs_release_seqid(struct nfs_seqid *seqid)
 {
 	if (!list_empty(&seqid->list)) {
 		struct rpc_sequence *sequence = seqid->sequence->sequence;
 
 		spin_lock(&sequence->lock);
-		list_del(&seqid->list);
+		list_del_init(&seqid->list);
 		spin_unlock(&sequence->lock);
 		rpc_wake_up(&sequence->wait);
 	}
+}
+
+void nfs_free_seqid(struct nfs_seqid *seqid)
+{
+	nfs_release_seqid(seqid);
 	kfree(seqid);
 }
 
@@ -877,6 +955,10 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_
 			case -NFS4ERR_EXPIRED:
 			case -NFS4ERR_NO_GRACE:
 			case -NFS4ERR_STALE_CLIENTID:
+			case -NFS4ERR_BADSESSION:
+			case -NFS4ERR_BADSLOT:
+			case -NFS4ERR_BAD_HIGH_SLOT:
+			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 				goto out;
 			default:
 				printk(KERN_ERR "%s: unhandled error %d. Zeroing state\n",
@@ -959,6 +1041,10 @@ restart:
 			case -NFS4ERR_NO_GRACE:
 				nfs4_state_mark_reclaim_nograce(sp->so_client, state);
 			case -NFS4ERR_STALE_CLIENTID:
+			case -NFS4ERR_BADSESSION:
+			case -NFS4ERR_BADSLOT:
+			case -NFS4ERR_BAD_HIGH_SLOT:
+			case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 				goto out_err;
 		}
 		nfs4_put_open_state(state);
@@ -1011,6 +1097,14 @@ static void nfs4_state_start_reclaim_reboot(struct nfs_client *clp)
 	nfs4_state_mark_reclaim_helper(clp, nfs4_state_mark_reclaim_reboot);
 }
 
+static void nfs4_reclaim_complete(struct nfs_client *clp,
+				 const struct nfs4_state_recovery_ops *ops)
+{
+	/* Notify the server we're done reclaiming our state */
+	if (ops->reclaim_complete)
+		(void)ops->reclaim_complete(clp);
+}
+
 static void nfs4_state_end_reclaim_reboot(struct nfs_client *clp)
 {
 	struct nfs4_state_owner *sp;
@@ -1020,6 +1114,9 @@ static void nfs4_state_end_reclaim_reboot(struct nfs_client *clp)
 	if (!test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state))
 		return;
 
+	nfs4_reclaim_complete(clp,
+		nfs4_reboot_recovery_ops[clp->cl_minorversion]);
+
 	for (pos = rb_first(&clp->cl_state_owners); pos != NULL; pos = rb_next(pos)) {
 		sp = rb_entry(pos, struct nfs4_state_owner, so_client_node);
 		spin_lock(&sp->so_lock);
@@ -1046,25 +1143,25 @@ static void nfs4_state_start_reclaim_nograce(struct nfs_client *clp)
 	nfs4_state_mark_reclaim_helper(clp, nfs4_state_mark_reclaim_nograce);
 }
 
-static void nfs4_state_end_reclaim_nograce(struct nfs_client *clp)
-{
-	clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state);
-}
-
-static void nfs4_recovery_handle_error(struct nfs_client *clp, int error)
+static int nfs4_recovery_handle_error(struct nfs_client *clp, int error)
 {
 	switch (error) {
 		case -NFS4ERR_CB_PATH_DOWN:
 			nfs_handle_cb_pathdown(clp);
-			break;
+			return 0;
+		case -NFS4ERR_NO_GRACE:
+			nfs4_state_end_reclaim_reboot(clp);
+			return 0;
 		case -NFS4ERR_STALE_CLIENTID:
 		case -NFS4ERR_LEASE_MOVED:
 			set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+			nfs4_state_end_reclaim_reboot(clp);
 			nfs4_state_start_reclaim_reboot(clp);
 			break;
 		case -NFS4ERR_EXPIRED:
 			set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
 			nfs4_state_start_reclaim_nograce(clp);
+			break;
 		case -NFS4ERR_BADSESSION:
 		case -NFS4ERR_BADSLOT:
 		case -NFS4ERR_BAD_HIGH_SLOT:
@@ -1072,8 +1169,11 @@ static void nfs4_recovery_handle_error(struct nfs_client *clp, int error)
 		case -NFS4ERR_CONN_NOT_BOUND_TO_SESSION:
 		case -NFS4ERR_SEQ_FALSE_RETRY:
 		case -NFS4ERR_SEQ_MISORDERED:
-			set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
+			set_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state);
+			/* Zero session reset errors */
+			return 0;
 	}
+	return error;
 }
 
 static int nfs4_do_reclaim(struct nfs_client *clp, const struct nfs4_state_recovery_ops *ops)
@@ -1093,8 +1193,7 @@ restart:
 		if (status < 0) {
 			set_bit(ops->owner_flag_bit, &sp->so_flags);
 			nfs4_put_state_owner(sp);
-			nfs4_recovery_handle_error(clp, status);
-			return status;
+			return nfs4_recovery_handle_error(clp, status);
 		}
 		nfs4_put_state_owner(sp);
 		goto restart;
@@ -1124,8 +1223,7 @@ static int nfs4_check_lease(struct nfs_client *clp)
 	status = ops->renew_lease(clp, cred);
 	put_rpccred(cred);
 out:
-	nfs4_recovery_handle_error(clp, status);
-	return status;
+	return nfs4_recovery_handle_error(clp, status);
 }
 
 static int nfs4_reclaim_lease(struct nfs_client *clp)
@@ -1151,55 +1249,59 @@ static int nfs4_reclaim_lease(struct nfs_client *clp)
 }
 
 #ifdef CONFIG_NFS_V4_1
-static void nfs4_session_recovery_handle_error(struct nfs_client *clp, int err)
+void nfs41_handle_sequence_flag_errors(struct nfs_client *clp, u32 flags)
 {
-	switch (err) {
-	case -NFS4ERR_STALE_CLIENTID:
+	if (!flags)
+		return;
+	else if (flags & SEQ4_STATUS_RESTART_RECLAIM_NEEDED) {
 		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
-		set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
-	}
+		nfs4_state_start_reclaim_reboot(clp);
+		nfs4_schedule_state_recovery(clp);
+	} else if (flags & (SEQ4_STATUS_EXPIRED_ALL_STATE_REVOKED |
+			    SEQ4_STATUS_EXPIRED_SOME_STATE_REVOKED |
+			    SEQ4_STATUS_ADMIN_STATE_REVOKED |
+			    SEQ4_STATUS_RECALLABLE_STATE_REVOKED |
+			    SEQ4_STATUS_LEASE_MOVED)) {
+		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
+		nfs4_state_start_reclaim_nograce(clp);
+		nfs4_schedule_state_recovery(clp);
+	} else if (flags & (SEQ4_STATUS_CB_PATH_DOWN |
+			    SEQ4_STATUS_BACKCHANNEL_FAULT |
+			    SEQ4_STATUS_CB_PATH_DOWN_SESSION))
+		nfs_expire_all_delegations(clp);
 }
 
 static int nfs4_reset_session(struct nfs_client *clp)
 {
 	int status;
 
+	nfs4_begin_drain_session(clp);
 	status = nfs4_proc_destroy_session(clp->cl_session);
 	if (status && status != -NFS4ERR_BADSESSION &&
 	    status != -NFS4ERR_DEADSESSION) {
-		nfs4_session_recovery_handle_error(clp, status);
+		status = nfs4_recovery_handle_error(clp, status);
 		goto out;
 	}
 
 	memset(clp->cl_session->sess_id.data, 0, NFS4_MAX_SESSIONID_LEN);
-	status = nfs4_proc_create_session(clp, 1);
+	status = nfs4_proc_create_session(clp);
 	if (status)
-		nfs4_session_recovery_handle_error(clp, status);
-		/* fall through*/
-out:
-	/* Wake up the next rpc task even on error */
-	rpc_wake_up_next(&clp->cl_session->fc_slot_table.slot_tbl_waitq);
-	return status;
-}
+		status = nfs4_recovery_handle_error(clp, status);
 
-static int nfs4_initialize_session(struct nfs_client *clp)
-{
-	int status;
+out:
+	/*
+	 * Let the state manager reestablish state
+	 */
+	if (!test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) &&
+	    status == 0)
+		nfs41_setup_state_renewal(clp);
 
-	status = nfs4_proc_create_session(clp, 0);
-	if (!status) {
-		nfs_mark_client_ready(clp, NFS_CS_READY);
-	} else if (status == -NFS4ERR_STALE_CLIENTID) {
-		set_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state);
-		set_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state);
-	} else {
-		nfs_mark_client_ready(clp, status);
-	}
 	return status;
 }
+
 #else /* CONFIG_NFS_V4_1 */
 static int nfs4_reset_session(struct nfs_client *clp) { return 0; }
-static int nfs4_initialize_session(struct nfs_client *clp) { return 0; }
+static int nfs4_end_drain_session(struct nfs_client *clp) { return 0; }
 #endif /* CONFIG_NFS_V4_1 */
 
 /* Set NFS4CLNT_LEASE_EXPIRED for all v4.0 errors and for recoverable errors
@@ -1234,7 +1336,8 @@ static void nfs4_state_manager(struct nfs_client *clp)
 			status = nfs4_reclaim_lease(clp);
 			if (status) {
 				nfs4_set_lease_expired(clp, status);
-				if (status == -EAGAIN)
+				if (test_bit(NFS4CLNT_LEASE_EXPIRED,
+							&clp->cl_state))
 					continue;
 				if (clp->cl_cons_state ==
 							NFS_CS_SESSION_INITING)
@@ -1242,57 +1345,54 @@ static void nfs4_state_manager(struct nfs_client *clp)
 				goto out_error;
 			}
 			clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state);
+			set_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state);
 		}
 
 		if (test_and_clear_bit(NFS4CLNT_CHECK_LEASE, &clp->cl_state)) {
 			status = nfs4_check_lease(clp);
-			if (status != 0)
+			if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
 				continue;
+			if (status < 0 && status != -NFS4ERR_CB_PATH_DOWN)
+				goto out_error;
 		}
+
 		/* Initialize or reset the session */
-		if (test_and_clear_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state)
+		if (test_and_clear_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state)
 		   && nfs4_has_session(clp)) {
-			if (clp->cl_cons_state == NFS_CS_SESSION_INITING)
-				status = nfs4_initialize_session(clp);
-			else
-				status = nfs4_reset_session(clp);
-			if (status) {
-				if (status == -NFS4ERR_STALE_CLIENTID)
-					continue;
+			status = nfs4_reset_session(clp);
+			if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state))
+				continue;
+			if (status < 0)
 				goto out_error;
-			}
 		}
+
 		/* First recover reboot state... */
-		if (test_and_clear_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) {
+		if (test_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state)) {
 			status = nfs4_do_reclaim(clp,
 				nfs4_reboot_recovery_ops[clp->cl_minorversion]);
-			if (status == -NFS4ERR_STALE_CLIENTID)
-				continue;
-			if (test_bit(NFS4CLNT_SESSION_SETUP, &clp->cl_state))
+			if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) ||
+			    test_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state))
 				continue;
 			nfs4_state_end_reclaim_reboot(clp);
-			continue;
+			if (test_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state))
+				continue;
+			if (status < 0)
+				goto out_error;
 		}
 
 		/* Now recover expired state... */
 		if (test_and_clear_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state)) {
 			status = nfs4_do_reclaim(clp,
 				nfs4_nograce_recovery_ops[clp->cl_minorversion]);
-			if (status < 0) {
-				set_bit(NFS4CLNT_RECLAIM_NOGRACE, &clp->cl_state);
-				if (status == -NFS4ERR_STALE_CLIENTID)
-					continue;
-				if (status == -NFS4ERR_EXPIRED)
-					continue;
-				if (test_bit(NFS4CLNT_SESSION_SETUP,
-								&clp->cl_state))
-					continue;
+			if (test_bit(NFS4CLNT_LEASE_EXPIRED, &clp->cl_state) ||
+			    test_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state) ||
+			    test_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state))
+				continue;
+			if (status < 0)
 				goto out_error;
-			} else
-				nfs4_state_end_reclaim_nograce(clp);
-			continue;
 		}
 
+		nfs4_end_drain_session(clp);
 		if (test_and_clear_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state)) {
 			nfs_client_return_marked_delegations(clp);
 			continue;
@@ -1309,8 +1409,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
 out_error:
 	printk(KERN_WARNING "Error: state manager failed on NFSv4 server %s"
 			" with error %d\n", clp->cl_hostname, -status);
-	if (test_bit(NFS4CLNT_RECLAIM_REBOOT, &clp->cl_state))
-		nfs4_state_end_reclaim_reboot(clp);
+	nfs4_end_drain_session(clp);
 	nfs4_clear_state_manager_bit(clp);
 }
 
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 20b4e30e6c82..e437fd6a819f 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -46,11 +46,13 @@
 #include <linux/proc_fs.h>
 #include <linux/kdev_t.h>
 #include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/msg_prot.h>
 #include <linux/nfs.h>
 #include <linux/nfs4.h>
 #include <linux/nfs_fs.h>
 #include <linux/nfs_idmap.h>
 #include "nfs4_fs.h"
+#include "internal.h"
 
 #define NFSDBG_FACILITY		NFSDBG_XDR
 
@@ -134,7 +136,7 @@ static int nfs4_stat_to_errno(int);
 #define decode_lookup_maxsz	(op_decode_hdr_maxsz)
 #define encode_share_access_maxsz \
 				(2)
-#define encode_createmode_maxsz	(1 + encode_attrs_maxsz)
+#define encode_createmode_maxsz	(1 + encode_attrs_maxsz + encode_verifier_maxsz)
 #define encode_opentype_maxsz	(1 + encode_createmode_maxsz)
 #define encode_claim_null_maxsz	(1 + nfs4_name_maxsz)
 #define encode_open_maxsz	(op_encode_hdr_maxsz + \
@@ -299,6 +301,8 @@ static int nfs4_stat_to_errno(int);
 				XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 4)
 #define decode_sequence_maxsz	(op_decode_hdr_maxsz + \
 				XDR_QUADLEN(NFS4_MAX_SESSIONID_LEN) + 5)
+#define encode_reclaim_complete_maxsz	(op_encode_hdr_maxsz + 4)
+#define decode_reclaim_complete_maxsz	(op_decode_hdr_maxsz + 4)
 #else /* CONFIG_NFS_V4_1 */
 #define encode_sequence_maxsz	0
 #define decode_sequence_maxsz	0
@@ -676,6 +680,25 @@ static int nfs4_stat_to_errno(int);
 					 decode_sequence_maxsz + \
 					 decode_putrootfh_maxsz + \
 					 decode_fsinfo_maxsz)
+#define NFS4_enc_reclaim_complete_sz	(compound_encode_hdr_maxsz + \
+					 encode_sequence_maxsz + \
+					 encode_reclaim_complete_maxsz)
+#define NFS4_dec_reclaim_complete_sz	(compound_decode_hdr_maxsz + \
+					 decode_sequence_maxsz + \
+					 decode_reclaim_complete_maxsz)
+
+const u32 nfs41_maxwrite_overhead = ((RPC_MAX_HEADER_WITH_AUTH +
+				      compound_encode_hdr_maxsz +
+				      encode_sequence_maxsz +
+				      encode_putfh_maxsz +
+				      encode_getattr_maxsz) *
+				     XDR_UNIT);
+
+const u32 nfs41_maxread_overhead = ((RPC_MAX_HEADER_WITH_AUTH +
+				     compound_decode_hdr_maxsz +
+				     decode_sequence_maxsz +
+				     decode_putfh_maxsz) *
+				    XDR_UNIT);
 #endif /* CONFIG_NFS_V4_1 */
 
 static const umode_t nfs_type2fmt[] = {
@@ -1140,6 +1163,7 @@ static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_opena
 static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_openargs *arg)
 {
 	__be32 *p;
+	struct nfs_client *clp;
 
 	p = reserve_space(xdr, 4);
 	switch(arg->open_flags & O_EXCL) {
@@ -1148,8 +1172,23 @@ static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_op
 		encode_attrs(xdr, arg->u.attrs, arg->server);
 		break;
 	default:
-		*p = cpu_to_be32(NFS4_CREATE_EXCLUSIVE);
-		encode_nfs4_verifier(xdr, &arg->u.verifier);
+		clp = arg->server->nfs_client;
+		if (clp->cl_minorversion > 0) {
+			if (nfs4_has_persistent_session(clp)) {
+				*p = cpu_to_be32(NFS4_CREATE_GUARDED);
+				encode_attrs(xdr, arg->u.attrs, arg->server);
+			} else {
+				struct iattr dummy;
+
+				*p = cpu_to_be32(NFS4_CREATE_EXCLUSIVE4_1);
+				encode_nfs4_verifier(xdr, &arg->u.verifier);
+				dummy.ia_valid = 0;
+				encode_attrs(xdr, &dummy, arg->server);
+			}
+		} else {
+			*p = cpu_to_be32(NFS4_CREATE_EXCLUSIVE);
+			encode_nfs4_verifier(xdr, &arg->u.verifier);
+		}
 	}
 }
 
@@ -1592,6 +1631,19 @@ static void encode_destroy_session(struct xdr_stream *xdr,
 	hdr->nops++;
 	hdr->replen += decode_destroy_session_maxsz;
 }
+
+static void encode_reclaim_complete(struct xdr_stream *xdr,
+				    struct nfs41_reclaim_complete_args *args,
+				    struct compound_hdr *hdr)
+{
+	__be32 *p;
+
+	p = reserve_space(xdr, 8);
+	*p++ = cpu_to_be32(OP_RECLAIM_COMPLETE);
+	*p++ = cpu_to_be32(args->one_fs);
+	hdr->nops++;
+	hdr->replen += decode_reclaim_complete_maxsz;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static void encode_sequence(struct xdr_stream *xdr,
@@ -2096,7 +2148,7 @@ nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p,
 	encode_compound_hdr(&xdr, req, &hdr);
 	encode_sequence(&xdr, &args->seq_args, &hdr);
 	encode_putfh(&xdr, args->fh, &hdr);
-	replen = hdr.replen + nfs4_fattr_bitmap_maxsz + 1;
+	replen = hdr.replen + op_decode_hdr_maxsz + nfs4_fattr_bitmap_maxsz + 1;
 	encode_getattr_two(&xdr, FATTR4_WORD0_ACL, 0, &hdr);
 
 	xdr_inline_pages(&req->rq_rcv_buf, replen << 2,
@@ -2420,6 +2472,26 @@ static int nfs4_xdr_enc_get_lease_time(struct rpc_rqst *req, uint32_t *p,
 	encode_nops(&hdr);
 	return 0;
 }
+
+/*
+ * a RECLAIM_COMPLETE request
+ */
+static int nfs4_xdr_enc_reclaim_complete(struct rpc_rqst *req, uint32_t *p,
+				     struct nfs41_reclaim_complete_args *args)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr = {
+		.minorversion = nfs4_xdr_minorversion(&args->seq_args)
+	};
+
+	xdr_init_encode(&xdr, &req->rq_snd_buf, p);
+	encode_compound_hdr(&xdr, req, &hdr);
+	encode_sequence(&xdr, &args->seq_args, &hdr);
+	encode_reclaim_complete(&xdr, args, &hdr);
+	encode_nops(&hdr);
+	return 0;
+}
+
 #endif /* CONFIG_NFS_V4_1 */
 
 static void print_overflow_msg(const char *func, const struct xdr_stream *xdr)
@@ -4528,6 +4600,11 @@ static int decode_destroy_session(struct xdr_stream *xdr, void *dummy)
 {
 	return decode_op_hdr(xdr, OP_DESTROY_SESSION);
 }
+
+static int decode_reclaim_complete(struct xdr_stream *xdr, void *dummy)
+{
+	return decode_op_hdr(xdr, OP_RECLAIM_COMPLETE);
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 static int decode_sequence(struct xdr_stream *xdr,
@@ -4583,8 +4660,8 @@ static int decode_sequence(struct xdr_stream *xdr,
 	dummy = be32_to_cpup(p++);
 	/* target highest slot id - currently not processed */
 	dummy = be32_to_cpup(p++);
-	/* result flags - currently not processed */
-	dummy = be32_to_cpup(p);
+	/* result flags */
+	res->sr_status_flags = be32_to_cpup(p);
 	status = 0;
 out_err:
 	res->sr_status = status;
@@ -5309,7 +5386,7 @@ out:
 }
 
 /*
- * FSINFO request
+ * Decode FSINFO response
  */
 static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
 			       struct nfs4_fsinfo_res *res)
@@ -5330,7 +5407,7 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p,
 }
 
 /*
- * PATHCONF request
+ * Decode PATHCONF response
  */
 static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p,
 				 struct nfs4_pathconf_res *res)
@@ -5351,7 +5428,7 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p,
 }
 
 /*
- * STATFS request
+ * Decode STATFS response
  */
 static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p,
 			       struct nfs4_statfs_res *res)
@@ -5372,7 +5449,7 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p,
 }
 
 /*
- * GETATTR_BITMAP request
+ * Decode GETATTR_BITMAP response
  */
 static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, __be32 *p, struct nfs4_server_caps_res *res)
 {
@@ -5411,7 +5488,7 @@ static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, __be32 *p, void *dummy)
 }
 
 /*
- * a SETCLIENTID request
+ * Decode SETCLIENTID response
  */
 static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, __be32 *p,
 		struct nfs_client *clp)
@@ -5428,7 +5505,7 @@ static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, __be32 *p,
 }
 
 /*
- * a SETCLIENTID_CONFIRM request
+ * Decode SETCLIENTID_CONFIRM response
  */
 static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo)
 {
@@ -5448,7 +5525,7 @@ static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, __be32 *p, str
 }
 
 /*
- * DELEGRETURN request
+ * Decode DELEGRETURN response
  */
 static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_delegreturnres *res)
 {
@@ -5474,7 +5551,7 @@ out:
 }
 
 /*
- * FS_LOCATIONS request
+ * Decode FS_LOCATIONS response
  */
 static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p,
 				     struct nfs4_fs_locations_res *res)
@@ -5504,7 +5581,7 @@ out:
 
 #if defined(CONFIG_NFS_V4_1)
 /*
- * EXCHANGE_ID request
+ * Decode EXCHANGE_ID response
  */
 static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
 				    void *res)
@@ -5521,7 +5598,7 @@ static int nfs4_xdr_dec_exchange_id(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
- * a CREATE_SESSION request
+ * Decode CREATE_SESSION response
  */
 static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p,
 				       struct nfs41_create_session_res *res)
@@ -5538,7 +5615,7 @@ static int nfs4_xdr_dec_create_session(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
- * a DESTROY_SESSION request
+ * Decode DESTROY_SESSION response
  */
 static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p,
 					void *dummy)
@@ -5555,7 +5632,7 @@ static int nfs4_xdr_dec_destroy_session(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
- * a SEQUENCE request
+ * Decode SEQUENCE response
  */
 static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, uint32_t *p,
 				 struct nfs4_sequence_res *res)
@@ -5572,7 +5649,7 @@ static int nfs4_xdr_dec_sequence(struct rpc_rqst *rqstp, uint32_t *p,
 }
 
 /*
- * a GET_LEASE_TIME request
+ * Decode GET_LEASE_TIME response
  */
 static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
 				       struct nfs4_get_lease_time_res *res)
@@ -5591,6 +5668,25 @@ static int nfs4_xdr_dec_get_lease_time(struct rpc_rqst *rqstp, uint32_t *p,
 		status = decode_fsinfo(&xdr, res->lr_fsinfo);
 	return status;
 }
+
+/*
+ * Decode RECLAIM_COMPLETE response
+ */
+static int nfs4_xdr_dec_reclaim_complete(struct rpc_rqst *rqstp, uint32_t *p,
+					 struct nfs41_reclaim_complete_res *res)
+{
+	struct xdr_stream xdr;
+	struct compound_hdr hdr;
+	int status;
+
+	xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p);
+	status = decode_compound_hdr(&xdr, &hdr);
+	if (!status)
+		status = decode_sequence(&xdr, &res->seq_res, rqstp);
+	if (!status)
+		status = decode_reclaim_complete(&xdr, (void *)NULL);
+	return status;
+}
 #endif /* CONFIG_NFS_V4_1 */
 
 __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus)
@@ -5767,6 +5863,7 @@ struct rpc_procinfo	nfs4_procedures[] = {
   PROC(DESTROY_SESSION,	enc_destroy_session,	dec_destroy_session),
   PROC(SEQUENCE,	enc_sequence,	dec_sequence),
   PROC(GET_LEASE_TIME,	enc_get_lease_time,	dec_get_lease_time),
+  PROC(RECLAIM_COMPLETE, enc_reclaim_complete,  dec_reclaim_complete),
 #endif /* CONFIG_NFS_V4_1 */
 };
 
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 12c9e66d3f1d..db9b360ae19d 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -356,25 +356,19 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data
 	struct nfs_readres *resp = &data->res;
 
 	if (resp->eof || resp->count == argp->count)
-		goto out;
+		return;
 
 	/* This is a short read! */
 	nfs_inc_stats(data->inode, NFSIOS_SHORTREAD);
 	/* Has the server at least made some progress? */
 	if (resp->count == 0)
-		goto out;
+		return;
 
 	/* Yes, so retry the read at the end of the data */
 	argp->offset += resp->count;
 	argp->pgbase += resp->count;
 	argp->count -= resp->count;
-	nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client);
-	return;
-out:
-	nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client,
-				&data->res.seq_res);
-	return;
-
+	nfs_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client);
 }
 
 /*
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 90be551b80c1..ce907efc5508 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -175,14 +175,16 @@ static const match_table_t nfs_mount_option_tokens = {
 };
 
 enum {
-	Opt_xprt_udp, Opt_xprt_tcp, Opt_xprt_rdma,
+	Opt_xprt_udp, Opt_xprt_udp6, Opt_xprt_tcp, Opt_xprt_tcp6, Opt_xprt_rdma,
 
 	Opt_xprt_err
 };
 
 static const match_table_t nfs_xprt_protocol_tokens = {
 	{ Opt_xprt_udp, "udp" },
+	{ Opt_xprt_udp6, "udp6" },
 	{ Opt_xprt_tcp, "tcp" },
+	{ Opt_xprt_tcp6, "tcp6" },
 	{ Opt_xprt_rdma, "rdma" },
 
 	{ Opt_xprt_err, NULL }
@@ -492,6 +494,45 @@ static const char *nfs_pseudoflavour_to_name(rpc_authflavor_t flavour)
 	return sec_flavours[i].str;
 }
 
+static void nfs_show_mountd_netid(struct seq_file *m, struct nfs_server *nfss,
+				  int showdefaults)
+{
+	struct sockaddr *sap = (struct sockaddr *) &nfss->mountd_address;
+
+	seq_printf(m, ",mountproto=");
+	switch (sap->sa_family) {
+	case AF_INET:
+		switch (nfss->mountd_protocol) {
+		case IPPROTO_UDP:
+			seq_printf(m, RPCBIND_NETID_UDP);
+			break;
+		case IPPROTO_TCP:
+			seq_printf(m, RPCBIND_NETID_TCP);
+			break;
+		default:
+			if (showdefaults)
+				seq_printf(m, "auto");
+		}
+		break;
+	case AF_INET6:
+		switch (nfss->mountd_protocol) {
+		case IPPROTO_UDP:
+			seq_printf(m, RPCBIND_NETID_UDP6);
+			break;
+		case IPPROTO_TCP:
+			seq_printf(m, RPCBIND_NETID_TCP6);
+			break;
+		default:
+			if (showdefaults)
+				seq_printf(m, "auto");
+		}
+		break;
+	default:
+		if (showdefaults)
+			seq_printf(m, "auto");
+	}
+}
+
 static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss,
 				    int showdefaults)
 {
@@ -505,7 +546,7 @@ static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss,
 	}
 	case AF_INET6: {
 		struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap;
-		seq_printf(m, ",mountaddr=%pI6", &sin6->sin6_addr);
+		seq_printf(m, ",mountaddr=%pI6c", &sin6->sin6_addr);
 		break;
 	}
 	default:
@@ -518,17 +559,7 @@ static void nfs_show_mountd_options(struct seq_file *m, struct nfs_server *nfss,
 	if (nfss->mountd_port || showdefaults)
 		seq_printf(m, ",mountport=%u", nfss->mountd_port);
 
-	switch (nfss->mountd_protocol) {
-	case IPPROTO_UDP:
-		seq_printf(m, ",mountproto=udp");
-		break;
-	case IPPROTO_TCP:
-		seq_printf(m, ",mountproto=tcp");
-		break;
-	default:
-		if (showdefaults)
-			seq_printf(m, ",mountproto=auto");
-	}
+	nfs_show_mountd_netid(m, nfss, showdefaults);
 }
 
 /*
@@ -578,7 +609,7 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss,
 			seq_puts(m, nfs_infop->nostr);
 	}
 	seq_printf(m, ",proto=%s",
-		   rpc_peeraddr2str(nfss->client, RPC_DISPLAY_PROTO));
+		   rpc_peeraddr2str(nfss->client, RPC_DISPLAY_NETID));
 	if (version == 4) {
 		if (nfss->port != NFS_PORT)
 			seq_printf(m, ",port=%u", nfss->port);
@@ -714,8 +745,6 @@ static void nfs_umount_begin(struct super_block *sb)
 	struct nfs_server *server;
 	struct rpc_clnt *rpc;
 
-	lock_kernel();
-
 	server = NFS_SB(sb);
 	/* -EIO all pending I/O */
 	rpc = server->client_acl;
@@ -724,8 +753,6 @@ static void nfs_umount_begin(struct super_block *sb)
 	rpc = server->client;
 	if (!IS_ERR(rpc))
 		rpc_killall_tasks(rpc);
-
-	unlock_kernel();
 }
 
 static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(unsigned int version)
@@ -734,8 +761,6 @@ static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(unsigned int ve
 
 	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (data) {
-		data->rsize		= NFS_MAX_FILE_IO_SIZE;
-		data->wsize		= NFS_MAX_FILE_IO_SIZE;
 		data->acregmin		= NFS_DEF_ACREGMIN;
 		data->acregmax		= NFS_DEF_ACREGMAX;
 		data->acdirmin		= NFS_DEF_ACDIRMIN;
@@ -887,6 +912,8 @@ static int nfs_parse_mount_options(char *raw,
 {
 	char *p, *string, *secdata;
 	int rc, sloppy = 0, invalid_option = 0;
+	unsigned short protofamily = AF_UNSPEC;
+	unsigned short mountfamily = AF_UNSPEC;
 
 	if (!raw) {
 		dfprintk(MOUNT, "NFS: mount options string was NULL.\n");
@@ -1232,12 +1259,17 @@ static int nfs_parse_mount_options(char *raw,
 			token = match_token(string,
 					    nfs_xprt_protocol_tokens, args);
 
+			protofamily = AF_INET;
 			switch (token) {
+			case Opt_xprt_udp6:
+				protofamily = AF_INET6;
 			case Opt_xprt_udp:
 				mnt->flags &= ~NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_UDP;
 				kfree(string);
 				break;
+			case Opt_xprt_tcp6:
+				protofamily = AF_INET6;
 			case Opt_xprt_tcp:
 				mnt->flags |= NFS_MOUNT_TCP;
 				mnt->nfs_server.protocol = XPRT_TRANSPORT_TCP;
@@ -1265,10 +1297,15 @@ static int nfs_parse_mount_options(char *raw,
 					    nfs_xprt_protocol_tokens, args);
 			kfree(string);
 
+			mountfamily = AF_INET;
 			switch (token) {
+			case Opt_xprt_udp6:
+				mountfamily = AF_INET6;
 			case Opt_xprt_udp:
 				mnt->mount_server.protocol = XPRT_TRANSPORT_UDP;
 				break;
+			case Opt_xprt_tcp6:
+				mountfamily = AF_INET6;
 			case Opt_xprt_tcp:
 				mnt->mount_server.protocol = XPRT_TRANSPORT_TCP;
 				break;
@@ -1367,8 +1404,33 @@ static int nfs_parse_mount_options(char *raw,
 	if (!sloppy && invalid_option)
 		return 0;
 
+	/*
+	 * verify that any proto=/mountproto= options match the address
+	 * familiies in the addr=/mountaddr= options.
+	 */
+	if (protofamily != AF_UNSPEC &&
+	    protofamily != mnt->nfs_server.address.ss_family)
+		goto out_proto_mismatch;
+
+	if (mountfamily != AF_UNSPEC) {
+		if (mnt->mount_server.addrlen) {
+			if (mountfamily != mnt->mount_server.address.ss_family)
+				goto out_mountproto_mismatch;
+		} else {
+			if (mountfamily != mnt->nfs_server.address.ss_family)
+				goto out_mountproto_mismatch;
+		}
+	}
+
 	return 1;
 
+out_mountproto_mismatch:
+	printk(KERN_INFO "NFS: mount server address does not match mountproto= "
+			 "option\n");
+	return 0;
+out_proto_mismatch:
+	printk(KERN_INFO "NFS: server address does not match proto= option\n");
+	return 0;
 out_invalid_address:
 	printk(KERN_INFO "NFS: bad IP address specified: %s\n", p);
 	return 0;
@@ -1881,7 +1943,6 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
 	if (data == NULL)
 		return -ENOMEM;
 
-	lock_kernel();
 	/* fill out struct with values from existing mount */
 	data->flags = nfss->flags;
 	data->rsize = nfss->rsize;
@@ -1907,7 +1968,6 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
 	error = nfs_compare_remount_data(nfss, data);
 out:
 	kfree(data);
-	unlock_kernel();
 	return error;
 }
 
diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
index 1064c91ae810..6da3d3ff6edd 100644
--- a/fs/nfs/unlink.c
+++ b/fs/nfs/unlink.c
@@ -83,7 +83,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata)
 	struct inode *dir = data->dir;
 
 	if (!NFS_PROTO(dir)->unlink_done(task, dir))
-		nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client);
+		nfs_restart_rpc(task, NFS_SERVER(dir)->nfs_client);
 }
 
 /**
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index c84b5cc1a943..d171696017f4 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -774,7 +774,7 @@ int nfs_updatepage(struct file *file, struct page *page,
 	 */
 	if (nfs_write_pageuptodate(page, inode) &&
 			inode->i_flock == NULL &&
-			!(file->f_flags & O_SYNC)) {
+			!(file->f_flags & O_DSYNC)) {
 		count = max(count + offset, nfs_page_length(page));
 		offset = 0;
 	}
@@ -1216,7 +1216,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 				 */
 				argp->stable = NFS_FILE_SYNC;
 			}
-			nfs4_restart_rpc(task, server->nfs_client);
+			nfs_restart_rpc(task, server->nfs_client);
 			return -EAGAIN;
 		}
 		if (time_before(complain, jiffies)) {
@@ -1228,7 +1228,6 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data)
 		/* Can't do anything about it except throw an error. */
 		task->tk_status = -EIO;
 	}
-	nfs4_sequence_free_slot(server->nfs_client, &data->res.seq_res);
 	return 0;
 }
 
@@ -1612,15 +1611,16 @@ int nfs_migrate_page(struct address_space *mapping, struct page *newpage,
 	if (ret)
 		goto out_unlock;
 	page_cache_get(newpage);
+	spin_lock(&mapping->host->i_lock);
 	req->wb_page = newpage;
 	SetPagePrivate(newpage);
-	set_page_private(newpage, page_private(page));
+	set_page_private(newpage, (unsigned long)req);
 	ClearPagePrivate(page);
 	set_page_private(page, 0);
+	spin_unlock(&mapping->host->i_lock);
 	page_cache_release(page);
 out_unlock:
 	nfs_clear_page_tag_locked(req);
-	nfs_release_request(req);
 out:
 	return ret;
 }
diff --git a/fs/nfsctl.c b/fs/nfsctl.c
index 8f9a20556f79..d3854d94b7cf 100644
--- a/fs/nfsctl.c
+++ b/fs/nfsctl.c
@@ -7,8 +7,6 @@
 #include <linux/types.h>
 #include <linux/file.h>
 #include <linux/fs.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
 #include <linux/nfsd/syscall.h>
 #include <linux/cred.h>
 #include <linux/sched.h>
diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c
index 36fcabbf5186..79717a40daba 100644
--- a/fs/nfsd/auth.c
+++ b/fs/nfsd/auth.c
@@ -1,15 +1,7 @@
-/*
- * linux/fs/nfsd/auth.c
- *
- * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
- */
+/* Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de> */
 
-#include <linux/types.h>
 #include <linux/sched.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/sunrpc/svcauth.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/export.h>
+#include "nfsd.h"
 #include "auth.h"
 
 int nfsexp_flags(struct svc_rqst *rqstp, struct svc_export *exp)
diff --git a/fs/nfsd/cache.h b/fs/nfsd/cache.h
new file mode 100644
index 000000000000..d892be61016c
--- /dev/null
+++ b/fs/nfsd/cache.h
@@ -0,0 +1,83 @@
+/*
+ * Request reply cache. This was heavily inspired by the
+ * implementation in 4.3BSD/4.4BSD.
+ *
+ * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef NFSCACHE_H
+#define NFSCACHE_H
+
+#include <linux/sunrpc/svc.h>
+
+/*
+ * Representation of a reply cache entry.
+ */
+struct svc_cacherep {
+	struct hlist_node	c_hash;
+	struct list_head	c_lru;
+
+	unsigned char		c_state,	/* unused, inprog, done */
+				c_type,		/* status, buffer */
+				c_secure : 1;	/* req came from port < 1024 */
+	struct sockaddr_in	c_addr;
+	__be32			c_xid;
+	u32			c_prot;
+	u32			c_proc;
+	u32			c_vers;
+	unsigned long		c_timestamp;
+	union {
+		struct kvec	u_vec;
+		__be32		u_status;
+	}			c_u;
+};
+
+#define c_replvec		c_u.u_vec
+#define c_replstat		c_u.u_status
+
+/* cache entry states */
+enum {
+	RC_UNUSED,
+	RC_INPROG,
+	RC_DONE
+};
+
+/* return values */
+enum {
+	RC_DROPIT,
+	RC_REPLY,
+	RC_DOIT,
+	RC_INTR
+};
+
+/*
+ * Cache types.
+ * We may want to add more types one day, e.g. for diropres and
+ * attrstat replies. Using cache entries with fixed length instead
+ * of buffer pointers may be more efficient.
+ */
+enum {
+	RC_NOCACHE,
+	RC_REPLSTAT,
+	RC_REPLBUFF,
+};
+
+/*
+ * If requests are retransmitted within this interval, they're dropped.
+ */
+#define RC_DELAY		(HZ/5)
+
+int	nfsd_reply_cache_init(void);
+void	nfsd_reply_cache_shutdown(void);
+int	nfsd_cache_lookup(struct svc_rqst *, int);
+void	nfsd_cache_update(struct svc_rqst *, int, __be32 *);
+
+#ifdef CONFIG_NFSD_V4
+void	nfsd4_set_statp(struct svc_rqst *rqstp, __be32 *statp);
+#else  /* CONFIG_NFSD_V4 */
+static inline void nfsd4_set_statp(struct svc_rqst *rqstp, __be32 *statp)
+{
+}
+#endif /* CONFIG_NFSD_V4 */
+
+#endif /* NFSCACHE_H */
diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index c1c9e035d4a4..c487810a2366 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -1,7 +1,5 @@
 #define MSNFS	/* HACK HACK */
 /*
- * linux/fs/nfsd/export.c
- *
  * NFS exporting and validation.
  *
  * We maintain a list of clients, each of which has a list of
@@ -14,29 +12,16 @@
  * Copyright (C) 1995, 1996 Olaf Kirch, <okir@monad.swb.de>
  */
 
-#include <linux/unistd.h>
-#include <linux/slab.h>
-#include <linux/stat.h>
-#include <linux/in.h>
-#include <linux/seq_file.h>
-#include <linux/syscalls.h>
-#include <linux/rwsem.h>
-#include <linux/dcache.h>
 #include <linux/namei.h>
-#include <linux/mount.h>
-#include <linux/hash.h>
 #include <linux/module.h>
 #include <linux/exportfs.h>
 
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/nfsfh.h>
 #include <linux/nfsd/syscall.h>
-#include <linux/lockd/bind.h>
-#include <linux/sunrpc/msg_prot.h>
-#include <linux/sunrpc/gss_api.h>
 #include <net/ipv6.h>
 
+#include "nfsd.h"
+#include "nfsfh.h"
+
 #define NFSDDBG_FACILITY	NFSDDBG_EXPORT
 
 typedef struct auth_domain	svc_client;
@@ -369,16 +354,25 @@ static struct svc_export *svc_export_update(struct svc_export *new,
 					    struct svc_export *old);
 static struct svc_export *svc_export_lookup(struct svc_export *);
 
-static int check_export(struct inode *inode, int flags, unsigned char *uuid)
+static int check_export(struct inode *inode, int *flags, unsigned char *uuid)
 {
 
-	/* We currently export only dirs and regular files.
-	 * This is what umountd does.
+	/*
+	 * We currently export only dirs, regular files, and (for v4
+	 * pseudoroot) symlinks.
 	 */
 	if (!S_ISDIR(inode->i_mode) &&
+	    !S_ISLNK(inode->i_mode) &&
 	    !S_ISREG(inode->i_mode))
 		return -ENOTDIR;
 
+	/*
+	 * Mountd should never pass down a writeable V4ROOT export, but,
+	 * just to make sure:
+	 */
+	if (*flags & NFSEXP_V4ROOT)
+		*flags |= NFSEXP_READONLY;
+
 	/* There are two requirements on a filesystem to be exportable.
 	 * 1:  We must be able to identify the filesystem from a number.
 	 *       either a device number (so FS_REQUIRES_DEV needed)
@@ -387,7 +381,7 @@ static int check_export(struct inode *inode, int flags, unsigned char *uuid)
 	 *       This means that s_export_op must be set.
 	 */
 	if (!(inode->i_sb->s_type->fs_flags & FS_REQUIRES_DEV) &&
-	    !(flags & NFSEXP_FSID) &&
+	    !(*flags & NFSEXP_FSID) &&
 	    uuid == NULL) {
 		dprintk("exp_export: export of non-dev fs without fsid\n");
 		return -EINVAL;
@@ -602,7 +596,7 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen)
 				goto out4;
 		}
 
-		err = check_export(exp.ex_path.dentry->d_inode, exp.ex_flags,
+		err = check_export(exp.ex_path.dentry->d_inode, &exp.ex_flags,
 				   exp.ex_uuid);
 		if (err)
 			goto out4;
@@ -1041,7 +1035,7 @@ exp_export(struct nfsctl_export *nxp)
 		goto finish;
 	}
 
-	err = check_export(path.dentry->d_inode, nxp->ex_flags, NULL);
+	err = check_export(path.dentry->d_inode, &nxp->ex_flags, NULL);
 	if (err) goto finish;
 
 	err = -ENOMEM;
@@ -1320,6 +1314,23 @@ rqst_exp_parent(struct svc_rqst *rqstp, struct path *path)
 	return exp;
 }
 
+static struct svc_export *find_fsidzero_export(struct svc_rqst *rqstp)
+{
+	struct svc_export *exp;
+	u32 fsidv[2];
+
+	mk_fsid(FSID_NUM, fsidv, 0, 0, 0, NULL);
+
+	exp = rqst_exp_find(rqstp, FSID_NUM, fsidv);
+	/*
+	 * We shouldn't have accepting an nfsv4 request at all if we
+	 * don't have a pseudoexport!:
+	 */
+	if (IS_ERR(exp) && PTR_ERR(exp) == -ENOENT)
+		exp = ERR_PTR(-ESERVERFAULT);
+	return exp;
+}
+
 /*
  * Called when we need the filehandle for the root of the pseudofs,
  * for a given NFSv4 client.   The root is defined to be the
@@ -1330,11 +1341,8 @@ exp_pseudoroot(struct svc_rqst *rqstp, struct svc_fh *fhp)
 {
 	struct svc_export *exp;
 	__be32 rv;
-	u32 fsidv[2];
 
-	mk_fsid(FSID_NUM, fsidv, 0, 0, 0, NULL);
-
-	exp = rqst_exp_find(rqstp, FSID_NUM, fsidv);
+	exp = find_fsidzero_export(rqstp);
 	if (IS_ERR(exp))
 		return nfserrno(PTR_ERR(exp));
 	rv = fh_compose(fhp, exp, exp->ex_path.dentry, NULL);
@@ -1425,6 +1433,7 @@ static struct flags {
 	{ NFSEXP_CROSSMOUNT, {"crossmnt", ""}},
 	{ NFSEXP_NOSUBTREECHECK, {"no_subtree_check", ""}},
 	{ NFSEXP_NOAUTHNLM, {"insecure_locks", ""}},
+	{ NFSEXP_V4ROOT, {"v4root", ""}},
 #ifdef MSNFS
 	{ NFSEXP_MSNFS, {"msnfs", ""}},
 #endif
diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c
index b2786a5f9afe..0c6d81670137 100644
--- a/fs/nfsd/lockd.c
+++ b/fs/nfsd/lockd.c
@@ -1,6 +1,4 @@
 /*
- * linux/fs/nfsd/lockd.c
- *
  * This file contains all the stubs needed when communicating with lockd.
  * This level of indirection is necessary so we can run nfsd+lockd without
  * requiring the nfs client to be compiled in/loaded, and vice versa.
@@ -8,14 +6,10 @@
  * Copyright (C) 1996, Olaf Kirch <okir@monad.swb.de>
  */
 
-#include <linux/types.h>
-#include <linux/fs.h>
 #include <linux/file.h>
-#include <linux/mount.h>
-#include <linux/sunrpc/clnt.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
 #include <linux/lockd/bind.h>
+#include "nfsd.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_LOCKD
 
diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c
index 4e3219e84116..f20589d2ae27 100644
--- a/fs/nfsd/nfs2acl.c
+++ b/fs/nfsd/nfs2acl.c
@@ -1,19 +1,15 @@
 /*
- * linux/fs/nfsd/nfs2acl.c
- *
  * Process version 2 NFSACL requests.
  *
  * Copyright (C) 2002-2003 Andreas Gruenbacher <agruen@suse.de>
  */
 
-#include <linux/sunrpc/svc.h>
-#include <linux/nfs.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/cache.h>
-#include <linux/nfsd/xdr.h>
-#include <linux/nfsd/xdr3.h>
-#include <linux/posix_acl.h>
+#include "nfsd.h"
+/* FIXME: nfsacl.h is a broken header */
 #include <linux/nfsacl.h>
+#include "cache.h"
+#include "xdr3.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_PROC
 #define RETURN_STATUS(st)	{ resp->status = (st); return (st); }
@@ -217,6 +213,16 @@ static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p,
  * XDR encode functions
  */
 
+/*
+ * There must be an encoding function for void results so svc_process
+ * will work properly.
+ */
+int
+nfsaclsvc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy)
+{
+	return xdr_ressize_check(rqstp, p);
+}
+
 /* GETACL */
 static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, __be32 *p,
 		struct nfsd3_getaclres *resp)
@@ -308,7 +314,6 @@ static int nfsaclsvc_release_access(struct svc_rqst *rqstp, __be32 *p,
 }
 
 #define nfsaclsvc_decode_voidargs	NULL
-#define nfsaclsvc_encode_voidres	NULL
 #define nfsaclsvc_release_void		NULL
 #define nfsd3_fhandleargs	nfsd_fhandle
 #define nfsd3_attrstatres	nfsd_attrstat
@@ -346,5 +351,5 @@ struct svc_version	nfsd_acl_version2 = {
 		.vs_proc	= nfsd_acl_procedures2,
 		.vs_dispatch	= nfsd_dispatch,
 		.vs_xdrsize	= NFS3_SVC_XDRSIZE,
-		.vs_hidden	= 1,
+		.vs_hidden	= 0,
 };
diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c
index 9981dbb377a3..e0c4846bad92 100644
--- a/fs/nfsd/nfs3acl.c
+++ b/fs/nfsd/nfs3acl.c
@@ -1,18 +1,15 @@
 /*
- * linux/fs/nfsd/nfs3acl.c
- *
  * Process version 3 NFSACL requests.
  *
  * Copyright (C) 2002-2003 Andreas Gruenbacher <agruen@suse.de>
  */
 
-#include <linux/sunrpc/svc.h>
-#include <linux/nfs3.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/cache.h>
-#include <linux/nfsd/xdr3.h>
-#include <linux/posix_acl.h>
+#include "nfsd.h"
+/* FIXME: nfsacl.h is a broken header */
 #include <linux/nfsacl.h>
+#include "cache.h"
+#include "xdr3.h"
+#include "vfs.h"
 
 #define RETURN_STATUS(st)	{ resp->status = (st); return (st); }
 
@@ -264,6 +261,6 @@ struct svc_version	nfsd_acl_version3 = {
 		.vs_proc	= nfsd_acl_procedures3,
 		.vs_dispatch	= nfsd_dispatch,
 		.vs_xdrsize	= NFS3_SVC_XDRSIZE,
-		.vs_hidden	= 1,
+		.vs_hidden	= 0,
 };
 
diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
index a713c418a922..3d68f45a37b9 100644
--- a/fs/nfsd/nfs3proc.c
+++ b/fs/nfsd/nfs3proc.c
@@ -1,30 +1,16 @@
 /*
- * linux/fs/nfsd/nfs3proc.c
- *
  * Process version 3 NFS requests.
  *
  * Copyright (C) 1996, 1997, 1998 Olaf Kirch <okir@monad.swb.de>
  */
 
-#include <linux/linkage.h>
-#include <linux/time.h>
-#include <linux/errno.h>
 #include <linux/fs.h>
 #include <linux/ext2_fs.h>
-#include <linux/stat.h>
-#include <linux/fcntl.h>
-#include <linux/net.h>
-#include <linux/in.h>
-#include <linux/unistd.h>
-#include <linux/slab.h>
-#include <linux/major.h>
 #include <linux/magic.h>
 
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/cache.h>
-#include <linux/nfsd/xdr3.h>
-#include <linux/nfs3.h>
+#include "cache.h"
+#include "xdr3.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_PROC
 
diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c
index d0a2ce1b4324..2a533a0af2a9 100644
--- a/fs/nfsd/nfs3xdr.c
+++ b/fs/nfsd/nfs3xdr.c
@@ -1,6 +1,4 @@
 /*
- * linux/fs/nfsd/nfs3xdr.c
- *
  * XDR support for nfsd/protocol version 3.
  *
  * Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
@@ -8,19 +6,8 @@
  * 2003-08-09 Jamie Lokier: Use htonl() for nanoseconds, not htons()!
  */
 
-#include <linux/types.h>
-#include <linux/time.h>
-#include <linux/nfs3.h>
-#include <linux/list.h>
-#include <linux/spinlock.h>
-#include <linux/dcache.h>
 #include <linux/namei.h>
-#include <linux/mm.h>
-#include <linux/vfs.h>
-#include <linux/sunrpc/xdr.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/xdr3.h>
+#include "xdr3.h"
 #include "auth.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_XDR
diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c
index 725d02f210e2..88150685df34 100644
--- a/fs/nfsd/nfs4acl.c
+++ b/fs/nfsd/nfs4acl.c
@@ -1,6 +1,4 @@
 /*
- *  fs/nfs4acl/acl.c
- *
  *  Common NFSv4 ACL handling code.
  *
  *  Copyright (c) 2002, 2003 The Regents of the University of Michigan.
@@ -36,15 +34,7 @@
  *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
-#include <linux/string.h>
-#include <linux/slab.h>
-#include <linux/list.h>
-#include <linux/types.h>
-#include <linux/fs.h>
-#include <linux/module.h>
 #include <linux/nfs_fs.h>
-#include <linux/posix_acl.h>
-#include <linux/nfs4.h>
 #include <linux/nfs4_acl.h>
 
 
@@ -389,7 +379,7 @@ sort_pacl(struct posix_acl *pacl)
 	sort_pacl_range(pacl, 1, i-1);
 
 	BUG_ON(pacl->a_entries[i].e_tag != ACL_GROUP_OBJ);
-	j = i++;
+	j = ++i;
 	while (pacl->a_entries[j].e_tag == ACL_GROUP)
 		j++;
 	sort_pacl_range(pacl, i, j-1);
diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c
index 24e8d78f8dde..c6eed2a3b093 100644
--- a/fs/nfsd/nfs4callback.c
+++ b/fs/nfsd/nfs4callback.c
@@ -1,6 +1,4 @@
 /*
- *  linux/fs/nfsd/nfs4callback.c
- *
  *  Copyright (c) 2001 The Regents of the University of Michigan.
  *  All rights reserved.
  *
@@ -33,22 +31,9 @@
  *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
 
-#include <linux/module.h>
-#include <linux/list.h>
-#include <linux/inet.h>
-#include <linux/errno.h>
-#include <linux/delay.h>
-#include <linux/sched.h>
-#include <linux/kthread.h>
-#include <linux/sunrpc/xdr.h>
-#include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/clnt.h>
-#include <linux/sunrpc/svcsock.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/state.h>
-#include <linux/sunrpc/sched.h>
-#include <linux/nfs4.h>
-#include <linux/sunrpc/xprtsock.h>
+#include "nfsd.h"
+#include "state.h"
 
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c
index ba2c199592fd..6e2983b27f3c 100644
--- a/fs/nfsd/nfs4idmap.c
+++ b/fs/nfsd/nfs4idmap.c
@@ -1,6 +1,4 @@
 /*
- *  fs/nfsd/nfs4idmap.c
- *
  *  Mapping of UID/GIDs to name and vice versa.
  *
  *  Copyright (c) 2002, 2003 The Regents of the University of
@@ -35,22 +33,9 @@
  */
 
 #include <linux/module.h>
-#include <linux/init.h>
-
-#include <linux/mm.h>
-#include <linux/errno.h>
-#include <linux/string.h>
-#include <linux/sunrpc/clnt.h>
-#include <linux/nfs.h>
-#include <linux/nfs4.h>
-#include <linux/nfs_fs.h>
-#include <linux/nfs_page.h>
-#include <linux/sunrpc/cache.h>
 #include <linux/nfsd_idmap.h>
-#include <linux/list.h>
-#include <linux/time.h>
 #include <linux/seq_file.h>
-#include <linux/sunrpc/svcauth.h>
+#include <linux/sched.h>
 
 /*
  * Cache entry
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index bebc0c2e1b0a..37514c469846 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -1,6 +1,4 @@
 /*
- *  fs/nfsd/nfs4proc.c
- *
  *  Server-side procedures for NFSv4.
  *
  *  Copyright (c) 2002 The Regents of the University of Michigan.
@@ -34,20 +32,11 @@
  *  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
  *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  */
-
-#include <linux/param.h>
-#include <linux/major.h>
-#include <linux/slab.h>
 #include <linux/file.h>
 
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/cache.h>
-#include <linux/nfs4.h>
-#include <linux/nfsd/state.h>
-#include <linux/nfsd/xdr4.h>
-#include <linux/nfs4_acl.h>
-#include <linux/sunrpc/gss_api.h>
+#include "cache.h"
+#include "xdr4.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_PROC
 
@@ -170,7 +159,7 @@ do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs
 		accmode |= NFSD_MAY_READ;
 	if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE)
 		accmode |= (NFSD_MAY_WRITE | NFSD_MAY_TRUNC);
-	if (open->op_share_deny & NFS4_SHARE_DENY_WRITE)
+	if (open->op_share_deny & NFS4_SHARE_DENY_READ)
 		accmode |= NFSD_MAY_WRITE;
 
 	status = fh_verify(rqstp, current_fh, S_IFREG, accmode);
diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c
index b5348405046b..5a754f7b71ed 100644
--- a/fs/nfsd/nfs4recover.c
+++ b/fs/nfsd/nfs4recover.c
@@ -1,6 +1,4 @@
 /*
-*  linux/fs/nfsd/nfs4recover.c
-*
 *  Copyright (c) 2004 The Regents of the University of Michigan.
 *  All rights reserved.
 *
@@ -33,20 +31,14 @@
 *
 */
 
-#include <linux/err.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfs4.h>
-#include <linux/nfsd/state.h>
-#include <linux/nfsd/xdr4.h>
-#include <linux/param.h>
 #include <linux/file.h>
 #include <linux/namei.h>
-#include <asm/uaccess.h>
-#include <linux/scatterlist.h>
 #include <linux/crypto.h>
 #include <linux/sched.h>
-#include <linux/mount.h>
+
+#include "nfsd.h"
+#include "state.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 2153f9bdbebd..f19ed866c95f 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1,6 +1,4 @@
 /*
-*  linux/fs/nfsd/nfs4state.c
-*
 *  Copyright (c) 2001 The Regents of the University of Michigan.
 *  All rights reserved.
 *
@@ -34,28 +32,14 @@
 *
 */
 
-#include <linux/param.h>
-#include <linux/major.h>
-#include <linux/slab.h>
-
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/cache.h>
 #include <linux/file.h>
-#include <linux/mount.h>
-#include <linux/workqueue.h>
 #include <linux/smp_lock.h>
-#include <linux/kthread.h>
-#include <linux/nfs4.h>
-#include <linux/nfsd/state.h>
-#include <linux/nfsd/xdr4.h>
 #include <linux/namei.h>
 #include <linux/swap.h>
-#include <linux/mutex.h>
-#include <linux/lockd/bind.h>
-#include <linux/module.h>
 #include <linux/sunrpc/svcauth_gss.h>
 #include <linux/sunrpc/clnt.h>
+#include "xdr4.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY                NFSDDBG_PROC
 
@@ -477,13 +461,14 @@ static int set_forechannel_drc_size(struct nfsd4_channel_attrs *fchan)
 
 /*
  * fchan holds the client values on input, and the server values on output
+ * sv_max_mesg is the maximum payload plus one page for overhead.
  */
 static int init_forechannel_attrs(struct svc_rqst *rqstp,
 				  struct nfsd4_channel_attrs *session_fchan,
 				  struct nfsd4_channel_attrs *fchan)
 {
 	int status = 0;
-	__u32   maxcount = svc_max_payload(rqstp);
+	__u32   maxcount = nfsd_serv->sv_max_mesg;
 
 	/* headerpadsz set to zero in encode routine */
 
@@ -523,6 +508,15 @@ free_session_slots(struct nfsd4_session *ses)
 		kfree(ses->se_slots[i]);
 }
 
+/*
+ * We don't actually need to cache the rpc and session headers, so we
+ * can allocate a little less for each slot:
+ */
+static inline int slot_bytes(struct nfsd4_channel_attrs *ca)
+{
+	return ca->maxresp_cached - NFSD_MIN_HDR_SEQ_SZ;
+}
+
 static int
 alloc_init_session(struct svc_rqst *rqstp, struct nfs4_client *clp,
 		   struct nfsd4_create_session *cses)
@@ -554,7 +548,7 @@ alloc_init_session(struct svc_rqst *rqstp, struct nfs4_client *clp,
 	memcpy(new, &tmp, sizeof(*new));
 
 	/* allocate each struct nfsd4_slot and data cache in one piece */
-	cachesize = new->se_fchannel.maxresp_cached - NFSD_MIN_HDR_SEQ_SZ;
+	cachesize = slot_bytes(&new->se_fchannel);
 	for (i = 0; i < new->se_fchannel.maxreqs; i++) {
 		sp = kzalloc(sizeof(*sp) + cachesize, GFP_KERNEL);
 		if (!sp)
@@ -628,10 +622,12 @@ void
 free_session(struct kref *kref)
 {
 	struct nfsd4_session *ses;
+	int mem;
 
 	ses = container_of(kref, struct nfsd4_session, se_ref);
 	spin_lock(&nfsd_drc_lock);
-	nfsd_drc_mem_used -= ses->se_fchannel.maxreqs * NFSD_SLOT_CACHE_SIZE;
+	mem = ses->se_fchannel.maxreqs * slot_bytes(&ses->se_fchannel);
+	nfsd_drc_mem_used -= mem;
 	spin_unlock(&nfsd_drc_lock);
 	free_session_slots(ses);
 	kfree(ses);
@@ -2404,11 +2400,8 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, struct nfs4_sta
 
 	memcpy(&open->op_delegate_stateid, &dp->dl_stateid, sizeof(dp->dl_stateid));
 
-	dprintk("NFSD: delegation stateid=(%08x/%08x/%08x/%08x)\n\n",
-	             dp->dl_stateid.si_boot,
-	             dp->dl_stateid.si_stateownerid,
-	             dp->dl_stateid.si_fileid,
-	             dp->dl_stateid.si_generation);
+	dprintk("NFSD: delegation stateid=" STATEID_FMT "\n",
+		STATEID_VAL(&dp->dl_stateid));
 out:
 	if (open->op_claim_type == NFS4_OPEN_CLAIM_PREVIOUS
 			&& flag == NFS4_OPEN_DELEGATE_NONE
@@ -2498,9 +2491,8 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nf
 
 	status = nfs_ok;
 
-	dprintk("nfs4_process_open2: stateid=(%08x/%08x/%08x/%08x)\n",
-	            stp->st_stateid.si_boot, stp->st_stateid.si_stateownerid,
-	            stp->st_stateid.si_fileid, stp->st_stateid.si_generation);
+	dprintk("%s: stateid=" STATEID_FMT "\n", __func__,
+		STATEID_VAL(&stp->st_stateid));
 out:
 	if (fp)
 		put_nfs4_file(fp);
@@ -2666,9 +2658,8 @@ STALE_STATEID(stateid_t *stateid)
 {
 	if (time_after((unsigned long)boot_time,
 			(unsigned long)stateid->si_boot)) {
-		dprintk("NFSD: stale stateid (%08x/%08x/%08x/%08x)!\n",
-			stateid->si_boot, stateid->si_stateownerid,
-			stateid->si_fileid, stateid->si_generation);
+		dprintk("NFSD: stale stateid " STATEID_FMT "!\n",
+			STATEID_VAL(stateid));
 		return 1;
 	}
 	return 0;
@@ -2680,9 +2671,8 @@ EXPIRED_STATEID(stateid_t *stateid)
 	if (time_before((unsigned long)boot_time,
 			((unsigned long)stateid->si_boot)) &&
 	    time_before((unsigned long)(stateid->si_boot + lease_time), get_seconds())) {
-		dprintk("NFSD: expired stateid (%08x/%08x/%08x/%08x)!\n",
-			stateid->si_boot, stateid->si_stateownerid,
-			stateid->si_fileid, stateid->si_generation);
+		dprintk("NFSD: expired stateid " STATEID_FMT "!\n",
+			STATEID_VAL(stateid));
 		return 1;
 	}
 	return 0;
@@ -2696,9 +2686,8 @@ stateid_error_map(stateid_t *stateid)
 	if (EXPIRED_STATEID(stateid))
 		return nfserr_expired;
 
-	dprintk("NFSD: bad stateid (%08x/%08x/%08x/%08x)!\n",
-		stateid->si_boot, stateid->si_stateownerid,
-		stateid->si_fileid, stateid->si_generation);
+	dprintk("NFSD: bad stateid " STATEID_FMT "!\n",
+		STATEID_VAL(stateid));
 	return nfserr_bad_stateid;
 }
 
@@ -2884,10 +2873,8 @@ nfs4_preprocess_seqid_op(struct nfsd4_compound_state *cstate, u32 seqid,
 	struct svc_fh *current_fh = &cstate->current_fh;
 	__be32 status;
 
-	dprintk("NFSD: preprocess_seqid_op: seqid=%d " 
-			"stateid = (%08x/%08x/%08x/%08x)\n", seqid,
-		stateid->si_boot, stateid->si_stateownerid, stateid->si_fileid,
-		stateid->si_generation);
+	dprintk("NFSD: %s: seqid=%d stateid = " STATEID_FMT "\n", __func__,
+		seqid, STATEID_VAL(stateid));
 
 	*stpp = NULL;
 	*sopp = NULL;
@@ -3019,12 +3006,8 @@ nfsd4_open_confirm(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
 	sop->so_confirmed = 1;
 	update_stateid(&stp->st_stateid);
 	memcpy(&oc->oc_resp_stateid, &stp->st_stateid, sizeof(stateid_t));
-	dprintk("NFSD: nfsd4_open_confirm: success, seqid=%d " 
-		"stateid=(%08x/%08x/%08x/%08x)\n", oc->oc_seqid,
-		         stp->st_stateid.si_boot,
-		         stp->st_stateid.si_stateownerid,
-		         stp->st_stateid.si_fileid,
-		         stp->st_stateid.si_generation);
+	dprintk("NFSD: %s: success, seqid=%d stateid=" STATEID_FMT "\n",
+		__func__, oc->oc_seqid, STATEID_VAL(&stp->st_stateid));
 
 	nfsd4_create_clid_dir(sop->so_client);
 out:
@@ -3283,9 +3266,8 @@ find_delegation_stateid(struct inode *ino, stateid_t *stid)
 	struct nfs4_file *fp;
 	struct nfs4_delegation *dl;
 
-	dprintk("NFSD:find_delegation_stateid stateid=(%08x/%08x/%08x/%08x)\n",
-                    stid->si_boot, stid->si_stateownerid,
-                    stid->si_fileid, stid->si_generation);
+	dprintk("NFSD: %s: stateid=" STATEID_FMT "\n", __func__,
+		STATEID_VAL(stid));
 
 	fp = find_file(ino);
 	if (!fp)
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 0fbd50cee1f6..a8587e90fd5a 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -40,24 +40,16 @@
  * at the end of nfs4svc_decode_compoundargs.
  */
 
-#include <linux/param.h>
-#include <linux/smp.h>
-#include <linux/fs.h>
 #include <linux/namei.h>
-#include <linux/vfs.h>
+#include <linux/statfs.h>
 #include <linux/utsname.h>
-#include <linux/sunrpc/xdr.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/sunrpc/clnt.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/state.h>
-#include <linux/nfsd/xdr4.h>
 #include <linux/nfsd_idmap.h>
-#include <linux/nfs4.h>
 #include <linux/nfs4_acl.h>
-#include <linux/sunrpc/gss_api.h>
 #include <linux/sunrpc/svcauth_gss.h>
 
+#include "xdr4.h"
+#include "vfs.h"
+
 #define NFSDDBG_FACILITY		NFSDDBG_XDR
 
 /*
@@ -2204,11 +2196,14 @@ nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd,
 	 * we will not follow the cross mount and will fill the attribtutes
 	 * directly from the mountpoint dentry.
 	 */
-	if (d_mountpoint(dentry) && !attributes_need_mount(cd->rd_bmval))
-		ignore_crossmnt = 1;
-	else if (d_mountpoint(dentry)) {
+	if (nfsd_mountpoint(dentry, exp)) {
 		int err;
 
+		if (!(exp->ex_flags & NFSEXP_V4ROOT)
+				&& !attributes_need_mount(cd->rd_bmval)) {
+			ignore_crossmnt = 1;
+			goto out_encode;
+		}
 		/*
 		 * Why the heck aren't we just using nfsd_lookup??
 		 * Different "."/".." handling?  Something else?
@@ -2224,6 +2219,7 @@ nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd,
 			goto out_put;
 
 	}
+out_encode:
 	nfserr = nfsd4_encode_fattr(NULL, exp, dentry, p, buflen, cd->rd_bmval,
 					cd->rd_rqstp, ignore_crossmnt);
 out_put:
diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c
index 4638635c5d87..da08560c4818 100644
--- a/fs/nfsd/nfscache.c
+++ b/fs/nfsd/nfscache.c
@@ -1,6 +1,4 @@
 /*
- * linux/fs/nfsd/nfscache.c
- *
  * Request reply cache. This is currently a global cache, but this may
  * change in the future and be a per-client cache.
  *
@@ -10,16 +8,8 @@
  * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
  */
 
-#include <linux/kernel.h>
-#include <linux/time.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-#include <linux/spinlock.h>
-#include <linux/list.h>
-
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/cache.h>
+#include "nfsd.h"
+#include "cache.h"
 
 /* Size of reply cache. Common values are:
  * 4.3BSD:	128
diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 5c01fc148ce8..2604c3e70ea5 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -1,46 +1,20 @@
 /*
- * linux/fs/nfsd/nfsctl.c
- *
  * Syscall interface to knfsd.
  *
  * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
  */
 
-#include <linux/module.h>
-
-#include <linux/linkage.h>
-#include <linux/time.h>
-#include <linux/errno.h>
-#include <linux/fs.h>
 #include <linux/namei.h>
-#include <linux/fcntl.h>
-#include <linux/net.h>
-#include <linux/in.h>
-#include <linux/syscalls.h>
-#include <linux/unistd.h>
-#include <linux/slab.h>
-#include <linux/proc_fs.h>
-#include <linux/seq_file.h>
-#include <linux/pagemap.h>
-#include <linux/init.h>
-#include <linux/inet.h>
-#include <linux/string.h>
 #include <linux/ctype.h>
 
-#include <linux/nfs.h>
 #include <linux/nfsd_idmap.h>
-#include <linux/lockd/bind.h>
-#include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/svcsock.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/cache.h>
-#include <linux/nfsd/xdr.h>
 #include <linux/nfsd/syscall.h>
 #include <linux/lockd/lockd.h>
 #include <linux/sunrpc/clnt.h>
 
-#include <asm/uaccess.h>
-#include <net/ipv6.h>
+#include "nfsd.h"
+#include "cache.h"
 
 /*
  *	We have a single directory with 9 nodes in it.
@@ -55,6 +29,7 @@ enum {
 	NFSD_Getfd,
 	NFSD_Getfs,
 	NFSD_List,
+	NFSD_Export_features,
 	NFSD_Fh,
 	NFSD_FO_UnlockIP,
 	NFSD_FO_UnlockFS,
@@ -173,6 +148,24 @@ static const struct file_operations exports_operations = {
 	.owner		= THIS_MODULE,
 };
 
+static int export_features_show(struct seq_file *m, void *v)
+{
+	seq_printf(m, "0x%x 0x%x\n", NFSEXP_ALLFLAGS, NFSEXP_SECINFO_FLAGS);
+	return 0;
+}
+
+static int export_features_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, export_features_show, NULL);
+}
+
+static struct file_operations export_features_operations = {
+	.open		= export_features_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
 extern int nfsd_pool_stats_open(struct inode *inode, struct file *file);
 extern int nfsd_pool_stats_release(struct inode *inode, struct file *file);
 
@@ -1330,6 +1323,8 @@ static int nfsd_fill_super(struct super_block * sb, void * data, int silent)
 		[NFSD_Getfd] = {".getfd", &transaction_ops, S_IWUSR|S_IRUSR},
 		[NFSD_Getfs] = {".getfs", &transaction_ops, S_IWUSR|S_IRUSR},
 		[NFSD_List] = {"exports", &exports_operations, S_IRUGO},
+		[NFSD_Export_features] = {"export_features",
+					&export_features_operations, S_IRUGO},
 		[NFSD_FO_UnlockIP] = {"unlock_ip",
 					&transaction_ops, S_IWUSR|S_IRUSR},
 		[NFSD_FO_UnlockFS] = {"unlock_filesystem",
diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h
new file mode 100644
index 000000000000..e942a1aaac92
--- /dev/null
+++ b/fs/nfsd/nfsd.h
@@ -0,0 +1,338 @@
+/*
+ * Hodge-podge collection of knfsd-related stuff.
+ * I will sort this out later.
+ *
+ * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef LINUX_NFSD_NFSD_H
+#define LINUX_NFSD_NFSD_H
+
+#include <linux/types.h>
+#include <linux/mount.h>
+
+#include <linux/nfsd/debug.h>
+#include <linux/nfsd/export.h>
+#include <linux/nfsd/stats.h>
+/*
+ * nfsd version
+ */
+#define NFSD_SUPPORTED_MINOR_VERSION	1
+
+struct readdir_cd {
+	__be32			err;	/* 0, nfserr, or nfserr_eof */
+};
+
+
+extern struct svc_program	nfsd_program;
+extern struct svc_version	nfsd_version2, nfsd_version3,
+				nfsd_version4;
+extern u32			nfsd_supported_minorversion;
+extern struct mutex		nfsd_mutex;
+extern struct svc_serv		*nfsd_serv;
+extern spinlock_t		nfsd_drc_lock;
+extern unsigned int		nfsd_drc_max_mem;
+extern unsigned int		nfsd_drc_mem_used;
+
+extern const struct seq_operations nfs_exports_op;
+
+/*
+ * Function prototypes.
+ */
+int		nfsd_svc(unsigned short port, int nrservs);
+int		nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp);
+
+int		nfsd_nrthreads(void);
+int		nfsd_nrpools(void);
+int		nfsd_get_nrthreads(int n, int *);
+int		nfsd_set_nrthreads(int n, int *);
+
+#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
+#ifdef CONFIG_NFSD_V2_ACL
+extern struct svc_version nfsd_acl_version2;
+#else
+#define nfsd_acl_version2 NULL
+#endif
+#ifdef CONFIG_NFSD_V3_ACL
+extern struct svc_version nfsd_acl_version3;
+#else
+#define nfsd_acl_version3 NULL
+#endif
+#endif
+
+enum vers_op {NFSD_SET, NFSD_CLEAR, NFSD_TEST, NFSD_AVAIL };
+int nfsd_vers(int vers, enum vers_op change);
+int nfsd_minorversion(u32 minorversion, enum vers_op change);
+void nfsd_reset_versions(void);
+int nfsd_create_serv(void);
+
+extern int nfsd_max_blksize;
+
+static inline int nfsd_v4client(struct svc_rqst *rq)
+{
+	return rq->rq_prog == NFS_PROGRAM && rq->rq_vers == 4;
+}
+
+/* 
+ * NFSv4 State
+ */
+#ifdef CONFIG_NFSD_V4
+extern unsigned int max_delegations;
+int nfs4_state_init(void);
+void nfsd4_free_slabs(void);
+int nfs4_state_start(void);
+void nfs4_state_shutdown(void);
+time_t nfs4_lease_time(void);
+void nfs4_reset_lease(time_t leasetime);
+int nfs4_reset_recoverydir(char *recdir);
+#else
+static inline int nfs4_state_init(void) { return 0; }
+static inline void nfsd4_free_slabs(void) { }
+static inline int nfs4_state_start(void) { return 0; }
+static inline void nfs4_state_shutdown(void) { }
+static inline time_t nfs4_lease_time(void) { return 0; }
+static inline void nfs4_reset_lease(time_t leasetime) { }
+static inline int nfs4_reset_recoverydir(char *recdir) { return 0; }
+#endif
+
+/*
+ * lockd binding
+ */
+void		nfsd_lockd_init(void);
+void		nfsd_lockd_shutdown(void);
+
+
+/*
+ * These macros provide pre-xdr'ed values for faster operation.
+ */
+#define	nfs_ok			cpu_to_be32(NFS_OK)
+#define	nfserr_perm		cpu_to_be32(NFSERR_PERM)
+#define	nfserr_noent		cpu_to_be32(NFSERR_NOENT)
+#define	nfserr_io		cpu_to_be32(NFSERR_IO)
+#define	nfserr_nxio		cpu_to_be32(NFSERR_NXIO)
+#define	nfserr_eagain		cpu_to_be32(NFSERR_EAGAIN)
+#define	nfserr_acces		cpu_to_be32(NFSERR_ACCES)
+#define	nfserr_exist		cpu_to_be32(NFSERR_EXIST)
+#define	nfserr_xdev		cpu_to_be32(NFSERR_XDEV)
+#define	nfserr_nodev		cpu_to_be32(NFSERR_NODEV)
+#define	nfserr_notdir		cpu_to_be32(NFSERR_NOTDIR)
+#define	nfserr_isdir		cpu_to_be32(NFSERR_ISDIR)
+#define	nfserr_inval		cpu_to_be32(NFSERR_INVAL)
+#define	nfserr_fbig		cpu_to_be32(NFSERR_FBIG)
+#define	nfserr_nospc		cpu_to_be32(NFSERR_NOSPC)
+#define	nfserr_rofs		cpu_to_be32(NFSERR_ROFS)
+#define	nfserr_mlink		cpu_to_be32(NFSERR_MLINK)
+#define	nfserr_opnotsupp	cpu_to_be32(NFSERR_OPNOTSUPP)
+#define	nfserr_nametoolong	cpu_to_be32(NFSERR_NAMETOOLONG)
+#define	nfserr_notempty		cpu_to_be32(NFSERR_NOTEMPTY)
+#define	nfserr_dquot		cpu_to_be32(NFSERR_DQUOT)
+#define	nfserr_stale		cpu_to_be32(NFSERR_STALE)
+#define	nfserr_remote		cpu_to_be32(NFSERR_REMOTE)
+#define	nfserr_wflush		cpu_to_be32(NFSERR_WFLUSH)
+#define	nfserr_badhandle	cpu_to_be32(NFSERR_BADHANDLE)
+#define	nfserr_notsync		cpu_to_be32(NFSERR_NOT_SYNC)
+#define	nfserr_badcookie	cpu_to_be32(NFSERR_BAD_COOKIE)
+#define	nfserr_notsupp		cpu_to_be32(NFSERR_NOTSUPP)
+#define	nfserr_toosmall		cpu_to_be32(NFSERR_TOOSMALL)
+#define	nfserr_serverfault	cpu_to_be32(NFSERR_SERVERFAULT)
+#define	nfserr_badtype		cpu_to_be32(NFSERR_BADTYPE)
+#define	nfserr_jukebox		cpu_to_be32(NFSERR_JUKEBOX)
+#define	nfserr_denied		cpu_to_be32(NFSERR_DENIED)
+#define	nfserr_deadlock		cpu_to_be32(NFSERR_DEADLOCK)
+#define nfserr_expired          cpu_to_be32(NFSERR_EXPIRED)
+#define	nfserr_bad_cookie	cpu_to_be32(NFSERR_BAD_COOKIE)
+#define	nfserr_same		cpu_to_be32(NFSERR_SAME)
+#define	nfserr_clid_inuse	cpu_to_be32(NFSERR_CLID_INUSE)
+#define	nfserr_stale_clientid	cpu_to_be32(NFSERR_STALE_CLIENTID)
+#define	nfserr_resource		cpu_to_be32(NFSERR_RESOURCE)
+#define	nfserr_moved		cpu_to_be32(NFSERR_MOVED)
+#define	nfserr_nofilehandle	cpu_to_be32(NFSERR_NOFILEHANDLE)
+#define	nfserr_minor_vers_mismatch	cpu_to_be32(NFSERR_MINOR_VERS_MISMATCH)
+#define nfserr_share_denied	cpu_to_be32(NFSERR_SHARE_DENIED)
+#define nfserr_stale_stateid	cpu_to_be32(NFSERR_STALE_STATEID)
+#define nfserr_old_stateid	cpu_to_be32(NFSERR_OLD_STATEID)
+#define nfserr_bad_stateid	cpu_to_be32(NFSERR_BAD_STATEID)
+#define nfserr_bad_seqid	cpu_to_be32(NFSERR_BAD_SEQID)
+#define	nfserr_symlink		cpu_to_be32(NFSERR_SYMLINK)
+#define	nfserr_not_same		cpu_to_be32(NFSERR_NOT_SAME)
+#define	nfserr_restorefh	cpu_to_be32(NFSERR_RESTOREFH)
+#define	nfserr_attrnotsupp	cpu_to_be32(NFSERR_ATTRNOTSUPP)
+#define	nfserr_bad_xdr		cpu_to_be32(NFSERR_BAD_XDR)
+#define	nfserr_openmode		cpu_to_be32(NFSERR_OPENMODE)
+#define	nfserr_locks_held	cpu_to_be32(NFSERR_LOCKS_HELD)
+#define	nfserr_op_illegal	cpu_to_be32(NFSERR_OP_ILLEGAL)
+#define	nfserr_grace		cpu_to_be32(NFSERR_GRACE)
+#define	nfserr_no_grace		cpu_to_be32(NFSERR_NO_GRACE)
+#define	nfserr_reclaim_bad	cpu_to_be32(NFSERR_RECLAIM_BAD)
+#define	nfserr_badname		cpu_to_be32(NFSERR_BADNAME)
+#define	nfserr_cb_path_down	cpu_to_be32(NFSERR_CB_PATH_DOWN)
+#define	nfserr_locked		cpu_to_be32(NFSERR_LOCKED)
+#define	nfserr_wrongsec		cpu_to_be32(NFSERR_WRONGSEC)
+#define nfserr_badiomode		cpu_to_be32(NFS4ERR_BADIOMODE)
+#define nfserr_badlayout		cpu_to_be32(NFS4ERR_BADLAYOUT)
+#define nfserr_bad_session_digest	cpu_to_be32(NFS4ERR_BAD_SESSION_DIGEST)
+#define nfserr_badsession		cpu_to_be32(NFS4ERR_BADSESSION)
+#define nfserr_badslot			cpu_to_be32(NFS4ERR_BADSLOT)
+#define nfserr_complete_already		cpu_to_be32(NFS4ERR_COMPLETE_ALREADY)
+#define nfserr_conn_not_bound_to_session cpu_to_be32(NFS4ERR_CONN_NOT_BOUND_TO_SESSION)
+#define nfserr_deleg_already_wanted	cpu_to_be32(NFS4ERR_DELEG_ALREADY_WANTED)
+#define nfserr_back_chan_busy		cpu_to_be32(NFS4ERR_BACK_CHAN_BUSY)
+#define nfserr_layouttrylater		cpu_to_be32(NFS4ERR_LAYOUTTRYLATER)
+#define nfserr_layoutunavailable	cpu_to_be32(NFS4ERR_LAYOUTUNAVAILABLE)
+#define nfserr_nomatching_layout	cpu_to_be32(NFS4ERR_NOMATCHING_LAYOUT)
+#define nfserr_recallconflict		cpu_to_be32(NFS4ERR_RECALLCONFLICT)
+#define nfserr_unknown_layouttype	cpu_to_be32(NFS4ERR_UNKNOWN_LAYOUTTYPE)
+#define nfserr_seq_misordered		cpu_to_be32(NFS4ERR_SEQ_MISORDERED)
+#define nfserr_sequence_pos		cpu_to_be32(NFS4ERR_SEQUENCE_POS)
+#define nfserr_req_too_big		cpu_to_be32(NFS4ERR_REQ_TOO_BIG)
+#define nfserr_rep_too_big		cpu_to_be32(NFS4ERR_REP_TOO_BIG)
+#define nfserr_rep_too_big_to_cache	cpu_to_be32(NFS4ERR_REP_TOO_BIG_TO_CACHE)
+#define nfserr_retry_uncached_rep	cpu_to_be32(NFS4ERR_RETRY_UNCACHED_REP)
+#define nfserr_unsafe_compound		cpu_to_be32(NFS4ERR_UNSAFE_COMPOUND)
+#define nfserr_too_many_ops		cpu_to_be32(NFS4ERR_TOO_MANY_OPS)
+#define nfserr_op_not_in_session	cpu_to_be32(NFS4ERR_OP_NOT_IN_SESSION)
+#define nfserr_hash_alg_unsupp		cpu_to_be32(NFS4ERR_HASH_ALG_UNSUPP)
+#define nfserr_clientid_busy		cpu_to_be32(NFS4ERR_CLIENTID_BUSY)
+#define nfserr_pnfs_io_hole		cpu_to_be32(NFS4ERR_PNFS_IO_HOLE)
+#define nfserr_seq_false_retry		cpu_to_be32(NFS4ERR_SEQ_FALSE_RETRY)
+#define nfserr_bad_high_slot		cpu_to_be32(NFS4ERR_BAD_HIGH_SLOT)
+#define nfserr_deadsession		cpu_to_be32(NFS4ERR_DEADSESSION)
+#define nfserr_encr_alg_unsupp		cpu_to_be32(NFS4ERR_ENCR_ALG_UNSUPP)
+#define nfserr_pnfs_no_layout		cpu_to_be32(NFS4ERR_PNFS_NO_LAYOUT)
+#define nfserr_not_only_op		cpu_to_be32(NFS4ERR_NOT_ONLY_OP)
+#define nfserr_wrong_cred		cpu_to_be32(NFS4ERR_WRONG_CRED)
+#define nfserr_wrong_type		cpu_to_be32(NFS4ERR_WRONG_TYPE)
+#define nfserr_dirdeleg_unavail		cpu_to_be32(NFS4ERR_DIRDELEG_UNAVAIL)
+#define nfserr_reject_deleg		cpu_to_be32(NFS4ERR_REJECT_DELEG)
+#define nfserr_returnconflict		cpu_to_be32(NFS4ERR_RETURNCONFLICT)
+#define nfserr_deleg_revoked		cpu_to_be32(NFS4ERR_DELEG_REVOKED)
+
+/* error codes for internal use */
+/* if a request fails due to kmalloc failure, it gets dropped.
+ *  Client should resend eventually
+ */
+#define	nfserr_dropit		cpu_to_be32(30000)
+/* end-of-file indicator in readdir */
+#define	nfserr_eof		cpu_to_be32(30001)
+/* replay detected */
+#define	nfserr_replay_me	cpu_to_be32(11001)
+/* nfs41 replay detected */
+#define	nfserr_replay_cache	cpu_to_be32(11002)
+
+/* Check for dir entries '.' and '..' */
+#define isdotent(n, l)	(l < 3 && n[0] == '.' && (l == 1 || n[1] == '.'))
+
+/*
+ * Time of server startup
+ */
+extern struct timeval	nfssvc_boot;
+
+#ifdef CONFIG_NFSD_V4
+
+/* before processing a COMPOUND operation, we have to check that there
+ * is enough space in the buffer for XDR encode to succeed.  otherwise,
+ * we might process an operation with side effects, and be unable to
+ * tell the client that the operation succeeded.
+ *
+ * COMPOUND_SLACK_SPACE - this is the minimum bytes of buffer space
+ * needed to encode an "ordinary" _successful_ operation.  (GETATTR,
+ * READ, READDIR, and READLINK have their own buffer checks.)  if we
+ * fall below this level, we fail the next operation with NFS4ERR_RESOURCE.
+ *
+ * COMPOUND_ERR_SLACK_SPACE - this is the minimum bytes of buffer space
+ * needed to encode an operation which has failed with NFS4ERR_RESOURCE.
+ * care is taken to ensure that we never fall below this level for any
+ * reason.
+ */
+#define	COMPOUND_SLACK_SPACE		140    /* OP_GETFH */
+#define COMPOUND_ERR_SLACK_SPACE	12     /* OP_SETATTR */
+
+#define NFSD_LEASE_TIME                 (nfs4_lease_time())
+#define NFSD_LAUNDROMAT_MINTIMEOUT      10   /* seconds */
+
+/*
+ * The following attributes are currently not supported by the NFSv4 server:
+ *    ARCHIVE       (deprecated anyway)
+ *    HIDDEN        (unlikely to be supported any time soon)
+ *    MIMETYPE      (unlikely to be supported any time soon)
+ *    QUOTA_*       (will be supported in a forthcoming patch)
+ *    SYSTEM        (unlikely to be supported any time soon)
+ *    TIME_BACKUP   (unlikely to be supported any time soon)
+ *    TIME_CREATE   (unlikely to be supported any time soon)
+ */
+#define NFSD4_SUPPORTED_ATTRS_WORD0                                                         \
+(FATTR4_WORD0_SUPPORTED_ATTRS   | FATTR4_WORD0_TYPE         | FATTR4_WORD0_FH_EXPIRE_TYPE   \
+ | FATTR4_WORD0_CHANGE          | FATTR4_WORD0_SIZE         | FATTR4_WORD0_LINK_SUPPORT     \
+ | FATTR4_WORD0_SYMLINK_SUPPORT | FATTR4_WORD0_NAMED_ATTR   | FATTR4_WORD0_FSID             \
+ | FATTR4_WORD0_UNIQUE_HANDLES  | FATTR4_WORD0_LEASE_TIME   | FATTR4_WORD0_RDATTR_ERROR     \
+ | FATTR4_WORD0_ACLSUPPORT      | FATTR4_WORD0_CANSETTIME   | FATTR4_WORD0_CASE_INSENSITIVE \
+ | FATTR4_WORD0_CASE_PRESERVING | FATTR4_WORD0_CHOWN_RESTRICTED                             \
+ | FATTR4_WORD0_FILEHANDLE      | FATTR4_WORD0_FILEID       | FATTR4_WORD0_FILES_AVAIL      \
+ | FATTR4_WORD0_FILES_FREE      | FATTR4_WORD0_FILES_TOTAL  | FATTR4_WORD0_FS_LOCATIONS | FATTR4_WORD0_HOMOGENEOUS      \
+ | FATTR4_WORD0_MAXFILESIZE     | FATTR4_WORD0_MAXLINK      | FATTR4_WORD0_MAXNAME          \
+ | FATTR4_WORD0_MAXREAD         | FATTR4_WORD0_MAXWRITE     | FATTR4_WORD0_ACL)
+
+#define NFSD4_SUPPORTED_ATTRS_WORD1                                                         \
+(FATTR4_WORD1_MODE              | FATTR4_WORD1_NO_TRUNC     | FATTR4_WORD1_NUMLINKS         \
+ | FATTR4_WORD1_OWNER	        | FATTR4_WORD1_OWNER_GROUP  | FATTR4_WORD1_RAWDEV           \
+ | FATTR4_WORD1_SPACE_AVAIL     | FATTR4_WORD1_SPACE_FREE   | FATTR4_WORD1_SPACE_TOTAL      \
+ | FATTR4_WORD1_SPACE_USED      | FATTR4_WORD1_TIME_ACCESS  | FATTR4_WORD1_TIME_ACCESS_SET  \
+ | FATTR4_WORD1_TIME_DELTA   | FATTR4_WORD1_TIME_METADATA    \
+ | FATTR4_WORD1_TIME_MODIFY     | FATTR4_WORD1_TIME_MODIFY_SET | FATTR4_WORD1_MOUNTED_ON_FILEID)
+
+#define NFSD4_SUPPORTED_ATTRS_WORD2 0
+
+#define NFSD4_1_SUPPORTED_ATTRS_WORD0 \
+	NFSD4_SUPPORTED_ATTRS_WORD0
+
+#define NFSD4_1_SUPPORTED_ATTRS_WORD1 \
+	NFSD4_SUPPORTED_ATTRS_WORD1
+
+#define NFSD4_1_SUPPORTED_ATTRS_WORD2 \
+	(NFSD4_SUPPORTED_ATTRS_WORD2 | FATTR4_WORD2_SUPPATTR_EXCLCREAT)
+
+static inline u32 nfsd_suppattrs0(u32 minorversion)
+{
+	return minorversion ? NFSD4_1_SUPPORTED_ATTRS_WORD0
+			    : NFSD4_SUPPORTED_ATTRS_WORD0;
+}
+
+static inline u32 nfsd_suppattrs1(u32 minorversion)
+{
+	return minorversion ? NFSD4_1_SUPPORTED_ATTRS_WORD1
+			    : NFSD4_SUPPORTED_ATTRS_WORD1;
+}
+
+static inline u32 nfsd_suppattrs2(u32 minorversion)
+{
+	return minorversion ? NFSD4_1_SUPPORTED_ATTRS_WORD2
+			    : NFSD4_SUPPORTED_ATTRS_WORD2;
+}
+
+/* These will return ERR_INVAL if specified in GETATTR or READDIR. */
+#define NFSD_WRITEONLY_ATTRS_WORD1							    \
+(FATTR4_WORD1_TIME_ACCESS_SET   | FATTR4_WORD1_TIME_MODIFY_SET)
+
+/* These are the only attrs allowed in CREATE/OPEN/SETATTR. */
+#define NFSD_WRITEABLE_ATTRS_WORD0                                                          \
+(FATTR4_WORD0_SIZE              | FATTR4_WORD0_ACL                                         )
+#define NFSD_WRITEABLE_ATTRS_WORD1                                                          \
+(FATTR4_WORD1_MODE              | FATTR4_WORD1_OWNER         | FATTR4_WORD1_OWNER_GROUP     \
+ | FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_MODIFY_SET)
+#define NFSD_WRITEABLE_ATTRS_WORD2 0
+
+#define NFSD_SUPPATTR_EXCLCREAT_WORD0 \
+	NFSD_WRITEABLE_ATTRS_WORD0
+/*
+ * we currently store the exclusive create verifier in the v_{a,m}time
+ * attributes so the client can't set these at create time using EXCLUSIVE4_1
+ */
+#define NFSD_SUPPATTR_EXCLCREAT_WORD1 \
+	(NFSD_WRITEABLE_ATTRS_WORD1 & \
+	 ~(FATTR4_WORD1_TIME_ACCESS_SET | FATTR4_WORD1_TIME_MODIFY_SET))
+#define NFSD_SUPPATTR_EXCLCREAT_WORD2 \
+	NFSD_WRITEABLE_ATTRS_WORD2
+
+#endif /* CONFIG_NFSD_V4 */
+
+#endif /* LINUX_NFSD_NFSD_H */
diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c
index 01965b2f3a76..55c8e63af0be 100644
--- a/fs/nfsd/nfsfh.c
+++ b/fs/nfsd/nfsfh.c
@@ -1,6 +1,4 @@
 /*
- * linux/fs/nfsd/nfsfh.c
- *
  * NFS server file handle treatment.
  *
  * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
@@ -9,19 +7,11 @@
  * ... and again Southern-Winter 2001 to support export_operations
  */
 
-#include <linux/slab.h>
-#include <linux/fs.h>
-#include <linux/unistd.h>
-#include <linux/string.h>
-#include <linux/stat.h>
-#include <linux/dcache.h>
 #include <linux/exportfs.h>
-#include <linux/mount.h>
 
-#include <linux/sunrpc/clnt.h>
-#include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/svcauth_gss.h>
-#include <linux/nfsd/nfsd.h>
+#include "nfsd.h"
+#include "vfs.h"
 #include "auth.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_FH
@@ -96,8 +86,10 @@ nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int type)
 static __be32 nfsd_setuser_and_check_port(struct svc_rqst *rqstp,
 					  struct svc_export *exp)
 {
+	int flags = nfsexp_flags(rqstp, exp);
+
 	/* Check if the request originated from a secure port. */
-	if (!rqstp->rq_secure && EX_SECURE(exp)) {
+	if (!rqstp->rq_secure && !(flags & NFSEXP_INSECURE_PORT)) {
 		RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]);
 		dprintk(KERN_WARNING
 		       "nfsd: request from insecure port %s!\n",
@@ -109,6 +101,36 @@ static __be32 nfsd_setuser_and_check_port(struct svc_rqst *rqstp,
 	return nfserrno(nfsd_setuser(rqstp, exp));
 }
 
+static inline __be32 check_pseudo_root(struct svc_rqst *rqstp,
+	struct dentry *dentry, struct svc_export *exp)
+{
+	if (!(exp->ex_flags & NFSEXP_V4ROOT))
+		return nfs_ok;
+	/*
+	 * v2/v3 clients have no need for the V4ROOT export--they use
+	 * the mount protocl instead; also, further V4ROOT checks may be
+	 * in v4-specific code, in which case v2/v3 clients could bypass
+	 * them.
+	 */
+	if (!nfsd_v4client(rqstp))
+		return nfserr_stale;
+	/*
+	 * We're exposing only the directories and symlinks that have to be
+	 * traversed on the way to real exports:
+	 */
+	if (unlikely(!S_ISDIR(dentry->d_inode->i_mode) &&
+		     !S_ISLNK(dentry->d_inode->i_mode)))
+		return nfserr_stale;
+	/*
+	 * A pseudoroot export gives permission to access only one
+	 * single directory; the kernel has to make another upcall
+	 * before granting access to anything else under it:
+	 */
+	if (unlikely(dentry != exp->ex_path.dentry))
+		return nfserr_stale;
+	return nfs_ok;
+}
+
 /*
  * Use the given filehandle to look up the corresponding export and
  * dentry.  On success, the results are used to set fh_export and
@@ -232,14 +254,6 @@ static __be32 nfsd_set_fh_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp)
 		goto out;
 	}
 
-	if (exp->ex_flags & NFSEXP_NOSUBTREECHECK) {
-		error = nfsd_setuser_and_check_port(rqstp, exp);
-		if (error) {
-			dput(dentry);
-			goto out;
-		}
-	}
-
 	if (S_ISDIR(dentry->d_inode->i_mode) &&
 			(dentry->d_flags & DCACHE_DISCONNECTED)) {
 		printk("nfsd: find_fh_dentry returned a DISCONNECTED directory: %s/%s\n",
@@ -294,28 +308,32 @@ fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access)
 		error = nfsd_set_fh_dentry(rqstp, fhp);
 		if (error)
 			goto out;
-		dentry = fhp->fh_dentry;
-		exp = fhp->fh_export;
-	} else {
-		/*
-		 * just rechecking permissions
-		 * (e.g. nfsproc_create calls fh_verify, then nfsd_create
-		 * does as well)
-		 */
-		dprintk("nfsd: fh_verify - just checking\n");
-		dentry = fhp->fh_dentry;
-		exp = fhp->fh_export;
-		/*
-		 * Set user creds for this exportpoint; necessary even
-		 * in the "just checking" case because this may be a
-		 * filehandle that was created by fh_compose, and that
-		 * is about to be used in another nfsv4 compound
-		 * operation.
-		 */
-		error = nfsd_setuser_and_check_port(rqstp, exp);
-		if (error)
-			goto out;
 	}
+	dentry = fhp->fh_dentry;
+	exp = fhp->fh_export;
+	/*
+	 * We still have to do all these permission checks, even when
+	 * fh_dentry is already set:
+	 * 	- fh_verify may be called multiple times with different
+	 * 	  "access" arguments (e.g. nfsd_proc_create calls
+	 * 	  fh_verify(...,NFSD_MAY_EXEC) first, then later (in
+	 * 	  nfsd_create) calls fh_verify(...,NFSD_MAY_CREATE).
+	 *	- in the NFSv4 case, the filehandle may have been filled
+	 *	  in by fh_compose, and given a dentry, but further
+	 *	  compound operations performed with that filehandle
+	 *	  still need permissions checks.  In the worst case, a
+	 *	  mountpoint crossing may have changed the export
+	 *	  options, and we may now need to use a different uid
+	 *	  (for example, if different id-squashing options are in
+	 *	  effect on the new filesystem).
+	 */
+	error = check_pseudo_root(rqstp, dentry, exp);
+	if (error)
+		goto out;
+
+	error = nfsd_setuser_and_check_port(rqstp, exp);
+	if (error)
+		goto out;
 
 	error = nfsd_mode_check(rqstp, dentry->d_inode->i_mode, type);
 	if (error)
diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h
new file mode 100644
index 000000000000..cdfb8c6a4206
--- /dev/null
+++ b/fs/nfsd/nfsfh.h
@@ -0,0 +1,208 @@
+/* Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de> */
+
+#ifndef _LINUX_NFSD_FH_INT_H
+#define _LINUX_NFSD_FH_INT_H
+
+#include <linux/nfsd/nfsfh.h>
+
+enum nfsd_fsid {
+	FSID_DEV = 0,
+	FSID_NUM,
+	FSID_MAJOR_MINOR,
+	FSID_ENCODE_DEV,
+	FSID_UUID4_INUM,
+	FSID_UUID8,
+	FSID_UUID16,
+	FSID_UUID16_INUM,
+};
+
+enum fsid_source {
+	FSIDSOURCE_DEV,
+	FSIDSOURCE_FSID,
+	FSIDSOURCE_UUID,
+};
+extern enum fsid_source fsid_source(struct svc_fh *fhp);
+
+
+/* This might look a little large to "inline" but in all calls except
+ * one, 'vers' is constant so moste of the function disappears.
+ */
+static inline void mk_fsid(int vers, u32 *fsidv, dev_t dev, ino_t ino,
+			   u32 fsid, unsigned char *uuid)
+{
+	u32 *up;
+	switch(vers) {
+	case FSID_DEV:
+		fsidv[0] = htonl((MAJOR(dev)<<16) |
+				 MINOR(dev));
+		fsidv[1] = ino_t_to_u32(ino);
+		break;
+	case FSID_NUM:
+		fsidv[0] = fsid;
+		break;
+	case FSID_MAJOR_MINOR:
+		fsidv[0] = htonl(MAJOR(dev));
+		fsidv[1] = htonl(MINOR(dev));
+		fsidv[2] = ino_t_to_u32(ino);
+		break;
+
+	case FSID_ENCODE_DEV:
+		fsidv[0] = new_encode_dev(dev);
+		fsidv[1] = ino_t_to_u32(ino);
+		break;
+
+	case FSID_UUID4_INUM:
+		/* 4 byte fsid and inode number */
+		up = (u32*)uuid;
+		fsidv[0] = ino_t_to_u32(ino);
+		fsidv[1] = up[0] ^ up[1] ^ up[2] ^ up[3];
+		break;
+
+	case FSID_UUID8:
+		/* 8 byte fsid  */
+		up = (u32*)uuid;
+		fsidv[0] = up[0] ^ up[2];
+		fsidv[1] = up[1] ^ up[3];
+		break;
+
+	case FSID_UUID16:
+		/* 16 byte fsid - NFSv3+ only */
+		memcpy(fsidv, uuid, 16);
+		break;
+
+	case FSID_UUID16_INUM:
+		/* 8 byte inode and 16 byte fsid */
+		*(u64*)fsidv = (u64)ino;
+		memcpy(fsidv+2, uuid, 16);
+		break;
+	default: BUG();
+	}
+}
+
+static inline int key_len(int type)
+{
+	switch(type) {
+	case FSID_DEV:		return 8;
+	case FSID_NUM: 		return 4;
+	case FSID_MAJOR_MINOR:	return 12;
+	case FSID_ENCODE_DEV:	return 8;
+	case FSID_UUID4_INUM:	return 8;
+	case FSID_UUID8:	return 8;
+	case FSID_UUID16:	return 16;
+	case FSID_UUID16_INUM:	return 24;
+	default: return 0;
+	}
+}
+
+/*
+ * Shorthand for dprintk()'s
+ */
+extern char * SVCFH_fmt(struct svc_fh *fhp);
+
+/*
+ * Function prototypes
+ */
+__be32	fh_verify(struct svc_rqst *, struct svc_fh *, int, int);
+__be32	fh_compose(struct svc_fh *, struct svc_export *, struct dentry *, struct svc_fh *);
+__be32	fh_update(struct svc_fh *);
+void	fh_put(struct svc_fh *);
+
+static __inline__ struct svc_fh *
+fh_copy(struct svc_fh *dst, struct svc_fh *src)
+{
+	WARN_ON(src->fh_dentry || src->fh_locked);
+			
+	*dst = *src;
+	return dst;
+}
+
+static inline void
+fh_copy_shallow(struct knfsd_fh *dst, struct knfsd_fh *src)
+{
+	dst->fh_size = src->fh_size;
+	memcpy(&dst->fh_base, &src->fh_base, src->fh_size);
+}
+
+static __inline__ struct svc_fh *
+fh_init(struct svc_fh *fhp, int maxsize)
+{
+	memset(fhp, 0, sizeof(*fhp));
+	fhp->fh_maxsize = maxsize;
+	return fhp;
+}
+
+#ifdef CONFIG_NFSD_V3
+/*
+ * Fill in the pre_op attr for the wcc data
+ */
+static inline void
+fill_pre_wcc(struct svc_fh *fhp)
+{
+	struct inode    *inode;
+
+	inode = fhp->fh_dentry->d_inode;
+	if (!fhp->fh_pre_saved) {
+		fhp->fh_pre_mtime = inode->i_mtime;
+		fhp->fh_pre_ctime = inode->i_ctime;
+		fhp->fh_pre_size  = inode->i_size;
+		fhp->fh_pre_change = inode->i_version;
+		fhp->fh_pre_saved = 1;
+	}
+}
+
+extern void fill_post_wcc(struct svc_fh *);
+#else
+#define	fill_pre_wcc(ignored)
+#define fill_post_wcc(notused)
+#endif /* CONFIG_NFSD_V3 */
+
+
+/*
+ * Lock a file handle/inode
+ * NOTE: both fh_lock and fh_unlock are done "by hand" in
+ * vfs.c:nfsd_rename as it needs to grab 2 i_mutex's at once
+ * so, any changes here should be reflected there.
+ */
+
+static inline void
+fh_lock_nested(struct svc_fh *fhp, unsigned int subclass)
+{
+	struct dentry	*dentry = fhp->fh_dentry;
+	struct inode	*inode;
+
+	BUG_ON(!dentry);
+
+	if (fhp->fh_locked) {
+		printk(KERN_WARNING "fh_lock: %s/%s already locked!\n",
+			dentry->d_parent->d_name.name, dentry->d_name.name);
+		return;
+	}
+
+	inode = dentry->d_inode;
+	mutex_lock_nested(&inode->i_mutex, subclass);
+	fill_pre_wcc(fhp);
+	fhp->fh_locked = 1;
+}
+
+static inline void
+fh_lock(struct svc_fh *fhp)
+{
+	fh_lock_nested(fhp, I_MUTEX_NORMAL);
+}
+
+/*
+ * Unlock a file handle/inode
+ */
+static inline void
+fh_unlock(struct svc_fh *fhp)
+{
+	BUG_ON(!fhp->fh_dentry);
+
+	if (fhp->fh_locked) {
+		fill_post_wcc(fhp);
+		mutex_unlock(&fhp->fh_dentry->d_inode->i_mutex);
+		fhp->fh_locked = 0;
+	}
+}
+
+#endif /* _LINUX_NFSD_FH_INT_H */
diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c
index 0eb9c820b7a6..a047ad6111ef 100644
--- a/fs/nfsd/nfsproc.c
+++ b/fs/nfsd/nfsproc.c
@@ -1,29 +1,14 @@
 /*
- * nfsproc2.c	Process version 2 NFS requests.
- * linux/fs/nfsd/nfs2proc.c
- * 
  * Process version 2 NFS requests.
  *
  * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
  */
 
-#include <linux/linkage.h>
-#include <linux/time.h>
-#include <linux/errno.h>
-#include <linux/fs.h>
-#include <linux/stat.h>
-#include <linux/fcntl.h>
-#include <linux/net.h>
-#include <linux/in.h>
 #include <linux/namei.h>
-#include <linux/unistd.h>
-#include <linux/slab.h>
 
-#include <linux/sunrpc/clnt.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/cache.h>
-#include <linux/nfsd/xdr.h>
+#include "cache.h"
+#include "xdr.h"
+#include "vfs.h"
 
 typedef struct svc_rqst	svc_rqst;
 typedef struct svc_buf	svc_buf;
@@ -758,6 +743,7 @@ nfserrno (int errno)
 		{ nfserr_io, -ETXTBSY },
 		{ nfserr_notsupp, -EOPNOTSUPP },
 		{ nfserr_toosmall, -ETOOSMALL },
+		{ nfserr_serverfault, -ESERVERFAULT },
 	};
 	int	i;
 
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 67ea83eedd43..171699eb07c8 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -1,6 +1,4 @@
 /*
- * linux/fs/nfsd/nfssvc.c
- *
  * Central processing for nfsd.
  *
  * Authors:	Olaf Kirch (okir@monad.swb.de)
@@ -8,33 +6,19 @@
  * Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
  */
 
-#include <linux/module.h>
 #include <linux/sched.h>
-#include <linux/time.h>
-#include <linux/errno.h>
-#include <linux/nfs.h>
-#include <linux/in.h>
-#include <linux/uio.h>
-#include <linux/unistd.h>
-#include <linux/slab.h>
-#include <linux/smp.h>
 #include <linux/freezer.h>
 #include <linux/fs_struct.h>
-#include <linux/kthread.h>
 #include <linux/swap.h>
 
-#include <linux/sunrpc/types.h>
 #include <linux/sunrpc/stats.h>
-#include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/svcsock.h>
-#include <linux/sunrpc/cache.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/stats.h>
-#include <linux/nfsd/cache.h>
-#include <linux/nfsd/syscall.h>
 #include <linux/lockd/bind.h>
 #include <linux/nfsacl.h>
 #include <linux/seq_file.h>
+#include "nfsd.h"
+#include "cache.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY	NFSDDBG_SVC
 
diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c
index afd08e2c90a5..4ce005dbf3e6 100644
--- a/fs/nfsd/nfsxdr.c
+++ b/fs/nfsd/nfsxdr.c
@@ -1,20 +1,10 @@
 /*
- * linux/fs/nfsd/nfsxdr.c
- *
  * XDR support for nfsd
  *
  * Copyright (C) 1995, 1996 Olaf Kirch <okir@monad.swb.de>
  */
 
-#include <linux/types.h>
-#include <linux/time.h>
-#include <linux/nfs.h>
-#include <linux/vfs.h>
-#include <linux/sunrpc/xdr.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/xdr.h>
-#include <linux/mm.h>
+#include "xdr.h"
 #include "auth.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_XDR
diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h
new file mode 100644
index 000000000000..fefeae27f25e
--- /dev/null
+++ b/fs/nfsd/state.h
@@ -0,0 +1,408 @@
+/*
+ *  Copyright (c) 2001 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Kendrick Smith <kmsmith@umich.edu>
+ *  Andy Adamson <andros@umich.edu>
+ *  
+ *  Redistribution and use in source and binary forms, with or without
+ *  modification, are permitted provided that the following conditions
+ *  are met:
+ *  
+ *  1. Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *  2. Redistributions in binary form must reproduce the above copyright
+ *     notice, this list of conditions and the following disclaimer in the
+ *     documentation and/or other materials provided with the distribution.
+ *  3. Neither the name of the University nor the names of its
+ *     contributors may be used to endorse or promote products derived
+ *     from this software without specific prior written permission.
+ *
+ *  THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ *  WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ *  DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ *  FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *  CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *  SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
+ *  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+ *  LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+ *  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+ *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#ifndef _NFSD4_STATE_H
+#define _NFSD4_STATE_H
+
+#include <linux/nfsd/nfsfh.h>
+#include "nfsfh.h"
+
+typedef struct {
+	u32             cl_boot;
+	u32             cl_id;
+} clientid_t;
+
+typedef struct {
+	u32             so_boot;
+	u32             so_stateownerid;
+	u32             so_fileid;
+} stateid_opaque_t;
+
+typedef struct {
+	u32                     si_generation;
+	stateid_opaque_t        si_opaque;
+} stateid_t;
+#define si_boot           si_opaque.so_boot
+#define si_stateownerid   si_opaque.so_stateownerid
+#define si_fileid         si_opaque.so_fileid
+
+#define STATEID_FMT	"(%08x/%08x/%08x/%08x)"
+#define STATEID_VAL(s) \
+	(s)->si_boot, \
+	(s)->si_stateownerid, \
+	(s)->si_fileid, \
+	(s)->si_generation
+
+struct nfsd4_cb_sequence {
+	/* args/res */
+	u32			cbs_minorversion;
+	struct nfs4_client	*cbs_clp;
+};
+
+struct nfs4_delegation {
+	struct list_head	dl_perfile;
+	struct list_head	dl_perclnt;
+	struct list_head	dl_recall_lru;  /* delegation recalled */
+	atomic_t		dl_count;       /* ref count */
+	struct nfs4_client	*dl_client;
+	struct nfs4_file	*dl_file;
+	struct file_lock	*dl_flock;
+	struct file		*dl_vfs_file;
+	u32			dl_type;
+	time_t			dl_time;
+/* For recall: */
+	u32			dl_ident;
+	stateid_t		dl_stateid;
+	struct knfsd_fh		dl_fh;
+	int			dl_retries;
+};
+
+/* client delegation callback info */
+struct nfs4_cb_conn {
+	/* SETCLIENTID info */
+	struct sockaddr_storage	cb_addr;
+	size_t			cb_addrlen;
+	u32                     cb_prog;
+	u32			cb_minorversion;
+	u32                     cb_ident;	/* minorversion 0 only */
+	/* RPC client info */
+	atomic_t		cb_set;     /* successful CB_NULL call */
+	struct rpc_clnt *       cb_client;
+};
+
+/* Maximum number of slots per session. 160 is useful for long haul TCP */
+#define NFSD_MAX_SLOTS_PER_SESSION     160
+/* Maximum number of operations per session compound */
+#define NFSD_MAX_OPS_PER_COMPOUND	16
+/* Maximum  session per slot cache size */
+#define NFSD_SLOT_CACHE_SIZE		1024
+/* Maximum number of NFSD_SLOT_CACHE_SIZE slots per session */
+#define NFSD_CACHE_SIZE_SLOTS_PER_SESSION	32
+#define NFSD_MAX_MEM_PER_SESSION  \
+		(NFSD_CACHE_SIZE_SLOTS_PER_SESSION * NFSD_SLOT_CACHE_SIZE)
+
+struct nfsd4_slot {
+	bool	sl_inuse;
+	bool	sl_cachethis;
+	u16	sl_opcnt;
+	u32	sl_seqid;
+	__be32	sl_status;
+	u32	sl_datalen;
+	char	sl_data[];
+};
+
+struct nfsd4_channel_attrs {
+	u32		headerpadsz;
+	u32		maxreq_sz;
+	u32		maxresp_sz;
+	u32		maxresp_cached;
+	u32		maxops;
+	u32		maxreqs;
+	u32		nr_rdma_attrs;
+	u32		rdma_attrs;
+};
+
+struct nfsd4_create_session {
+	clientid_t			clientid;
+	struct nfs4_sessionid		sessionid;
+	u32				seqid;
+	u32				flags;
+	struct nfsd4_channel_attrs	fore_channel;
+	struct nfsd4_channel_attrs	back_channel;
+	u32				callback_prog;
+	u32				uid;
+	u32				gid;
+};
+
+/* The single slot clientid cache structure */
+struct nfsd4_clid_slot {
+	u32				sl_seqid;
+	__be32				sl_status;
+	struct nfsd4_create_session	sl_cr_ses;
+};
+
+struct nfsd4_session {
+	struct kref		se_ref;
+	struct list_head	se_hash;	/* hash by sessionid */
+	struct list_head	se_perclnt;
+	u32			se_flags;
+	struct nfs4_client	*se_client;	/* for expire_client */
+	struct nfs4_sessionid	se_sessionid;
+	struct nfsd4_channel_attrs se_fchannel;
+	struct nfsd4_channel_attrs se_bchannel;
+	struct nfsd4_slot	*se_slots[];	/* forward channel slots */
+};
+
+static inline void
+nfsd4_put_session(struct nfsd4_session *ses)
+{
+	extern void free_session(struct kref *kref);
+	kref_put(&ses->se_ref, free_session);
+}
+
+static inline void
+nfsd4_get_session(struct nfsd4_session *ses)
+{
+	kref_get(&ses->se_ref);
+}
+
+/* formatted contents of nfs4_sessionid */
+struct nfsd4_sessionid {
+	clientid_t	clientid;
+	u32		sequence;
+	u32		reserved;
+};
+
+#define HEXDIR_LEN     33 /* hex version of 16 byte md5 of cl_name plus '\0' */
+
+/*
+ * struct nfs4_client - one per client.  Clientids live here.
+ * 	o Each nfs4_client is hashed by clientid.
+ *
+ * 	o Each nfs4_clients is also hashed by name 
+ * 	  (the opaque quantity initially sent by the client to identify itself).
+ * 	  
+ *	o cl_perclient list is used to ensure no dangling stateowner references
+ *	  when we expire the nfs4_client
+ */
+struct nfs4_client {
+	struct list_head	cl_idhash; 	/* hash by cl_clientid.id */
+	struct list_head	cl_strhash; 	/* hash by cl_name */
+	struct list_head	cl_openowners;
+	struct list_head	cl_delegations;
+	struct list_head        cl_lru;         /* tail queue */
+	struct xdr_netobj	cl_name; 	/* id generated by client */
+	char                    cl_recdir[HEXDIR_LEN]; /* recovery dir */
+	nfs4_verifier		cl_verifier; 	/* generated by client */
+	time_t                  cl_time;        /* time of last lease renewal */
+	struct sockaddr_storage	cl_addr; 	/* client ipaddress */
+	u32			cl_flavor;	/* setclientid pseudoflavor */
+	char			*cl_principal;	/* setclientid principal name */
+	struct svc_cred		cl_cred; 	/* setclientid principal */
+	clientid_t		cl_clientid;	/* generated by server */
+	nfs4_verifier		cl_confirm;	/* generated by server */
+	struct nfs4_cb_conn	cl_cb_conn;     /* callback info */
+	atomic_t		cl_count;	/* ref count */
+	u32			cl_firststate;	/* recovery dir creation */
+
+	/* for nfs41 */
+	struct list_head	cl_sessions;
+	struct nfsd4_clid_slot	cl_cs_slot;	/* create_session slot */
+	u32			cl_exchange_flags;
+	struct nfs4_sessionid	cl_sessionid;
+
+	/* for nfs41 callbacks */
+	/* We currently support a single back channel with a single slot */
+	unsigned long		cl_cb_slot_busy;
+	u32			cl_cb_seq_nr;
+	struct svc_xprt		*cl_cb_xprt;	/* 4.1 callback transport */
+	struct rpc_wait_queue	cl_cb_waitq;	/* backchannel callers may */
+						/* wait here for slots */
+};
+
+/* struct nfs4_client_reset
+ * one per old client. Populates reset_str_hashtbl. Filled from conf_id_hashtbl
+ * upon lease reset, or from upcall to state_daemon (to read in state
+ * from non-volitile storage) upon reboot.
+ */
+struct nfs4_client_reclaim {
+	struct list_head	cr_strhash;	/* hash by cr_name */
+	char			cr_recdir[HEXDIR_LEN]; /* recover dir */
+};
+
+static inline void
+update_stateid(stateid_t *stateid)
+{
+	stateid->si_generation++;
+}
+
+/* A reasonable value for REPLAY_ISIZE was estimated as follows:  
+ * The OPEN response, typically the largest, requires 
+ *   4(status) + 8(stateid) + 20(changeinfo) + 4(rflags) +  8(verifier) + 
+ *   4(deleg. type) + 8(deleg. stateid) + 4(deleg. recall flag) + 
+ *   20(deleg. space limit) + ~32(deleg. ace) = 112 bytes 
+ */
+
+#define NFSD4_REPLAY_ISIZE       112 
+
+/*
+ * Replay buffer, where the result of the last seqid-mutating operation 
+ * is cached. 
+ */
+struct nfs4_replay {
+	__be32			rp_status;
+	unsigned int		rp_buflen;
+	char			*rp_buf;
+	unsigned		intrp_allocated;
+	struct knfsd_fh		rp_openfh;
+	char			rp_ibuf[NFSD4_REPLAY_ISIZE];
+};
+
+/*
+* nfs4_stateowner can either be an open_owner, or a lock_owner
+*
+*    so_idhash:  stateid_hashtbl[] for open owner, lockstateid_hashtbl[]
+*         for lock_owner
+*    so_strhash: ownerstr_hashtbl[] for open_owner, lock_ownerstr_hashtbl[]
+*         for lock_owner
+*    so_perclient: nfs4_client->cl_perclient entry - used when nfs4_client
+*         struct is reaped.
+*    so_perfilestate: heads the list of nfs4_stateid (either open or lock) 
+*         and is used to ensure no dangling nfs4_stateid references when we 
+*         release a stateowner.
+*    so_perlockowner: (open) nfs4_stateid->st_perlockowner entry - used when
+*         close is called to reap associated byte-range locks
+*    so_close_lru: (open) stateowner is placed on this list instead of being
+*         reaped (when so_perfilestate is empty) to hold the last close replay.
+*         reaped by laundramat thread after lease period.
+*/
+struct nfs4_stateowner {
+	struct kref		so_ref;
+	struct list_head        so_idhash;   /* hash by so_id */
+	struct list_head        so_strhash;   /* hash by op_name */
+	struct list_head        so_perclient;
+	struct list_head        so_stateids;
+	struct list_head        so_perstateid; /* for lockowners only */
+	struct list_head	so_close_lru; /* tail queue */
+	time_t			so_time; /* time of placement on so_close_lru */
+	int			so_is_open_owner; /* 1=openowner,0=lockowner */
+	u32                     so_id;
+	struct nfs4_client *    so_client;
+	/* after increment in ENCODE_SEQID_OP_TAIL, represents the next
+	 * sequence id expected from the client: */
+	u32                     so_seqid;
+	struct xdr_netobj       so_owner;     /* open owner name */
+	int                     so_confirmed; /* successful OPEN_CONFIRM? */
+	struct nfs4_replay	so_replay;
+};
+
+/*
+*  nfs4_file: a file opened by some number of (open) nfs4_stateowners.
+*    o fi_perfile list is used to search for conflicting 
+*      share_acces, share_deny on the file.
+*/
+struct nfs4_file {
+	atomic_t		fi_ref;
+	struct list_head        fi_hash;    /* hash by "struct inode *" */
+	struct list_head        fi_stateids;
+	struct list_head	fi_delegations;
+	struct inode		*fi_inode;
+	u32                     fi_id;      /* used with stateowner->so_id 
+					     * for stateid_hashtbl hash */
+	bool			fi_had_conflict;
+};
+
+/*
+* nfs4_stateid can either be an open stateid or (eventually) a lock stateid
+*
+* (open)nfs4_stateid: one per (open)nfs4_stateowner, nfs4_file
+*
+* 	st_hash: stateid_hashtbl[] entry or lockstateid_hashtbl entry
+* 	st_perfile: file_hashtbl[] entry.
+* 	st_perfile_state: nfs4_stateowner->so_perfilestate
+*       st_perlockowner: (open stateid) list of lock nfs4_stateowners
+* 	st_access_bmap: used only for open stateid
+* 	st_deny_bmap: used only for open stateid
+*	st_openstp: open stateid lock stateid was derived from
+*
+* XXX: open stateids and lock stateids have diverged sufficiently that
+* we should consider defining separate structs for the two cases.
+*/
+
+struct nfs4_stateid {
+	struct list_head              st_hash; 
+	struct list_head              st_perfile;
+	struct list_head              st_perstateowner;
+	struct list_head              st_lockowners;
+	struct nfs4_stateowner      * st_stateowner;
+	struct nfs4_file            * st_file;
+	stateid_t                     st_stateid;
+	struct file                 * st_vfs_file;
+	unsigned long                 st_access_bmap;
+	unsigned long                 st_deny_bmap;
+	struct nfs4_stateid         * st_openstp;
+};
+
+/* flags for preprocess_seqid_op() */
+#define HAS_SESSION             0x00000001
+#define CONFIRM                 0x00000002
+#define OPEN_STATE              0x00000004
+#define LOCK_STATE              0x00000008
+#define RD_STATE	        0x00000010
+#define WR_STATE	        0x00000020
+#define CLOSE_STATE             0x00000040
+
+#define seqid_mutating_err(err)                       \
+	(((err) != nfserr_stale_clientid) &&    \
+	((err) != nfserr_bad_seqid) &&          \
+	((err) != nfserr_stale_stateid) &&      \
+	((err) != nfserr_bad_stateid))
+
+struct nfsd4_compound_state;
+
+extern __be32 nfs4_preprocess_stateid_op(struct nfsd4_compound_state *cstate,
+		stateid_t *stateid, int flags, struct file **filp);
+extern void nfs4_lock_state(void);
+extern void nfs4_unlock_state(void);
+extern int nfs4_in_grace(void);
+extern __be32 nfs4_check_open_reclaim(clientid_t *clid);
+extern void put_nfs4_client(struct nfs4_client *clp);
+extern void nfs4_free_stateowner(struct kref *kref);
+extern int set_callback_cred(void);
+extern void nfsd4_probe_callback(struct nfs4_client *clp);
+extern void nfsd4_cb_recall(struct nfs4_delegation *dp);
+extern void nfs4_put_delegation(struct nfs4_delegation *dp);
+extern __be32 nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname);
+extern void nfsd4_init_recdir(char *recdir_name);
+extern int nfsd4_recdir_load(void);
+extern void nfsd4_shutdown_recdir(void);
+extern int nfs4_client_to_reclaim(const char *name);
+extern int nfs4_has_reclaimed_state(const char *name, bool use_exchange_id);
+extern void nfsd4_recdir_purge_old(void);
+extern int nfsd4_create_clid_dir(struct nfs4_client *clp);
+extern void nfsd4_remove_clid_dir(struct nfs4_client *clp);
+
+static inline void
+nfs4_put_stateowner(struct nfs4_stateowner *so)
+{
+	kref_put(&so->so_ref, nfs4_free_stateowner);
+}
+
+static inline void
+nfs4_get_stateowner(struct nfs4_stateowner *so)
+{
+	kref_get(&so->so_ref);
+}
+
+#endif   /* NFSD4_STATE_H */
diff --git a/fs/nfsd/stats.c b/fs/nfsd/stats.c
index 71944cddf680..5232d3e8fb2f 100644
--- a/fs/nfsd/stats.c
+++ b/fs/nfsd/stats.c
@@ -1,6 +1,4 @@
 /*
- * linux/fs/nfsd/stats.c
- *
  * procfs-based user access to knfsd statistics
  *
  * /proc/net/rpc/nfsd
@@ -23,18 +21,13 @@
  * Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
  */
 
-#include <linux/kernel.h>
-#include <linux/time.h>
-#include <linux/proc_fs.h>
 #include <linux/seq_file.h>
-#include <linux/stat.h>
 #include <linux/module.h>
-
-#include <linux/sunrpc/svc.h>
 #include <linux/sunrpc/stats.h>
-#include <linux/nfsd/nfsd.h>
 #include <linux/nfsd/stats.h>
 
+#include "nfsd.h"
+
 struct nfsd_stats	nfsdstats;
 struct svc_stat		nfsd_svcstats = {
 	.program	= &nfsd_program,
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index a293f0273263..c194793b642b 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1,7 +1,5 @@
 #define MSNFS	/* HACK HACK */
 /*
- * linux/fs/nfsd/vfs.c
- *
  * File operations used by nfsd. Some of these have been ripped from
  * other parts of the kernel because they weren't exported, others
  * are partial duplicates with added or changed functionality.
@@ -16,48 +14,31 @@
  * Zerocpy NFS support (C) 2002 Hirokazu Takahashi <taka@valinux.co.jp>
  */
 
-#include <linux/string.h>
-#include <linux/time.h>
-#include <linux/errno.h>
 #include <linux/fs.h>
 #include <linux/file.h>
-#include <linux/mount.h>
-#include <linux/major.h>
 #include <linux/splice.h>
-#include <linux/proc_fs.h>
-#include <linux/stat.h>
 #include <linux/fcntl.h>
-#include <linux/net.h>
-#include <linux/unistd.h>
-#include <linux/slab.h>
-#include <linux/pagemap.h>
-#include <linux/in.h>
-#include <linux/module.h>
 #include <linux/namei.h>
-#include <linux/vfs.h>
 #include <linux/delay.h>
-#include <linux/sunrpc/svc.h>
-#include <linux/nfsd/nfsd.h>
-#ifdef CONFIG_NFSD_V3
-#include <linux/nfs3.h>
-#include <linux/nfsd/xdr3.h>
-#endif /* CONFIG_NFSD_V3 */
-#include <linux/nfsd/nfsfh.h>
 #include <linux/quotaops.h>
 #include <linux/fsnotify.h>
-#include <linux/posix_acl.h>
 #include <linux/posix_acl_xattr.h>
 #include <linux/xattr.h>
+#include <linux/jhash.h>
+#include <linux/ima.h>
+#include <asm/uaccess.h>
+
+#ifdef CONFIG_NFSD_V3
+#include "xdr3.h"
+#endif /* CONFIG_NFSD_V3 */
+
 #ifdef CONFIG_NFSD_V4
-#include <linux/nfs4.h>
 #include <linux/nfs4_acl.h>
 #include <linux/nfsd_idmap.h>
-#include <linux/security.h>
 #endif /* CONFIG_NFSD_V4 */
-#include <linux/jhash.h>
-#include <linux/ima.h>
 
-#include <asm/uaccess.h>
+#include "nfsd.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY		NFSDDBG_FILEOP
 
@@ -89,12 +70,6 @@ struct raparm_hbucket {
 #define RAPARM_HASH_MASK	(RAPARM_HASH_SIZE-1)
 static struct raparm_hbucket	raparm_hash[RAPARM_HASH_SIZE];
 
-static inline int
-nfsd_v4client(struct svc_rqst *rq)
-{
-    return rq->rq_prog == NFS_PROGRAM && rq->rq_vers == 4;
-}
-
 /* 
  * Called from nfsd_lookup and encode_dirent. Check if we have crossed 
  * a mount point.
@@ -116,8 +91,16 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp,
 
 	exp2 = rqst_exp_get_by_name(rqstp, &path);
 	if (IS_ERR(exp2)) {
-		if (PTR_ERR(exp2) != -ENOENT)
-			err = PTR_ERR(exp2);
+		err = PTR_ERR(exp2);
+		/*
+		 * We normally allow NFS clients to continue
+		 * "underneath" a mountpoint that is not exported.
+		 * The exception is V4ROOT, where no traversal is ever
+		 * allowed without an explicit export of the new
+		 * directory.
+		 */
+		if (err == -ENOENT && !(exp->ex_flags & NFSEXP_V4ROOT))
+			err = 0;
 		path_put(&path);
 		goto out;
 	}
@@ -141,6 +124,53 @@ out:
 	return err;
 }
 
+static void follow_to_parent(struct path *path)
+{
+	struct dentry *dp;
+
+	while (path->dentry == path->mnt->mnt_root && follow_up(path))
+		;
+	dp = dget_parent(path->dentry);
+	dput(path->dentry);
+	path->dentry = dp;
+}
+
+static int nfsd_lookup_parent(struct svc_rqst *rqstp, struct dentry *dparent, struct svc_export **exp, struct dentry **dentryp)
+{
+	struct svc_export *exp2;
+	struct path path = {.mnt = mntget((*exp)->ex_path.mnt),
+			    .dentry = dget(dparent)};
+
+	follow_to_parent(&path);
+
+	exp2 = rqst_exp_parent(rqstp, &path);
+	if (PTR_ERR(exp2) == -ENOENT) {
+		*dentryp = dget(dparent);
+	} else if (IS_ERR(exp2)) {
+		path_put(&path);
+		return PTR_ERR(exp2);
+	} else {
+		*dentryp = dget(path.dentry);
+		exp_put(*exp);
+		*exp = exp2;
+	}
+	path_put(&path);
+	return 0;
+}
+
+/*
+ * For nfsd purposes, we treat V4ROOT exports as though there was an
+ * export at *every* directory.
+ */
+int nfsd_mountpoint(struct dentry *dentry, struct svc_export *exp)
+{
+	if (d_mountpoint(dentry))
+		return 1;
+	if (!(exp->ex_flags & NFSEXP_V4ROOT))
+		return 0;
+	return dentry->d_inode != NULL;
+}
+
 __be32
 nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		   const char *name, unsigned int len,
@@ -169,35 +199,13 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp,
 			dentry = dget(dparent);
 		else if (dparent != exp->ex_path.dentry)
 			dentry = dget_parent(dparent);
-		else if (!EX_NOHIDE(exp))
+		else if (!EX_NOHIDE(exp) && !nfsd_v4client(rqstp))
 			dentry = dget(dparent); /* .. == . just like at / */
 		else {
 			/* checking mountpoint crossing is very different when stepping up */
-			struct svc_export *exp2 = NULL;
-			struct dentry *dp;
-			struct path path = {.mnt = mntget(exp->ex_path.mnt),
-					    .dentry = dget(dparent)};
-
-			while (path.dentry == path.mnt->mnt_root &&
-			       follow_up(&path))
-				;
-			dp = dget_parent(path.dentry);
-			dput(path.dentry);
-			path.dentry = dp;
-
-			exp2 = rqst_exp_parent(rqstp, &path);
-			if (PTR_ERR(exp2) == -ENOENT) {
-				dentry = dget(dparent);
-			} else if (IS_ERR(exp2)) {
-				host_err = PTR_ERR(exp2);
-				path_put(&path);
+			host_err = nfsd_lookup_parent(rqstp, dparent, &exp, &dentry);
+			if (host_err)
 				goto out_nfserr;
-			} else {
-				dentry = dget(path.dentry);
-				exp_put(exp);
-				exp = exp2;
-			}
-			path_put(&path);
 		}
 	} else {
 		fh_lock(fhp);
@@ -208,7 +216,7 @@ nfsd_lookup_dentry(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		/*
 		 * check if we have crossed a mount point ...
 		 */
-		if (d_mountpoint(dentry)) {
+		if (nfsd_mountpoint(dentry, exp)) {
 			if ((host_err = nfsd_cross_mnt(rqstp, &dentry, &exp))) {
 				dput(dentry);
 				goto out_nfserr;
@@ -744,8 +752,6 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type,
 			    flags, current_cred());
 	if (IS_ERR(*filp))
 		host_err = PTR_ERR(*filp);
-	else
-		ima_counts_get(*filp);
 out_nfserr:
 	err = nfserrno(host_err);
 out:
@@ -774,12 +780,9 @@ static inline int nfsd_dosync(struct file *filp, struct dentry *dp,
 	int (*fsync) (struct file *, struct dentry *, int);
 	int err;
 
-	err = filemap_fdatawrite(inode->i_mapping);
+	err = filemap_write_and_wait(inode->i_mapping);
 	if (err == 0 && fop && (fsync = fop->fsync))
 		err = fsync(filp, dp, 0);
-	if (err == 0)
-		err = filemap_fdatawait(inode->i_mapping);
-
 	return err;
 }
 
@@ -2124,8 +2127,7 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
 	 */
 	path.mnt = exp->ex_path.mnt;
 	path.dentry = dentry;
-	err = ima_path_check(&path, acc & (MAY_READ | MAY_WRITE | MAY_EXEC),
-			     IMA_COUNT_LEAVE);
+	err = ima_path_check(&path, acc & (MAY_READ | MAY_WRITE | MAY_EXEC));
 nfsd_out:
 	return err? nfserrno(err) : 0;
 }
diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h
new file mode 100644
index 000000000000..4b1de0a9ea75
--- /dev/null
+++ b/fs/nfsd/vfs.h
@@ -0,0 +1,101 @@
+/*
+ * Copyright (C) 1995-1997 Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef LINUX_NFSD_VFS_H
+#define LINUX_NFSD_VFS_H
+
+#include "nfsfh.h"
+
+/*
+ * Flags for nfsd_permission
+ */
+#define NFSD_MAY_NOP		0
+#define NFSD_MAY_EXEC		1 /* == MAY_EXEC */
+#define NFSD_MAY_WRITE		2 /* == MAY_WRITE */
+#define NFSD_MAY_READ		4 /* == MAY_READ */
+#define NFSD_MAY_SATTR		8
+#define NFSD_MAY_TRUNC		16
+#define NFSD_MAY_LOCK		32
+#define NFSD_MAY_OWNER_OVERRIDE	64
+#define NFSD_MAY_LOCAL_ACCESS	128 /* IRIX doing local access check on device special file*/
+#define NFSD_MAY_BYPASS_GSS_ON_ROOT 256
+
+#define NFSD_MAY_CREATE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE)
+#define NFSD_MAY_REMOVE		(NFSD_MAY_EXEC|NFSD_MAY_WRITE|NFSD_MAY_TRUNC)
+
+/*
+ * Callback function for readdir
+ */
+typedef int (*nfsd_dirop_t)(struct inode *, struct dentry *, int, int);
+
+/* nfsd/vfs.c */
+int		fh_lock_parent(struct svc_fh *, struct dentry *);
+int		nfsd_racache_init(int);
+void		nfsd_racache_shutdown(void);
+int		nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp,
+		                struct svc_export **expp);
+__be32		nfsd_lookup(struct svc_rqst *, struct svc_fh *,
+				const char *, unsigned int, struct svc_fh *);
+__be32		 nfsd_lookup_dentry(struct svc_rqst *, struct svc_fh *,
+				const char *, unsigned int,
+				struct svc_export **, struct dentry **);
+__be32		nfsd_setattr(struct svc_rqst *, struct svc_fh *,
+				struct iattr *, int, time_t);
+int nfsd_mountpoint(struct dentry *, struct svc_export *);
+#ifdef CONFIG_NFSD_V4
+__be32          nfsd4_set_nfs4_acl(struct svc_rqst *, struct svc_fh *,
+                    struct nfs4_acl *);
+int             nfsd4_get_nfs4_acl(struct svc_rqst *, struct dentry *, struct nfs4_acl **);
+#endif /* CONFIG_NFSD_V4 */
+__be32		nfsd_create(struct svc_rqst *, struct svc_fh *,
+				char *name, int len, struct iattr *attrs,
+				int type, dev_t rdev, struct svc_fh *res);
+#ifdef CONFIG_NFSD_V3
+__be32		nfsd_access(struct svc_rqst *, struct svc_fh *, u32 *, u32 *);
+__be32		nfsd_create_v3(struct svc_rqst *, struct svc_fh *,
+				char *name, int len, struct iattr *attrs,
+				struct svc_fh *res, int createmode,
+				u32 *verifier, int *truncp, int *created);
+__be32		nfsd_commit(struct svc_rqst *, struct svc_fh *,
+				loff_t, unsigned long);
+#endif /* CONFIG_NFSD_V3 */
+__be32		nfsd_open(struct svc_rqst *, struct svc_fh *, int,
+				int, struct file **);
+void		nfsd_close(struct file *);
+__be32 		nfsd_read(struct svc_rqst *, struct svc_fh *, struct file *,
+				loff_t, struct kvec *, int, unsigned long *);
+__be32 		nfsd_write(struct svc_rqst *, struct svc_fh *,struct file *,
+				loff_t, struct kvec *,int, unsigned long *, int *);
+__be32		nfsd_readlink(struct svc_rqst *, struct svc_fh *,
+				char *, int *);
+__be32		nfsd_symlink(struct svc_rqst *, struct svc_fh *,
+				char *name, int len, char *path, int plen,
+				struct svc_fh *res, struct iattr *);
+__be32		nfsd_link(struct svc_rqst *, struct svc_fh *,
+				char *, int, struct svc_fh *);
+__be32		nfsd_rename(struct svc_rqst *,
+				struct svc_fh *, char *, int,
+				struct svc_fh *, char *, int);
+__be32		nfsd_remove(struct svc_rqst *,
+				struct svc_fh *, char *, int);
+__be32		nfsd_unlink(struct svc_rqst *, struct svc_fh *, int type,
+				char *name, int len);
+int		nfsd_truncate(struct svc_rqst *, struct svc_fh *,
+				unsigned long size);
+__be32		nfsd_readdir(struct svc_rqst *, struct svc_fh *,
+			     loff_t *, struct readdir_cd *, filldir_t);
+__be32		nfsd_statfs(struct svc_rqst *, struct svc_fh *,
+				struct kstatfs *, int access);
+
+int		nfsd_notify_change(struct inode *, struct iattr *);
+__be32		nfsd_permission(struct svc_rqst *, struct svc_export *,
+				struct dentry *, int);
+int		nfsd_sync_dir(struct dentry *dp);
+
+#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
+struct posix_acl *nfsd_get_posix_acl(struct svc_fh *, int);
+int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *);
+#endif
+
+#endif /* LINUX_NFSD_VFS_H */
diff --git a/fs/nfsd/xdr.h b/fs/nfsd/xdr.h
new file mode 100644
index 000000000000..53b1863dd8f6
--- /dev/null
+++ b/fs/nfsd/xdr.h
@@ -0,0 +1,173 @@
+/* XDR types for nfsd. This is mainly a typing exercise. */
+
+#ifndef LINUX_NFSD_H
+#define LINUX_NFSD_H
+
+#include <linux/vfs.h>
+#include "nfsd.h"
+#include "nfsfh.h"
+
+struct nfsd_fhandle {
+	struct svc_fh		fh;
+};
+
+struct nfsd_sattrargs {
+	struct svc_fh		fh;
+	struct iattr		attrs;
+};
+
+struct nfsd_diropargs {
+	struct svc_fh		fh;
+	char *			name;
+	unsigned int		len;
+};
+
+struct nfsd_readargs {
+	struct svc_fh		fh;
+	__u32			offset;
+	__u32			count;
+	int			vlen;
+};
+
+struct nfsd_writeargs {
+	svc_fh			fh;
+	__u32			offset;
+	int			len;
+	int			vlen;
+};
+
+struct nfsd_createargs {
+	struct svc_fh		fh;
+	char *			name;
+	unsigned int		len;
+	struct iattr		attrs;
+};
+
+struct nfsd_renameargs {
+	struct svc_fh		ffh;
+	char *			fname;
+	unsigned int		flen;
+	struct svc_fh		tfh;
+	char *			tname;
+	unsigned int		tlen;
+};
+
+struct nfsd_readlinkargs {
+	struct svc_fh		fh;
+	char *			buffer;
+};
+	
+struct nfsd_linkargs {
+	struct svc_fh		ffh;
+	struct svc_fh		tfh;
+	char *			tname;
+	unsigned int		tlen;
+};
+
+struct nfsd_symlinkargs {
+	struct svc_fh		ffh;
+	char *			fname;
+	unsigned int		flen;
+	char *			tname;
+	unsigned int		tlen;
+	struct iattr		attrs;
+};
+
+struct nfsd_readdirargs {
+	struct svc_fh		fh;
+	__u32			cookie;
+	__u32			count;
+	__be32 *		buffer;
+};
+
+struct nfsd_attrstat {
+	struct svc_fh		fh;
+	struct kstat		stat;
+};
+
+struct nfsd_diropres  {
+	struct svc_fh		fh;
+	struct kstat		stat;
+};
+
+struct nfsd_readlinkres {
+	int			len;
+};
+
+struct nfsd_readres {
+	struct svc_fh		fh;
+	unsigned long		count;
+	struct kstat		stat;
+};
+
+struct nfsd_readdirres {
+	int			count;
+
+	struct readdir_cd	common;
+	__be32 *		buffer;
+	int			buflen;
+	__be32 *		offset;
+};
+
+struct nfsd_statfsres {
+	struct kstatfs		stats;
+};
+
+/*
+ * Storage requirements for XDR arguments and results.
+ */
+union nfsd_xdrstore {
+	struct nfsd_sattrargs	sattr;
+	struct nfsd_diropargs	dirop;
+	struct nfsd_readargs	read;
+	struct nfsd_writeargs	write;
+	struct nfsd_createargs	create;
+	struct nfsd_renameargs	rename;
+	struct nfsd_linkargs	link;
+	struct nfsd_symlinkargs	symlink;
+	struct nfsd_readdirargs	readdir;
+};
+
+#define NFS2_SVC_XDRSIZE	sizeof(union nfsd_xdrstore)
+
+
+int nfssvc_decode_void(struct svc_rqst *, __be32 *, void *);
+int nfssvc_decode_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *);
+int nfssvc_decode_sattrargs(struct svc_rqst *, __be32 *,
+				struct nfsd_sattrargs *);
+int nfssvc_decode_diropargs(struct svc_rqst *, __be32 *,
+				struct nfsd_diropargs *);
+int nfssvc_decode_readargs(struct svc_rqst *, __be32 *,
+				struct nfsd_readargs *);
+int nfssvc_decode_writeargs(struct svc_rqst *, __be32 *,
+				struct nfsd_writeargs *);
+int nfssvc_decode_createargs(struct svc_rqst *, __be32 *,
+				struct nfsd_createargs *);
+int nfssvc_decode_renameargs(struct svc_rqst *, __be32 *,
+				struct nfsd_renameargs *);
+int nfssvc_decode_readlinkargs(struct svc_rqst *, __be32 *,
+				struct nfsd_readlinkargs *);
+int nfssvc_decode_linkargs(struct svc_rqst *, __be32 *,
+				struct nfsd_linkargs *);
+int nfssvc_decode_symlinkargs(struct svc_rqst *, __be32 *,
+				struct nfsd_symlinkargs *);
+int nfssvc_decode_readdirargs(struct svc_rqst *, __be32 *,
+				struct nfsd_readdirargs *);
+int nfssvc_encode_void(struct svc_rqst *, __be32 *, void *);
+int nfssvc_encode_attrstat(struct svc_rqst *, __be32 *, struct nfsd_attrstat *);
+int nfssvc_encode_diropres(struct svc_rqst *, __be32 *, struct nfsd_diropres *);
+int nfssvc_encode_readlinkres(struct svc_rqst *, __be32 *, struct nfsd_readlinkres *);
+int nfssvc_encode_readres(struct svc_rqst *, __be32 *, struct nfsd_readres *);
+int nfssvc_encode_statfsres(struct svc_rqst *, __be32 *, struct nfsd_statfsres *);
+int nfssvc_encode_readdirres(struct svc_rqst *, __be32 *, struct nfsd_readdirres *);
+
+int nfssvc_encode_entry(void *, const char *name,
+			int namlen, loff_t offset, u64 ino, unsigned int);
+
+int nfssvc_release_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *);
+
+/* Helper functions for NFSv2 ACL code */
+__be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp);
+__be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp);
+
+#endif /* LINUX_NFSD_H */
diff --git a/fs/nfsd/xdr3.h b/fs/nfsd/xdr3.h
new file mode 100644
index 000000000000..7df980eb0562
--- /dev/null
+++ b/fs/nfsd/xdr3.h
@@ -0,0 +1,344 @@
+/*
+ * XDR types for NFSv3 in nfsd.
+ *
+ * Copyright (C) 1996-1998, Olaf Kirch <okir@monad.swb.de>
+ */
+
+#ifndef _LINUX_NFSD_XDR3_H
+#define _LINUX_NFSD_XDR3_H
+
+#include "xdr.h"
+
+struct nfsd3_sattrargs {
+	struct svc_fh		fh;
+	struct iattr		attrs;
+	int			check_guard;
+	time_t			guardtime;
+};
+
+struct nfsd3_diropargs {
+	struct svc_fh		fh;
+	char *			name;
+	unsigned int		len;
+};
+
+struct nfsd3_accessargs {
+	struct svc_fh		fh;
+	unsigned int		access;
+};
+
+struct nfsd3_readargs {
+	struct svc_fh		fh;
+	__u64			offset;
+	__u32			count;
+	int			vlen;
+};
+
+struct nfsd3_writeargs {
+	svc_fh			fh;
+	__u64			offset;
+	__u32			count;
+	int			stable;
+	__u32			len;
+	int			vlen;
+};
+
+struct nfsd3_createargs {
+	struct svc_fh		fh;
+	char *			name;
+	unsigned int		len;
+	int			createmode;
+	struct iattr		attrs;
+	__be32 *		verf;
+};
+
+struct nfsd3_mknodargs {
+	struct svc_fh		fh;
+	char *			name;
+	unsigned int		len;
+	__u32			ftype;
+	__u32			major, minor;
+	struct iattr		attrs;
+};
+
+struct nfsd3_renameargs {
+	struct svc_fh		ffh;
+	char *			fname;
+	unsigned int		flen;
+	struct svc_fh		tfh;
+	char *			tname;
+	unsigned int		tlen;
+};
+
+struct nfsd3_readlinkargs {
+	struct svc_fh		fh;
+	char *			buffer;
+};
+
+struct nfsd3_linkargs {
+	struct svc_fh		ffh;
+	struct svc_fh		tfh;
+	char *			tname;
+	unsigned int		tlen;
+};
+
+struct nfsd3_symlinkargs {
+	struct svc_fh		ffh;
+	char *			fname;
+	unsigned int		flen;
+	char *			tname;
+	unsigned int		tlen;
+	struct iattr		attrs;
+};
+
+struct nfsd3_readdirargs {
+	struct svc_fh		fh;
+	__u64			cookie;
+	__u32			dircount;
+	__u32			count;
+	__be32 *		verf;
+	__be32 *		buffer;
+};
+
+struct nfsd3_commitargs {
+	struct svc_fh		fh;
+	__u64			offset;
+	__u32			count;
+};
+
+struct nfsd3_getaclargs {
+	struct svc_fh		fh;
+	int			mask;
+};
+
+struct posix_acl;
+struct nfsd3_setaclargs {
+	struct svc_fh		fh;
+	int			mask;
+	struct posix_acl	*acl_access;
+	struct posix_acl	*acl_default;
+};
+
+struct nfsd3_attrstat {
+	__be32			status;
+	struct svc_fh		fh;
+	struct kstat            stat;
+};
+
+/* LOOKUP, CREATE, MKDIR, SYMLINK, MKNOD */
+struct nfsd3_diropres  {
+	__be32			status;
+	struct svc_fh		dirfh;
+	struct svc_fh		fh;
+};
+
+struct nfsd3_accessres {
+	__be32			status;
+	struct svc_fh		fh;
+	__u32			access;
+};
+
+struct nfsd3_readlinkres {
+	__be32			status;
+	struct svc_fh		fh;
+	__u32			len;
+};
+
+struct nfsd3_readres {
+	__be32			status;
+	struct svc_fh		fh;
+	unsigned long		count;
+	int			eof;
+};
+
+struct nfsd3_writeres {
+	__be32			status;
+	struct svc_fh		fh;
+	unsigned long		count;
+	int			committed;
+};
+
+struct nfsd3_renameres {
+	__be32			status;
+	struct svc_fh		ffh;
+	struct svc_fh		tfh;
+};
+
+struct nfsd3_linkres {
+	__be32			status;
+	struct svc_fh		tfh;
+	struct svc_fh		fh;
+};
+
+struct nfsd3_readdirres {
+	__be32			status;
+	struct svc_fh		fh;
+	int			count;
+	__be32			verf[2];
+
+	struct readdir_cd	common;
+	__be32 *		buffer;
+	int			buflen;
+	__be32 *		offset;
+	__be32 *		offset1;
+	struct svc_rqst *	rqstp;
+
+};
+
+struct nfsd3_fsstatres {
+	__be32			status;
+	struct kstatfs		stats;
+	__u32			invarsec;
+};
+
+struct nfsd3_fsinfores {
+	__be32			status;
+	__u32			f_rtmax;
+	__u32			f_rtpref;
+	__u32			f_rtmult;
+	__u32			f_wtmax;
+	__u32			f_wtpref;
+	__u32			f_wtmult;
+	__u32			f_dtpref;
+	__u64			f_maxfilesize;
+	__u32			f_properties;
+};
+
+struct nfsd3_pathconfres {
+	__be32			status;
+	__u32			p_link_max;
+	__u32			p_name_max;
+	__u32			p_no_trunc;
+	__u32			p_chown_restricted;
+	__u32			p_case_insensitive;
+	__u32			p_case_preserving;
+};
+
+struct nfsd3_commitres {
+	__be32			status;
+	struct svc_fh		fh;
+};
+
+struct nfsd3_getaclres {
+	__be32			status;
+	struct svc_fh		fh;
+	int			mask;
+	struct posix_acl	*acl_access;
+	struct posix_acl	*acl_default;
+};
+
+/* dummy type for release */
+struct nfsd3_fhandle_pair {
+	__u32			dummy;
+	struct svc_fh		fh1;
+	struct svc_fh		fh2;
+};
+
+/*
+ * Storage requirements for XDR arguments and results.
+ */
+union nfsd3_xdrstore {
+	struct nfsd3_sattrargs		sattrargs;
+	struct nfsd3_diropargs		diropargs;
+	struct nfsd3_readargs		readargs;
+	struct nfsd3_writeargs		writeargs;
+	struct nfsd3_createargs		createargs;
+	struct nfsd3_renameargs		renameargs;
+	struct nfsd3_linkargs		linkargs;
+	struct nfsd3_symlinkargs	symlinkargs;
+	struct nfsd3_readdirargs	readdirargs;
+	struct nfsd3_diropres 		diropres;
+	struct nfsd3_accessres		accessres;
+	struct nfsd3_readlinkres	readlinkres;
+	struct nfsd3_readres		readres;
+	struct nfsd3_writeres		writeres;
+	struct nfsd3_renameres		renameres;
+	struct nfsd3_linkres		linkres;
+	struct nfsd3_readdirres		readdirres;
+	struct nfsd3_fsstatres		fsstatres;
+	struct nfsd3_fsinfores		fsinfores;
+	struct nfsd3_pathconfres	pathconfres;
+	struct nfsd3_commitres		commitres;
+	struct nfsd3_getaclres		getaclres;
+};
+
+#define NFS3_SVC_XDRSIZE		sizeof(union nfsd3_xdrstore)
+
+int nfs3svc_decode_fhandle(struct svc_rqst *, __be32 *, struct nfsd_fhandle *);
+int nfs3svc_decode_sattrargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_sattrargs *);
+int nfs3svc_decode_diropargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_diropargs *);
+int nfs3svc_decode_accessargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_accessargs *);
+int nfs3svc_decode_readargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_readargs *);
+int nfs3svc_decode_writeargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_writeargs *);
+int nfs3svc_decode_createargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_createargs *);
+int nfs3svc_decode_mkdirargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_createargs *);
+int nfs3svc_decode_mknodargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_mknodargs *);
+int nfs3svc_decode_renameargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_renameargs *);
+int nfs3svc_decode_readlinkargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_readlinkargs *);
+int nfs3svc_decode_linkargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_linkargs *);
+int nfs3svc_decode_symlinkargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_symlinkargs *);
+int nfs3svc_decode_readdirargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_readdirargs *);
+int nfs3svc_decode_readdirplusargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_readdirargs *);
+int nfs3svc_decode_commitargs(struct svc_rqst *, __be32 *,
+				struct nfsd3_commitargs *);
+int nfs3svc_encode_voidres(struct svc_rqst *, __be32 *, void *);
+int nfs3svc_encode_attrstat(struct svc_rqst *, __be32 *,
+				struct nfsd3_attrstat *);
+int nfs3svc_encode_wccstat(struct svc_rqst *, __be32 *,
+				struct nfsd3_attrstat *);
+int nfs3svc_encode_diropres(struct svc_rqst *, __be32 *,
+				struct nfsd3_diropres *);
+int nfs3svc_encode_accessres(struct svc_rqst *, __be32 *,
+				struct nfsd3_accessres *);
+int nfs3svc_encode_readlinkres(struct svc_rqst *, __be32 *,
+				struct nfsd3_readlinkres *);
+int nfs3svc_encode_readres(struct svc_rqst *, __be32 *, struct nfsd3_readres *);
+int nfs3svc_encode_writeres(struct svc_rqst *, __be32 *, struct nfsd3_writeres *);
+int nfs3svc_encode_createres(struct svc_rqst *, __be32 *,
+				struct nfsd3_diropres *);
+int nfs3svc_encode_renameres(struct svc_rqst *, __be32 *,
+				struct nfsd3_renameres *);
+int nfs3svc_encode_linkres(struct svc_rqst *, __be32 *,
+				struct nfsd3_linkres *);
+int nfs3svc_encode_readdirres(struct svc_rqst *, __be32 *,
+				struct nfsd3_readdirres *);
+int nfs3svc_encode_fsstatres(struct svc_rqst *, __be32 *,
+				struct nfsd3_fsstatres *);
+int nfs3svc_encode_fsinfores(struct svc_rqst *, __be32 *,
+				struct nfsd3_fsinfores *);
+int nfs3svc_encode_pathconfres(struct svc_rqst *, __be32 *,
+				struct nfsd3_pathconfres *);
+int nfs3svc_encode_commitres(struct svc_rqst *, __be32 *,
+				struct nfsd3_commitres *);
+
+int nfs3svc_release_fhandle(struct svc_rqst *, __be32 *,
+				struct nfsd3_attrstat *);
+int nfs3svc_release_fhandle2(struct svc_rqst *, __be32 *,
+				struct nfsd3_fhandle_pair *);
+int nfs3svc_encode_entry(void *, const char *name,
+				int namlen, loff_t offset, u64 ino,
+				unsigned int);
+int nfs3svc_encode_entry_plus(void *, const char *name,
+				int namlen, loff_t offset, u64 ino,
+				unsigned int);
+/* Helper functions for NFSv3 ACL code */
+__be32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p,
+				struct svc_fh *fhp);
+__be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp);
+
+
+#endif /* _LINUX_NFSD_XDR3_H */
diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h
new file mode 100644
index 000000000000..efa337739534
--- /dev/null
+++ b/fs/nfsd/xdr4.h
@@ -0,0 +1,562 @@
+/*
+ *  Server-side types for NFSv4.
+ *
+ *  Copyright (c) 2002 The Regents of the University of Michigan.
+ *  All rights reserved.
+ *
+ *  Kendrick Smith <kmsmith@umich.edu>
+ *  Andy Adamson   <andros@umich.edu>
+ *
+ *  Redistribution and use in source and binary forms, with or without
+ *  modification, are permitted provided that the following conditions
+ *  are met:
+ *
+ *  1. Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *  2. Redistributions in binary form must reproduce the above copyright
+ *     notice, this list of conditions and the following disclaimer in the
+ *     documentation and/or other materials provided with the distribution.
+ *  3. Neither the name of the University nor the names of its
+ *     contributors may be used to endorse or promote products derived
+ *     from this software without specific prior written permission.
+ *
+ *  THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ *  WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ *  DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ *  FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ *  CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ *  SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
+ *  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
+ *  LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
+ *  NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+ *  SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ *
+ */
+
+#ifndef _LINUX_NFSD_XDR4_H
+#define _LINUX_NFSD_XDR4_H
+
+#include "state.h"
+#include "nfsd.h"
+
+#define NFSD4_MAX_TAGLEN	128
+#define XDR_LEN(n)                     (((n) + 3) & ~3)
+
+struct nfsd4_compound_state {
+	struct svc_fh		current_fh;
+	struct svc_fh		save_fh;
+	struct nfs4_stateowner	*replay_owner;
+	/* For sessions DRC */
+	struct nfsd4_session	*session;
+	struct nfsd4_slot	*slot;
+	__be32			*datap;
+	size_t			iovlen;
+	u32			minorversion;
+	u32			status;
+};
+
+static inline bool nfsd4_has_session(struct nfsd4_compound_state *cs)
+{
+	return cs->slot != NULL;
+}
+
+struct nfsd4_change_info {
+	u32		atomic;
+	bool		change_supported;
+	u32		before_ctime_sec;
+	u32		before_ctime_nsec;
+	u64		before_change;
+	u32		after_ctime_sec;
+	u32		after_ctime_nsec;
+	u64		after_change;
+};
+
+struct nfsd4_access {
+	u32		ac_req_access;      /* request */
+	u32		ac_supported;       /* response */
+	u32		ac_resp_access;     /* response */
+};
+
+struct nfsd4_close {
+	u32		cl_seqid;           /* request */
+	stateid_t	cl_stateid;         /* request+response */
+	struct nfs4_stateowner * cl_stateowner;	/* response */
+};
+
+struct nfsd4_commit {
+	u64		co_offset;          /* request */
+	u32		co_count;           /* request */
+	nfs4_verifier	co_verf;            /* response */
+};
+
+struct nfsd4_create {
+	u32		cr_namelen;         /* request */
+	char *		cr_name;            /* request */
+	u32		cr_type;            /* request */
+	union {                             /* request */
+		struct {
+			u32 namelen;
+			char *name;
+		} link;   /* NF4LNK */
+		struct {
+			u32 specdata1;
+			u32 specdata2;
+		} dev;    /* NF4BLK, NF4CHR */
+	} u;
+	u32		cr_bmval[3];        /* request */
+	struct iattr	cr_iattr;           /* request */
+	struct nfsd4_change_info  cr_cinfo; /* response */
+	struct nfs4_acl *cr_acl;
+};
+#define cr_linklen	u.link.namelen
+#define cr_linkname	u.link.name
+#define cr_specdata1	u.dev.specdata1
+#define cr_specdata2	u.dev.specdata2
+
+struct nfsd4_delegreturn {
+	stateid_t	dr_stateid;
+};
+
+struct nfsd4_getattr {
+	u32		ga_bmval[3];        /* request */
+	struct svc_fh	*ga_fhp;            /* response */
+};
+
+struct nfsd4_link {
+	u32		li_namelen;         /* request */
+	char *		li_name;            /* request */
+	struct nfsd4_change_info  li_cinfo; /* response */
+};
+
+struct nfsd4_lock_denied {
+	clientid_t	ld_clientid;
+	struct nfs4_stateowner   *ld_sop;
+	u64             ld_start;
+	u64             ld_length;
+	u32             ld_type;
+};
+
+struct nfsd4_lock {
+	/* request */
+	u32             lk_type;
+	u32             lk_reclaim;         /* boolean */
+	u64             lk_offset;
+	u64             lk_length;
+	u32             lk_is_new;
+	union {
+		struct {
+			u32             open_seqid;
+			stateid_t       open_stateid;
+			u32             lock_seqid;
+			clientid_t      clientid;
+			struct xdr_netobj owner;
+		} new;
+		struct {
+			stateid_t       lock_stateid;
+			u32             lock_seqid;
+		} old;
+	} v;
+
+	/* response */
+	union {
+		struct {
+			stateid_t               stateid;
+		} ok;
+		struct nfsd4_lock_denied        denied;
+	} u;
+	/* The lk_replay_owner is the open owner in the open_to_lock_owner
+	 * case and the lock owner otherwise: */
+	struct nfs4_stateowner *lk_replay_owner;
+};
+#define lk_new_open_seqid       v.new.open_seqid
+#define lk_new_open_stateid     v.new.open_stateid
+#define lk_new_lock_seqid       v.new.lock_seqid
+#define lk_new_clientid         v.new.clientid
+#define lk_new_owner            v.new.owner
+#define lk_old_lock_stateid     v.old.lock_stateid
+#define lk_old_lock_seqid       v.old.lock_seqid
+
+#define lk_rflags       u.ok.rflags
+#define lk_resp_stateid u.ok.stateid
+#define lk_denied       u.denied
+
+
+struct nfsd4_lockt {
+	u32				lt_type;
+	clientid_t			lt_clientid;
+	struct xdr_netobj		lt_owner;
+	u64				lt_offset;
+	u64				lt_length;
+	struct nfs4_stateowner * 	lt_stateowner;
+	struct nfsd4_lock_denied  	lt_denied;
+};
+
+ 
+struct nfsd4_locku {
+	u32             lu_type;
+	u32             lu_seqid;
+	stateid_t       lu_stateid;
+	u64             lu_offset;
+	u64             lu_length;
+	struct nfs4_stateowner  *lu_stateowner;
+};
+
+
+struct nfsd4_lookup {
+	u32		lo_len;             /* request */
+	char *		lo_name;            /* request */
+};
+
+struct nfsd4_putfh {
+	u32		pf_fhlen;           /* request */
+	char		*pf_fhval;          /* request */
+};
+
+struct nfsd4_open {
+	u32		op_claim_type;      /* request */
+	struct xdr_netobj op_fname;	    /* request - everything but CLAIM_PREV */
+	u32		op_delegate_type;   /* request - CLAIM_PREV only */
+	stateid_t       op_delegate_stateid; /* request - response */
+	u32		op_create;     	    /* request */
+	u32		op_createmode;      /* request */
+	u32		op_bmval[3];        /* request */
+	struct iattr	iattr;              /* UNCHECKED4, GUARDED4, EXCLUSIVE4_1 */
+	nfs4_verifier	verf;               /* EXCLUSIVE4 */
+	clientid_t	op_clientid;        /* request */
+	struct xdr_netobj op_owner;           /* request */
+	u32		op_seqid;           /* request */
+	u32		op_share_access;    /* request */
+	u32		op_share_deny;      /* request */
+	stateid_t	op_stateid;         /* response */
+	u32		op_recall;          /* recall */
+	struct nfsd4_change_info  op_cinfo; /* response */
+	u32		op_rflags;          /* response */
+	int		op_truncate;        /* used during processing */
+	struct nfs4_stateowner *op_stateowner; /* used during processing */
+	struct nfs4_acl *op_acl;
+};
+#define op_iattr	iattr
+#define op_verf		verf
+
+struct nfsd4_open_confirm {
+	stateid_t	oc_req_stateid		/* request */;
+	u32		oc_seqid    		/* request */;
+	stateid_t	oc_resp_stateid		/* response */;
+	struct nfs4_stateowner * oc_stateowner;	/* response */
+};
+
+struct nfsd4_open_downgrade {
+	stateid_t       od_stateid;
+	u32             od_seqid;
+	u32             od_share_access;
+	u32             od_share_deny;
+	struct nfs4_stateowner *od_stateowner;
+};
+
+
+struct nfsd4_read {
+	stateid_t	rd_stateid;         /* request */
+	u64		rd_offset;          /* request */
+	u32		rd_length;          /* request */
+	int		rd_vlen;
+	struct file     *rd_filp;
+	
+	struct svc_rqst *rd_rqstp;          /* response */
+	struct svc_fh * rd_fhp;             /* response */
+};
+
+struct nfsd4_readdir {
+	u64		rd_cookie;          /* request */
+	nfs4_verifier	rd_verf;            /* request */
+	u32		rd_dircount;        /* request */
+	u32		rd_maxcount;        /* request */
+	u32		rd_bmval[3];        /* request */
+	struct svc_rqst *rd_rqstp;          /* response */
+	struct svc_fh * rd_fhp;             /* response */
+
+	struct readdir_cd	common;
+	__be32 *		buffer;
+	int			buflen;
+	__be32 *		offset;
+};
+
+struct nfsd4_release_lockowner {
+	clientid_t        rl_clientid;
+	struct xdr_netobj rl_owner;
+};
+struct nfsd4_readlink {
+	struct svc_rqst *rl_rqstp;          /* request */
+	struct svc_fh *	rl_fhp;             /* request */
+};
+
+struct nfsd4_remove {
+	u32		rm_namelen;         /* request */
+	char *		rm_name;            /* request */
+	struct nfsd4_change_info  rm_cinfo; /* response */
+};
+
+struct nfsd4_rename {
+	u32		rn_snamelen;        /* request */
+	char *		rn_sname;           /* request */
+	u32		rn_tnamelen;        /* request */
+	char *		rn_tname;           /* request */
+	struct nfsd4_change_info  rn_sinfo; /* response */
+	struct nfsd4_change_info  rn_tinfo; /* response */
+};
+
+struct nfsd4_secinfo {
+	u32 si_namelen;					/* request */
+	char *si_name;					/* request */
+	struct svc_export *si_exp;			/* response */
+};
+
+struct nfsd4_setattr {
+	stateid_t	sa_stateid;         /* request */
+	u32		sa_bmval[3];        /* request */
+	struct iattr	sa_iattr;           /* request */
+	struct nfs4_acl *sa_acl;
+};
+
+struct nfsd4_setclientid {
+	nfs4_verifier	se_verf;            /* request */
+	u32		se_namelen;         /* request */
+	char *		se_name;            /* request */
+	u32		se_callback_prog;   /* request */
+	u32		se_callback_netid_len;  /* request */
+	char *		se_callback_netid_val;  /* request */
+	u32		se_callback_addr_len;   /* request */
+	char *		se_callback_addr_val;   /* request */
+	u32		se_callback_ident;  /* request */
+	clientid_t	se_clientid;        /* response */
+	nfs4_verifier	se_confirm;         /* response */
+};
+
+struct nfsd4_setclientid_confirm {
+	clientid_t	sc_clientid;
+	nfs4_verifier	sc_confirm;
+};
+
+/* also used for NVERIFY */
+struct nfsd4_verify {
+	u32		ve_bmval[3];        /* request */
+	u32		ve_attrlen;         /* request */
+	char *		ve_attrval;         /* request */
+};
+
+struct nfsd4_write {
+	stateid_t	wr_stateid;         /* request */
+	u64		wr_offset;          /* request */
+	u32		wr_stable_how;      /* request */
+	u32		wr_buflen;          /* request */
+	int		wr_vlen;
+
+	u32		wr_bytes_written;   /* response */
+	u32		wr_how_written;     /* response */
+	nfs4_verifier	wr_verifier;        /* response */
+};
+
+struct nfsd4_exchange_id {
+	nfs4_verifier	verifier;
+	struct xdr_netobj clname;
+	u32		flags;
+	clientid_t	clientid;
+	u32		seqid;
+	int		spa_how;
+};
+
+struct nfsd4_sequence {
+	struct nfs4_sessionid	sessionid;		/* request/response */
+	u32			seqid;			/* request/response */
+	u32			slotid;			/* request/response */
+	u32			maxslots;		/* request/response */
+	u32			cachethis;		/* request */
+#if 0
+	u32			target_maxslots;	/* response */
+	u32			status_flags;		/* response */
+#endif /* not yet */
+};
+
+struct nfsd4_destroy_session {
+	struct nfs4_sessionid	sessionid;
+};
+
+struct nfsd4_op {
+	int					opnum;
+	__be32					status;
+	union {
+		struct nfsd4_access		access;
+		struct nfsd4_close		close;
+		struct nfsd4_commit		commit;
+		struct nfsd4_create		create;
+		struct nfsd4_delegreturn	delegreturn;
+		struct nfsd4_getattr		getattr;
+		struct svc_fh *			getfh;
+		struct nfsd4_link		link;
+		struct nfsd4_lock		lock;
+		struct nfsd4_lockt		lockt;
+		struct nfsd4_locku		locku;
+		struct nfsd4_lookup		lookup;
+		struct nfsd4_verify		nverify;
+		struct nfsd4_open		open;
+		struct nfsd4_open_confirm	open_confirm;
+		struct nfsd4_open_downgrade	open_downgrade;
+		struct nfsd4_putfh		putfh;
+		struct nfsd4_read		read;
+		struct nfsd4_readdir		readdir;
+		struct nfsd4_readlink		readlink;
+		struct nfsd4_remove		remove;
+		struct nfsd4_rename		rename;
+		clientid_t			renew;
+		struct nfsd4_secinfo		secinfo;
+		struct nfsd4_setattr		setattr;
+		struct nfsd4_setclientid	setclientid;
+		struct nfsd4_setclientid_confirm setclientid_confirm;
+		struct nfsd4_verify		verify;
+		struct nfsd4_write		write;
+		struct nfsd4_release_lockowner	release_lockowner;
+
+		/* NFSv4.1 */
+		struct nfsd4_exchange_id	exchange_id;
+		struct nfsd4_create_session	create_session;
+		struct nfsd4_destroy_session	destroy_session;
+		struct nfsd4_sequence		sequence;
+	} u;
+	struct nfs4_replay *			replay;
+};
+
+struct nfsd4_compoundargs {
+	/* scratch variables for XDR decode */
+	__be32 *			p;
+	__be32 *			end;
+	struct page **			pagelist;
+	int				pagelen;
+	__be32				tmp[8];
+	__be32 *			tmpp;
+	struct tmpbuf {
+		struct tmpbuf *next;
+		void (*release)(const void *);
+		void *buf;
+	}				*to_free;
+
+	struct svc_rqst			*rqstp;
+
+	u32				taglen;
+	char *				tag;
+	u32				minorversion;
+	u32				opcnt;
+	struct nfsd4_op			*ops;
+	struct nfsd4_op			iops[8];
+};
+
+struct nfsd4_compoundres {
+	/* scratch variables for XDR encode */
+	__be32 *			p;
+	__be32 *			end;
+	struct xdr_buf *		xbuf;
+	struct svc_rqst *		rqstp;
+
+	u32				taglen;
+	char *				tag;
+	u32				opcnt;
+	__be32 *			tagp; /* tag, opcount encode location */
+	struct nfsd4_compound_state	cstate;
+};
+
+static inline bool nfsd4_is_solo_sequence(struct nfsd4_compoundres *resp)
+{
+	struct nfsd4_compoundargs *args = resp->rqstp->rq_argp;
+	return resp->opcnt == 1 && args->ops[0].opnum == OP_SEQUENCE;
+}
+
+static inline bool nfsd4_not_cached(struct nfsd4_compoundres *resp)
+{
+	return !resp->cstate.slot->sl_cachethis || nfsd4_is_solo_sequence(resp);
+}
+
+#define NFS4_SVC_XDRSIZE		sizeof(struct nfsd4_compoundargs)
+
+static inline void
+set_change_info(struct nfsd4_change_info *cinfo, struct svc_fh *fhp)
+{
+	BUG_ON(!fhp->fh_pre_saved || !fhp->fh_post_saved);
+	cinfo->atomic = 1;
+	cinfo->change_supported = IS_I_VERSION(fhp->fh_dentry->d_inode);
+	if (cinfo->change_supported) {
+		cinfo->before_change = fhp->fh_pre_change;
+		cinfo->after_change = fhp->fh_post_change;
+	} else {
+		cinfo->before_ctime_sec = fhp->fh_pre_ctime.tv_sec;
+		cinfo->before_ctime_nsec = fhp->fh_pre_ctime.tv_nsec;
+		cinfo->after_ctime_sec = fhp->fh_post_attr.ctime.tv_sec;
+		cinfo->after_ctime_nsec = fhp->fh_post_attr.ctime.tv_nsec;
+	}
+}
+
+int nfs4svc_encode_voidres(struct svc_rqst *, __be32 *, void *);
+int nfs4svc_decode_compoundargs(struct svc_rqst *, __be32 *,
+		struct nfsd4_compoundargs *);
+int nfs4svc_encode_compoundres(struct svc_rqst *, __be32 *,
+		struct nfsd4_compoundres *);
+void nfsd4_encode_operation(struct nfsd4_compoundres *, struct nfsd4_op *);
+void nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op);
+__be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp,
+		       struct dentry *dentry, __be32 *buffer, int *countp,
+		       u32 *bmval, struct svc_rqst *, int ignore_crossmnt);
+extern __be32 nfsd4_setclientid(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *,
+		struct nfsd4_setclientid *setclid);
+extern __be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *,
+		struct nfsd4_setclientid_confirm *setclientid_confirm);
+extern void nfsd4_store_cache_entry(struct nfsd4_compoundres *resp);
+extern __be32 nfsd4_replay_cache_entry(struct nfsd4_compoundres *resp,
+		struct nfsd4_sequence *seq);
+extern __be32 nfsd4_exchange_id(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *,
+struct nfsd4_exchange_id *);
+		extern __be32 nfsd4_create_session(struct svc_rqst *,
+		struct nfsd4_compound_state *,
+		struct nfsd4_create_session *);
+extern __be32 nfsd4_sequence(struct svc_rqst *,
+		struct nfsd4_compound_state *,
+		struct nfsd4_sequence *);
+extern __be32 nfsd4_destroy_session(struct svc_rqst *,
+		struct nfsd4_compound_state *,
+		struct nfsd4_destroy_session *);
+extern __be32 nfsd4_process_open1(struct nfsd4_compound_state *,
+		struct nfsd4_open *open);
+extern __be32 nfsd4_process_open2(struct svc_rqst *rqstp,
+		struct svc_fh *current_fh, struct nfsd4_open *open);
+extern __be32 nfsd4_open_confirm(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *, struct nfsd4_open_confirm *oc);
+extern __be32 nfsd4_close(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *,
+		struct nfsd4_close *close);
+extern __be32 nfsd4_open_downgrade(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *,
+		struct nfsd4_open_downgrade *od);
+extern __be32 nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *,
+		struct nfsd4_lock *lock);
+extern __be32 nfsd4_lockt(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *,
+		struct nfsd4_lockt *lockt);
+extern __be32 nfsd4_locku(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *,
+		struct nfsd4_locku *locku);
+extern __be32
+nfsd4_release_lockowner(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *,
+		struct nfsd4_release_lockowner *rlockowner);
+extern void nfsd4_release_compoundargs(struct nfsd4_compoundargs *);
+extern __be32 nfsd4_delegreturn(struct svc_rqst *rqstp,
+		struct nfsd4_compound_state *, struct nfsd4_delegreturn *dr);
+extern __be32 nfsd4_renew(struct svc_rqst *rqstp,
+			  struct nfsd4_compound_state *, clientid_t *clid);
+#endif
+
+/*
+ * Local variables:
+ *  c-basic-offset: 8
+ * End:
+ */
diff --git a/fs/nilfs2/alloc.c b/fs/nilfs2/alloc.c
index d69e6ae59251..3f959f1879d8 100644
--- a/fs/nilfs2/alloc.c
+++ b/fs/nilfs2/alloc.c
@@ -142,29 +142,75 @@ static void nilfs_palloc_desc_block_init(struct inode *inode,
 	}
 }
 
+static int nilfs_palloc_get_block(struct inode *inode, unsigned long blkoff,
+				  int create,
+				  void (*init_block)(struct inode *,
+						     struct buffer_head *,
+						     void *),
+				  struct buffer_head **bhp,
+				  struct nilfs_bh_assoc *prev,
+				  spinlock_t *lock)
+{
+	int ret;
+
+	spin_lock(lock);
+	if (prev->bh && blkoff == prev->blkoff) {
+		get_bh(prev->bh);
+		*bhp = prev->bh;
+		spin_unlock(lock);
+		return 0;
+	}
+	spin_unlock(lock);
+
+	ret = nilfs_mdt_get_block(inode, blkoff, create, init_block, bhp);
+	if (!ret) {
+		spin_lock(lock);
+		/*
+		 * The following code must be safe for change of the
+		 * cache contents during the get block call.
+		 */
+		brelse(prev->bh);
+		get_bh(*bhp);
+		prev->bh = *bhp;
+		prev->blkoff = blkoff;
+		spin_unlock(lock);
+	}
+	return ret;
+}
+
 static int nilfs_palloc_get_desc_block(struct inode *inode,
 				       unsigned long group,
 				       int create, struct buffer_head **bhp)
 {
-	return nilfs_mdt_get_block(inode,
-				   nilfs_palloc_desc_blkoff(inode, group),
-				   create, nilfs_palloc_desc_block_init, bhp);
+	struct nilfs_palloc_cache *cache = NILFS_MDT(inode)->mi_palloc_cache;
+
+	return nilfs_palloc_get_block(inode,
+				      nilfs_palloc_desc_blkoff(inode, group),
+				      create, nilfs_palloc_desc_block_init,
+				      bhp, &cache->prev_desc, &cache->lock);
 }
 
 static int nilfs_palloc_get_bitmap_block(struct inode *inode,
 					 unsigned long group,
 					 int create, struct buffer_head **bhp)
 {
-	return nilfs_mdt_get_block(inode,
-				   nilfs_palloc_bitmap_blkoff(inode, group),
-				   create, NULL, bhp);
+	struct nilfs_palloc_cache *cache = NILFS_MDT(inode)->mi_palloc_cache;
+
+	return nilfs_palloc_get_block(inode,
+				      nilfs_palloc_bitmap_blkoff(inode, group),
+				      create, NULL, bhp,
+				      &cache->prev_bitmap, &cache->lock);
 }
 
 int nilfs_palloc_get_entry_block(struct inode *inode, __u64 nr,
 				 int create, struct buffer_head **bhp)
 {
-	return nilfs_mdt_get_block(inode, nilfs_palloc_entry_blkoff(inode, nr),
-				   create, NULL, bhp);
+	struct nilfs_palloc_cache *cache = NILFS_MDT(inode)->mi_palloc_cache;
+
+	return nilfs_palloc_get_block(inode,
+				      nilfs_palloc_entry_blkoff(inode, nr),
+				      create, NULL, bhp,
+				      &cache->prev_entry, &cache->lock);
 }
 
 static struct nilfs_palloc_group_desc *
@@ -176,13 +222,6 @@ nilfs_palloc_block_get_group_desc(const struct inode *inode,
 		group % nilfs_palloc_groups_per_desc_block(inode);
 }
 
-static unsigned char *
-nilfs_palloc_block_get_bitmap(const struct inode *inode,
-			      const struct buffer_head *bh, void *kaddr)
-{
-	return (unsigned char *)(kaddr + bh_offset(bh));
-}
-
 void *nilfs_palloc_block_get_entry(const struct inode *inode, __u64 nr,
 				   const struct buffer_head *bh, void *kaddr)
 {
@@ -289,8 +328,7 @@ int nilfs_palloc_prepare_alloc_entry(struct inode *inode,
 				if (ret < 0)
 					goto out_desc;
 				bitmap_kaddr = kmap(bitmap_bh->b_page);
-				bitmap = nilfs_palloc_block_get_bitmap(
-					inode, bitmap_bh, bitmap_kaddr);
+				bitmap = bitmap_kaddr + bh_offset(bitmap_bh);
 				pos = nilfs_palloc_find_available_slot(
 					inode, group, group_offset, bitmap,
 					entries_per_group);
@@ -351,8 +389,7 @@ void nilfs_palloc_commit_free_entry(struct inode *inode,
 	desc = nilfs_palloc_block_get_group_desc(inode, group,
 						 req->pr_desc_bh, desc_kaddr);
 	bitmap_kaddr = kmap(req->pr_bitmap_bh->b_page);
-	bitmap = nilfs_palloc_block_get_bitmap(inode, req->pr_bitmap_bh,
-					       bitmap_kaddr);
+	bitmap = bitmap_kaddr + bh_offset(req->pr_bitmap_bh);
 
 	if (!nilfs_clear_bit_atomic(nilfs_mdt_bgl_lock(inode, group),
 				    group_offset, bitmap))
@@ -385,8 +422,7 @@ void nilfs_palloc_abort_alloc_entry(struct inode *inode,
 	desc = nilfs_palloc_block_get_group_desc(inode, group,
 						 req->pr_desc_bh, desc_kaddr);
 	bitmap_kaddr = kmap(req->pr_bitmap_bh->b_page);
-	bitmap = nilfs_palloc_block_get_bitmap(inode, req->pr_bitmap_bh,
-					       bitmap_kaddr);
+	bitmap = bitmap_kaddr + bh_offset(req->pr_bitmap_bh);
 	if (!nilfs_clear_bit_atomic(nilfs_mdt_bgl_lock(inode, group),
 				    group_offset, bitmap))
 		printk(KERN_WARNING "%s: entry numer %llu already freed\n",
@@ -472,8 +508,7 @@ int nilfs_palloc_freev(struct inode *inode, __u64 *entry_nrs, size_t nitems)
 		desc = nilfs_palloc_block_get_group_desc(
 			inode, group, desc_bh, desc_kaddr);
 		bitmap_kaddr = kmap(bitmap_bh->b_page);
-		bitmap = nilfs_palloc_block_get_bitmap(
-			inode, bitmap_bh, bitmap_kaddr);
+		bitmap = bitmap_kaddr + bh_offset(bitmap_bh);
 		for (j = i, n = 0;
 		     (j < nitems) && nilfs_palloc_group_is_in(inode, group,
 							      entry_nrs[j]);
@@ -502,3 +537,30 @@ int nilfs_palloc_freev(struct inode *inode, __u64 *entry_nrs, size_t nitems)
 	}
 	return 0;
 }
+
+void nilfs_palloc_setup_cache(struct inode *inode,
+			      struct nilfs_palloc_cache *cache)
+{
+	NILFS_MDT(inode)->mi_palloc_cache = cache;
+	spin_lock_init(&cache->lock);
+}
+
+void nilfs_palloc_clear_cache(struct inode *inode)
+{
+	struct nilfs_palloc_cache *cache = NILFS_MDT(inode)->mi_palloc_cache;
+
+	spin_lock(&cache->lock);
+	brelse(cache->prev_desc.bh);
+	brelse(cache->prev_bitmap.bh);
+	brelse(cache->prev_entry.bh);
+	cache->prev_desc.bh = NULL;
+	cache->prev_bitmap.bh = NULL;
+	cache->prev_entry.bh = NULL;
+	spin_unlock(&cache->lock);
+}
+
+void nilfs_palloc_destroy_cache(struct inode *inode)
+{
+	nilfs_palloc_clear_cache(inode);
+	NILFS_MDT(inode)->mi_palloc_cache = NULL;
+}
diff --git a/fs/nilfs2/alloc.h b/fs/nilfs2/alloc.h
index 4ace5475c2c7..f4543ac4f560 100644
--- a/fs/nilfs2/alloc.h
+++ b/fs/nilfs2/alloc.h
@@ -69,4 +69,25 @@ int nilfs_palloc_freev(struct inode *, __u64 *, size_t);
 #define nilfs_clear_bit_atomic		ext2_clear_bit_atomic
 #define nilfs_find_next_zero_bit	ext2_find_next_zero_bit
 
+/*
+ * persistent object allocator cache
+ */
+
+struct nilfs_bh_assoc {
+	unsigned long blkoff;
+	struct buffer_head *bh;
+};
+
+struct nilfs_palloc_cache {
+	spinlock_t lock;
+	struct nilfs_bh_assoc prev_desc;
+	struct nilfs_bh_assoc prev_bitmap;
+	struct nilfs_bh_assoc prev_entry;
+};
+
+void nilfs_palloc_setup_cache(struct inode *inode,
+			      struct nilfs_palloc_cache *cache);
+void nilfs_palloc_clear_cache(struct inode *inode);
+void nilfs_palloc_destroy_cache(struct inode *inode);
+
 #endif	/* _NILFS_ALLOC_H */
diff --git a/fs/nilfs2/bmap.c b/fs/nilfs2/bmap.c
index 08834df6ec68..effdbdbe6c11 100644
--- a/fs/nilfs2/bmap.c
+++ b/fs/nilfs2/bmap.c
@@ -402,19 +402,11 @@ int nilfs_bmap_test_and_clear_dirty(struct nilfs_bmap *bmap)
 void nilfs_bmap_add_blocks(const struct nilfs_bmap *bmap, int n)
 {
 	inode_add_bytes(bmap->b_inode, (1 << bmap->b_inode->i_blkbits) * n);
-	if (NILFS_MDT(bmap->b_inode))
-		nilfs_mdt_mark_dirty(bmap->b_inode);
-	else
-		mark_inode_dirty(bmap->b_inode);
 }
 
 void nilfs_bmap_sub_blocks(const struct nilfs_bmap *bmap, int n)
 {
 	inode_sub_bytes(bmap->b_inode, (1 << bmap->b_inode->i_blkbits) * n);
-	if (NILFS_MDT(bmap->b_inode))
-		nilfs_mdt_mark_dirty(bmap->b_inode);
-	else
-		mark_inode_dirty(bmap->b_inode);
 }
 
 __u64 nilfs_bmap_data_get_key(const struct nilfs_bmap *bmap,
@@ -425,8 +417,8 @@ __u64 nilfs_bmap_data_get_key(const struct nilfs_bmap *bmap,
 
 	key = page_index(bh->b_page) << (PAGE_CACHE_SHIFT -
 					 bmap->b_inode->i_blkbits);
-	for (pbh = page_buffers(bh->b_page); pbh != bh;
-	     pbh = pbh->b_this_page, key++);
+	for (pbh = page_buffers(bh->b_page); pbh != bh; pbh = pbh->b_this_page)
+		key++;
 
 	return key;
 }
diff --git a/fs/nilfs2/btnode.c b/fs/nilfs2/btnode.c
index 84c25382f8e3..471e269536ae 100644
--- a/fs/nilfs2/btnode.c
+++ b/fs/nilfs2/btnode.c
@@ -68,9 +68,34 @@ void nilfs_btnode_cache_clear(struct address_space *btnc)
 	truncate_inode_pages(btnc, 0);
 }
 
+struct buffer_head *
+nilfs_btnode_create_block(struct address_space *btnc, __u64 blocknr)
+{
+	struct inode *inode = NILFS_BTNC_I(btnc);
+	struct buffer_head *bh;
+
+	bh = nilfs_grab_buffer(inode, btnc, blocknr, 1 << BH_NILFS_Node);
+	if (unlikely(!bh))
+		return NULL;
+
+	if (unlikely(buffer_mapped(bh) || buffer_uptodate(bh) ||
+		     buffer_dirty(bh))) {
+		brelse(bh);
+		BUG();
+	}
+	memset(bh->b_data, 0, 1 << inode->i_blkbits);
+	bh->b_bdev = NILFS_I_NILFS(inode)->ns_bdev;
+	bh->b_blocknr = blocknr;
+	set_buffer_mapped(bh);
+	set_buffer_uptodate(bh);
+
+	unlock_page(bh->b_page);
+	page_cache_release(bh->b_page);
+	return bh;
+}
+
 int nilfs_btnode_submit_block(struct address_space *btnc, __u64 blocknr,
-			      sector_t pblocknr, struct buffer_head **pbh,
-			      int newblk)
+			      sector_t pblocknr, struct buffer_head **pbh)
 {
 	struct buffer_head *bh;
 	struct inode *inode = NILFS_BTNC_I(btnc);
@@ -81,19 +106,6 @@ int nilfs_btnode_submit_block(struct address_space *btnc, __u64 blocknr,
 		return -ENOMEM;
 
 	err = -EEXIST; /* internal code */
-	if (newblk) {
-		if (unlikely(buffer_mapped(bh) || buffer_uptodate(bh) ||
-			     buffer_dirty(bh))) {
-			brelse(bh);
-			BUG();
-		}
-		memset(bh->b_data, 0, 1 << inode->i_blkbits);
-		bh->b_bdev = NILFS_I_NILFS(inode)->ns_bdev;
-		bh->b_blocknr = blocknr;
-		set_buffer_mapped(bh);
-		set_buffer_uptodate(bh);
-		goto found;
-	}
 
 	if (buffer_uptodate(bh) || buffer_dirty(bh))
 		goto found;
@@ -135,27 +147,6 @@ out_locked:
 	return err;
 }
 
-int nilfs_btnode_get(struct address_space *btnc, __u64 blocknr,
-		     sector_t pblocknr, struct buffer_head **pbh, int newblk)
-{
-	struct buffer_head *bh;
-	int err;
-
-	err = nilfs_btnode_submit_block(btnc, blocknr, pblocknr, pbh, newblk);
-	if (err == -EEXIST) /* internal code (cache hit) */
-		return 0;
-	if (unlikely(err))
-		return err;
-
-	bh = *pbh;
-	wait_on_buffer(bh);
-	if (!buffer_uptodate(bh)) {
-		brelse(bh);
-		return -EIO;
-	}
-	return 0;
-}
-
 /**
  * nilfs_btnode_delete - delete B-tree node buffer
  * @bh: buffer to be deleted
@@ -244,12 +235,13 @@ retry:
 		unlock_page(obh->b_page);
 	}
 
-	err = nilfs_btnode_get(btnc, newkey, 0, &nbh, 1);
-	if (likely(!err)) {
-		BUG_ON(nbh == obh);
-		ctxt->newbh = nbh;
-	}
-	return err;
+	nbh = nilfs_btnode_create_block(btnc, newkey);
+	if (!nbh)
+		return -ENOMEM;
+
+	BUG_ON(nbh == obh);
+	ctxt->newbh = nbh;
+	return 0;
 
  failed_unlock:
 	unlock_page(obh->b_page);
diff --git a/fs/nilfs2/btnode.h b/fs/nilfs2/btnode.h
index 3e2275172ed6..07da83f07712 100644
--- a/fs/nilfs2/btnode.h
+++ b/fs/nilfs2/btnode.h
@@ -40,10 +40,10 @@ struct nilfs_btnode_chkey_ctxt {
 void nilfs_btnode_cache_init_once(struct address_space *);
 void nilfs_btnode_cache_init(struct address_space *, struct backing_dev_info *);
 void nilfs_btnode_cache_clear(struct address_space *);
+struct buffer_head *nilfs_btnode_create_block(struct address_space *btnc,
+					      __u64 blocknr);
 int nilfs_btnode_submit_block(struct address_space *, __u64, sector_t,
-			      struct buffer_head **, int);
-int nilfs_btnode_get(struct address_space *, __u64, sector_t,
-		     struct buffer_head **, int);
+			      struct buffer_head **);
 void nilfs_btnode_delete(struct buffer_head *);
 int nilfs_btnode_prepare_change_key(struct address_space *,
 				    struct nilfs_btnode_chkey_ctxt *);
diff --git a/fs/nilfs2/btree.c b/fs/nilfs2/btree.c
index e25b507a474f..7cdd98b8d514 100644
--- a/fs/nilfs2/btree.c
+++ b/fs/nilfs2/btree.c
@@ -114,7 +114,18 @@ static int nilfs_btree_get_block(const struct nilfs_btree *btree, __u64 ptr,
 {
 	struct address_space *btnc =
 		&NILFS_BMAP_I((struct nilfs_bmap *)btree)->i_btnode_cache;
-	return nilfs_btnode_get(btnc, ptr, 0, bhp, 0);
+	int err;
+
+	err = nilfs_btnode_submit_block(btnc, ptr, 0, bhp);
+	if (err)
+		return err == -EEXIST ? 0 : err;
+
+	wait_on_buffer(*bhp);
+	if (!buffer_uptodate(*bhp)) {
+		brelse(*bhp);
+		return -EIO;
+	}
+	return 0;
 }
 
 static int nilfs_btree_get_new_block(const struct nilfs_btree *btree,
@@ -122,12 +133,15 @@ static int nilfs_btree_get_new_block(const struct nilfs_btree *btree,
 {
 	struct address_space *btnc =
 		&NILFS_BMAP_I((struct nilfs_bmap *)btree)->i_btnode_cache;
-	int ret;
+	struct buffer_head *bh;
 
-	ret = nilfs_btnode_get(btnc, ptr, 0, bhp, 1);
-	if (!ret)
-		set_buffer_nilfs_volatile(*bhp);
-	return ret;
+	bh = nilfs_btnode_create_block(btnc, ptr);
+	if (!bh)
+		return -ENOMEM;
+
+	set_buffer_nilfs_volatile(bh);
+	*bhp = bh;
+	return 0;
 }
 
 static inline int
@@ -444,6 +458,18 @@ nilfs_btree_get_node(const struct nilfs_btree *btree,
 		nilfs_btree_get_nonroot_node(path, level);
 }
 
+static inline int
+nilfs_btree_bad_node(struct nilfs_btree_node *node, int level)
+{
+	if (unlikely(nilfs_btree_node_get_level(node) != level)) {
+		dump_stack();
+		printk(KERN_CRIT "NILFS: btree level mismatch: %d != %d\n",
+		       nilfs_btree_node_get_level(node), level);
+		return 1;
+	}
+	return 0;
+}
+
 static int nilfs_btree_do_lookup(const struct nilfs_btree *btree,
 				 struct nilfs_btree_path *path,
 				 __u64 key, __u64 *ptrp, int minlevel)
@@ -467,7 +493,8 @@ static int nilfs_btree_do_lookup(const struct nilfs_btree *btree,
 		if (ret < 0)
 			return ret;
 		node = nilfs_btree_get_nonroot_node(path, level);
-		BUG_ON(level != nilfs_btree_node_get_level(node));
+		if (nilfs_btree_bad_node(node, level))
+			return -EINVAL;
 		if (!found)
 			found = nilfs_btree_node_lookup(node, key, &index);
 		else
@@ -512,7 +539,8 @@ static int nilfs_btree_do_lookup_last(const struct nilfs_btree *btree,
 		if (ret < 0)
 			return ret;
 		node = nilfs_btree_get_nonroot_node(path, level);
-		BUG_ON(level != nilfs_btree_node_get_level(node));
+		if (nilfs_btree_bad_node(node, level))
+			return -EINVAL;
 		index = nilfs_btree_node_get_nchildren(node) - 1;
 		ptr = nilfs_btree_node_get_ptr(btree, node, index);
 		path[level].bp_index = index;
@@ -638,13 +666,11 @@ static void nilfs_btree_promote_key(struct nilfs_btree *btree,
 {
 	if (level < nilfs_btree_height(btree) - 1) {
 		do {
-			lock_buffer(path[level].bp_bh);
 			nilfs_btree_node_set_key(
 				nilfs_btree_get_nonroot_node(path, level),
 				path[level].bp_index, key);
 			if (!buffer_dirty(path[level].bp_bh))
 				nilfs_btnode_mark_dirty(path[level].bp_bh);
-			unlock_buffer(path[level].bp_bh);
 		} while ((path[level].bp_index == 0) &&
 			 (++level < nilfs_btree_height(btree) - 1));
 	}
@@ -663,13 +689,11 @@ static void nilfs_btree_do_insert(struct nilfs_btree *btree,
 	struct nilfs_btree_node *node;
 
 	if (level < nilfs_btree_height(btree) - 1) {
-		lock_buffer(path[level].bp_bh);
 		node = nilfs_btree_get_nonroot_node(path, level);
 		nilfs_btree_node_insert(btree, node, *keyp, *ptrp,
 					path[level].bp_index);
 		if (!buffer_dirty(path[level].bp_bh))
 			nilfs_btnode_mark_dirty(path[level].bp_bh);
-		unlock_buffer(path[level].bp_bh);
 
 		if (path[level].bp_index == 0)
 			nilfs_btree_promote_key(btree, path, level + 1,
@@ -689,9 +713,6 @@ static void nilfs_btree_carry_left(struct nilfs_btree *btree,
 	struct nilfs_btree_node *node, *left;
 	int nchildren, lnchildren, n, move;
 
-	lock_buffer(path[level].bp_bh);
-	lock_buffer(path[level].bp_sib_bh);
-
 	node = nilfs_btree_get_nonroot_node(path, level);
 	left = nilfs_btree_get_sib_node(path, level);
 	nchildren = nilfs_btree_node_get_nchildren(node);
@@ -712,9 +733,6 @@ static void nilfs_btree_carry_left(struct nilfs_btree *btree,
 	if (!buffer_dirty(path[level].bp_sib_bh))
 		nilfs_btnode_mark_dirty(path[level].bp_sib_bh);
 
-	unlock_buffer(path[level].bp_bh);
-	unlock_buffer(path[level].bp_sib_bh);
-
 	nilfs_btree_promote_key(btree, path, level + 1,
 				nilfs_btree_node_get_key(node, 0));
 
@@ -740,9 +758,6 @@ static void nilfs_btree_carry_right(struct nilfs_btree *btree,
 	struct nilfs_btree_node *node, *right;
 	int nchildren, rnchildren, n, move;
 
-	lock_buffer(path[level].bp_bh);
-	lock_buffer(path[level].bp_sib_bh);
-
 	node = nilfs_btree_get_nonroot_node(path, level);
 	right = nilfs_btree_get_sib_node(path, level);
 	nchildren = nilfs_btree_node_get_nchildren(node);
@@ -763,9 +778,6 @@ static void nilfs_btree_carry_right(struct nilfs_btree *btree,
 	if (!buffer_dirty(path[level].bp_sib_bh))
 		nilfs_btnode_mark_dirty(path[level].bp_sib_bh);
 
-	unlock_buffer(path[level].bp_bh);
-	unlock_buffer(path[level].bp_sib_bh);
-
 	path[level + 1].bp_index++;
 	nilfs_btree_promote_key(btree, path, level + 1,
 				nilfs_btree_node_get_key(right, 0));
@@ -794,9 +806,6 @@ static void nilfs_btree_split(struct nilfs_btree *btree,
 	__u64 newptr;
 	int nchildren, n, move;
 
-	lock_buffer(path[level].bp_bh);
-	lock_buffer(path[level].bp_sib_bh);
-
 	node = nilfs_btree_get_nonroot_node(path, level);
 	right = nilfs_btree_get_sib_node(path, level);
 	nchildren = nilfs_btree_node_get_nchildren(node);
@@ -815,9 +824,6 @@ static void nilfs_btree_split(struct nilfs_btree *btree,
 	if (!buffer_dirty(path[level].bp_sib_bh))
 		nilfs_btnode_mark_dirty(path[level].bp_sib_bh);
 
-	unlock_buffer(path[level].bp_bh);
-	unlock_buffer(path[level].bp_sib_bh);
-
 	newkey = nilfs_btree_node_get_key(right, 0);
 	newptr = path[level].bp_newreq.bpr_ptr;
 
@@ -852,8 +858,6 @@ static void nilfs_btree_grow(struct nilfs_btree *btree,
 	struct nilfs_btree_node *root, *child;
 	int n;
 
-	lock_buffer(path[level].bp_sib_bh);
-
 	root = nilfs_btree_get_root(btree);
 	child = nilfs_btree_get_sib_node(path, level);
 
@@ -865,8 +869,6 @@ static void nilfs_btree_grow(struct nilfs_btree *btree,
 	if (!buffer_dirty(path[level].bp_sib_bh))
 		nilfs_btnode_mark_dirty(path[level].bp_sib_bh);
 
-	unlock_buffer(path[level].bp_sib_bh);
-
 	path[level].bp_bh = path[level].bp_sib_bh;
 	path[level].bp_sib_bh = NULL;
 
@@ -1023,11 +1025,9 @@ static int nilfs_btree_prepare_insert(struct nilfs_btree *btree,
 
 		stats->bs_nblocks++;
 
-		lock_buffer(bh);
 		nilfs_btree_node_init(btree,
 				      (struct nilfs_btree_node *)bh->b_data,
 				      0, level, 0, NULL, NULL);
-		unlock_buffer(bh);
 		path[level].bp_sib_bh = bh;
 		path[level].bp_op = nilfs_btree_split;
 	}
@@ -1052,10 +1052,8 @@ static int nilfs_btree_prepare_insert(struct nilfs_btree *btree,
 	if (ret < 0)
 		goto err_out_curr_node;
 
-	lock_buffer(bh);
 	nilfs_btree_node_init(btree, (struct nilfs_btree_node *)bh->b_data,
 			      0, level, 0, NULL, NULL);
-	unlock_buffer(bh);
 	path[level].bp_sib_bh = bh;
 	path[level].bp_op = nilfs_btree_grow;
 
@@ -1154,13 +1152,11 @@ static void nilfs_btree_do_delete(struct nilfs_btree *btree,
 	struct nilfs_btree_node *node;
 
 	if (level < nilfs_btree_height(btree) - 1) {
-		lock_buffer(path[level].bp_bh);
 		node = nilfs_btree_get_nonroot_node(path, level);
 		nilfs_btree_node_delete(btree, node, keyp, ptrp,
 					path[level].bp_index);
 		if (!buffer_dirty(path[level].bp_bh))
 			nilfs_btnode_mark_dirty(path[level].bp_bh);
-		unlock_buffer(path[level].bp_bh);
 		if (path[level].bp_index == 0)
 			nilfs_btree_promote_key(btree, path, level + 1,
 				nilfs_btree_node_get_key(node, 0));
@@ -1180,9 +1176,6 @@ static void nilfs_btree_borrow_left(struct nilfs_btree *btree,
 
 	nilfs_btree_do_delete(btree, path, level, keyp, ptrp);
 
-	lock_buffer(path[level].bp_bh);
-	lock_buffer(path[level].bp_sib_bh);
-
 	node = nilfs_btree_get_nonroot_node(path, level);
 	left = nilfs_btree_get_sib_node(path, level);
 	nchildren = nilfs_btree_node_get_nchildren(node);
@@ -1197,9 +1190,6 @@ static void nilfs_btree_borrow_left(struct nilfs_btree *btree,
 	if (!buffer_dirty(path[level].bp_sib_bh))
 		nilfs_btnode_mark_dirty(path[level].bp_sib_bh);
 
-	unlock_buffer(path[level].bp_bh);
-	unlock_buffer(path[level].bp_sib_bh);
-
 	nilfs_btree_promote_key(btree, path, level + 1,
 				nilfs_btree_node_get_key(node, 0));
 
@@ -1217,9 +1207,6 @@ static void nilfs_btree_borrow_right(struct nilfs_btree *btree,
 
 	nilfs_btree_do_delete(btree, path, level, keyp, ptrp);
 
-	lock_buffer(path[level].bp_bh);
-	lock_buffer(path[level].bp_sib_bh);
-
 	node = nilfs_btree_get_nonroot_node(path, level);
 	right = nilfs_btree_get_sib_node(path, level);
 	nchildren = nilfs_btree_node_get_nchildren(node);
@@ -1234,9 +1221,6 @@ static void nilfs_btree_borrow_right(struct nilfs_btree *btree,
 	if (!buffer_dirty(path[level].bp_sib_bh))
 		nilfs_btnode_mark_dirty(path[level].bp_sib_bh);
 
-	unlock_buffer(path[level].bp_bh);
-	unlock_buffer(path[level].bp_sib_bh);
-
 	path[level + 1].bp_index++;
 	nilfs_btree_promote_key(btree, path, level + 1,
 				nilfs_btree_node_get_key(right, 0));
@@ -1255,9 +1239,6 @@ static void nilfs_btree_concat_left(struct nilfs_btree *btree,
 
 	nilfs_btree_do_delete(btree, path, level, keyp, ptrp);
 
-	lock_buffer(path[level].bp_bh);
-	lock_buffer(path[level].bp_sib_bh);
-
 	node = nilfs_btree_get_nonroot_node(path, level);
 	left = nilfs_btree_get_sib_node(path, level);
 
@@ -1268,9 +1249,6 @@ static void nilfs_btree_concat_left(struct nilfs_btree *btree,
 	if (!buffer_dirty(path[level].bp_sib_bh))
 		nilfs_btnode_mark_dirty(path[level].bp_sib_bh);
 
-	unlock_buffer(path[level].bp_bh);
-	unlock_buffer(path[level].bp_sib_bh);
-
 	nilfs_btnode_delete(path[level].bp_bh);
 	path[level].bp_bh = path[level].bp_sib_bh;
 	path[level].bp_sib_bh = NULL;
@@ -1286,9 +1264,6 @@ static void nilfs_btree_concat_right(struct nilfs_btree *btree,
 
 	nilfs_btree_do_delete(btree, path, level, keyp, ptrp);
 
-	lock_buffer(path[level].bp_bh);
-	lock_buffer(path[level].bp_sib_bh);
-
 	node = nilfs_btree_get_nonroot_node(path, level);
 	right = nilfs_btree_get_sib_node(path, level);
 
@@ -1299,9 +1274,6 @@ static void nilfs_btree_concat_right(struct nilfs_btree *btree,
 	if (!buffer_dirty(path[level].bp_bh))
 		nilfs_btnode_mark_dirty(path[level].bp_bh);
 
-	unlock_buffer(path[level].bp_bh);
-	unlock_buffer(path[level].bp_sib_bh);
-
 	nilfs_btnode_delete(path[level].bp_sib_bh);
 	path[level].bp_sib_bh = NULL;
 	path[level + 1].bp_index++;
@@ -1316,7 +1288,6 @@ static void nilfs_btree_shrink(struct nilfs_btree *btree,
 
 	nilfs_btree_do_delete(btree, path, level, keyp, ptrp);
 
-	lock_buffer(path[level].bp_bh);
 	root = nilfs_btree_get_root(btree);
 	child = nilfs_btree_get_nonroot_node(path, level);
 
@@ -1324,7 +1295,6 @@ static void nilfs_btree_shrink(struct nilfs_btree *btree,
 	nilfs_btree_node_set_level(root, level);
 	n = nilfs_btree_node_get_nchildren(child);
 	nilfs_btree_node_move_left(btree, root, child, n);
-	unlock_buffer(path[level].bp_bh);
 
 	nilfs_btnode_delete(path[level].bp_bh);
 	path[level].bp_bh = NULL;
@@ -1699,7 +1669,6 @@ nilfs_btree_commit_convert_and_insert(struct nilfs_bmap *bmap,
 		nilfs_bmap_commit_alloc_ptr(bmap, nreq, dat);
 
 		/* create child node at level 1 */
-		lock_buffer(bh);
 		node = (struct nilfs_btree_node *)bh->b_data;
 		nilfs_btree_node_init(btree, node, 0, 1, n, keys, ptrs);
 		nilfs_btree_node_insert(btree, node,
@@ -1709,7 +1678,6 @@ nilfs_btree_commit_convert_and_insert(struct nilfs_bmap *bmap,
 		if (!nilfs_bmap_dirty(bmap))
 			nilfs_bmap_set_dirty(bmap);
 
-		unlock_buffer(bh);
 		brelse(bh);
 
 		/* create root node at level 2 */
@@ -2050,7 +2018,7 @@ static void nilfs_btree_lookup_dirty_buffers(struct nilfs_bmap *bmap,
 	for (level = NILFS_BTREE_LEVEL_NODE_MIN;
 	     level < NILFS_BTREE_LEVEL_MAX;
 	     level++)
-		list_splice(&lists[level], listp->prev);
+		list_splice_tail(&lists[level], listp);
 }
 
 static int nilfs_btree_assign_p(struct nilfs_btree *btree,
diff --git a/fs/nilfs2/btree.h b/fs/nilfs2/btree.h
index 0e72bbbc6b64..4b82d84ade75 100644
--- a/fs/nilfs2/btree.h
+++ b/fs/nilfs2/btree.h
@@ -34,28 +34,6 @@ struct nilfs_btree;
 struct nilfs_btree_path;
 
 /**
- * struct nilfs_btree_node - B-tree node
- * @bn_flags: flags
- * @bn_level: level
- * @bn_nchildren: number of children
- * @bn_pad: padding
- */
-struct nilfs_btree_node {
-	__u8 bn_flags;
-	__u8 bn_level;
-	__le16 bn_nchildren;
-	__le32 bn_pad;
-};
-
-/* flags */
-#define NILFS_BTREE_NODE_ROOT	0x01
-
-/* level */
-#define NILFS_BTREE_LEVEL_DATA		0
-#define NILFS_BTREE_LEVEL_NODE_MIN	(NILFS_BTREE_LEVEL_DATA + 1)
-#define NILFS_BTREE_LEVEL_MAX		14
-
-/**
  * struct nilfs_btree - B-tree structure
  * @bt_bmap: bmap base structure
  */
diff --git a/fs/nilfs2/cpfile.c b/fs/nilfs2/cpfile.c
index 3f5d5d06f53c..18737818db63 100644
--- a/fs/nilfs2/cpfile.c
+++ b/fs/nilfs2/cpfile.c
@@ -328,19 +328,24 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile,
 			tnicps += nicps;
 			nilfs_mdt_mark_buffer_dirty(cp_bh);
 			nilfs_mdt_mark_dirty(cpfile);
-			if (!nilfs_cpfile_is_in_first(cpfile, cno) &&
-			    (count = nilfs_cpfile_block_sub_valid_checkpoints(
-				    cpfile, cp_bh, kaddr, nicps)) == 0) {
-				/* make hole */
-				kunmap_atomic(kaddr, KM_USER0);
-				brelse(cp_bh);
-				ret = nilfs_cpfile_delete_checkpoint_block(
-					cpfile, cno);
-				if (ret == 0)
-					continue;
-				printk(KERN_ERR "%s: cannot delete block\n",
-				       __func__);
-				break;
+			if (!nilfs_cpfile_is_in_first(cpfile, cno)) {
+				count =
+				  nilfs_cpfile_block_sub_valid_checkpoints(
+						cpfile, cp_bh, kaddr, nicps);
+				if (count == 0) {
+					/* make hole */
+					kunmap_atomic(kaddr, KM_USER0);
+					brelse(cp_bh);
+					ret =
+					  nilfs_cpfile_delete_checkpoint_block(
+								   cpfile, cno);
+					if (ret == 0)
+						continue;
+					printk(KERN_ERR
+					       "%s: cannot delete block\n",
+					       __func__);
+					break;
+				}
 			}
 		}
 
@@ -926,3 +931,29 @@ int nilfs_cpfile_get_stat(struct inode *cpfile, struct nilfs_cpstat *cpstat)
 	up_read(&NILFS_MDT(cpfile)->mi_sem);
 	return ret;
 }
+
+/**
+ * nilfs_cpfile_read - read cpfile inode
+ * @cpfile: cpfile inode
+ * @raw_inode: on-disk cpfile inode
+ */
+int nilfs_cpfile_read(struct inode *cpfile, struct nilfs_inode *raw_inode)
+{
+	return nilfs_read_inode_common(cpfile, raw_inode);
+}
+
+/**
+ * nilfs_cpfile_new - create cpfile
+ * @nilfs: nilfs object
+ * @cpsize: size of a checkpoint entry
+ */
+struct inode *nilfs_cpfile_new(struct the_nilfs *nilfs, size_t cpsize)
+{
+	struct inode *cpfile;
+
+	cpfile = nilfs_mdt_new(nilfs, NULL, NILFS_CPFILE_INO, 0);
+	if (cpfile)
+		nilfs_mdt_set_entry_size(cpfile, cpsize,
+					 sizeof(struct nilfs_cpfile_header));
+	return cpfile;
+}
diff --git a/fs/nilfs2/cpfile.h b/fs/nilfs2/cpfile.h
index debea896e701..bc0809e0ab43 100644
--- a/fs/nilfs2/cpfile.h
+++ b/fs/nilfs2/cpfile.h
@@ -40,4 +40,7 @@ int nilfs_cpfile_get_stat(struct inode *, struct nilfs_cpstat *);
 ssize_t nilfs_cpfile_get_cpinfo(struct inode *, __u64 *, int, void *, unsigned,
 				size_t);
 
+int nilfs_cpfile_read(struct inode *cpfile, struct nilfs_inode *raw_inode);
+struct inode *nilfs_cpfile_new(struct the_nilfs *nilfs, size_t cpsize);
+
 #endif	/* _NILFS_CPFILE_H */
diff --git a/fs/nilfs2/dat.c b/fs/nilfs2/dat.c
index 1ff8e15bd36b..187dd07ba86c 100644
--- a/fs/nilfs2/dat.c
+++ b/fs/nilfs2/dat.c
@@ -33,6 +33,16 @@
 #define NILFS_CNO_MIN	((__u64)1)
 #define NILFS_CNO_MAX	(~(__u64)0)
 
+struct nilfs_dat_info {
+	struct nilfs_mdt_info mi;
+	struct nilfs_palloc_cache palloc_cache;
+};
+
+static inline struct nilfs_dat_info *NILFS_DAT_I(struct inode *dat)
+{
+	return (struct nilfs_dat_info *)NILFS_MDT(dat);
+}
+
 static int nilfs_dat_prepare_entry(struct inode *dat,
 				   struct nilfs_palloc_req *req, int create)
 {
@@ -425,3 +435,40 @@ ssize_t nilfs_dat_get_vinfo(struct inode *dat, void *buf, unsigned visz,
 
 	return nvi;
 }
+
+/**
+ * nilfs_dat_read - read dat inode
+ * @dat: dat inode
+ * @raw_inode: on-disk dat inode
+ */
+int nilfs_dat_read(struct inode *dat, struct nilfs_inode *raw_inode)
+{
+	return nilfs_read_inode_common(dat, raw_inode);
+}
+
+/**
+ * nilfs_dat_new - create dat file
+ * @nilfs: nilfs object
+ * @entry_size: size of a dat entry
+ */
+struct inode *nilfs_dat_new(struct the_nilfs *nilfs, size_t entry_size)
+{
+	static struct lock_class_key dat_lock_key;
+	struct inode *dat;
+	struct nilfs_dat_info *di;
+	int err;
+
+	dat = nilfs_mdt_new(nilfs, NULL, NILFS_DAT_INO, sizeof(*di));
+	if (dat) {
+		err = nilfs_palloc_init_blockgroup(dat, entry_size);
+		if (unlikely(err)) {
+			nilfs_mdt_destroy(dat);
+			return NULL;
+		}
+
+		di = NILFS_DAT_I(dat);
+		lockdep_set_class(&di->mi.mi_sem, &dat_lock_key);
+		nilfs_palloc_setup_cache(dat, &di->palloc_cache);
+	}
+	return dat;
+}
diff --git a/fs/nilfs2/dat.h b/fs/nilfs2/dat.h
index 406070d3ff49..d31c3aab0efe 100644
--- a/fs/nilfs2/dat.h
+++ b/fs/nilfs2/dat.h
@@ -53,4 +53,7 @@ int nilfs_dat_freev(struct inode *, __u64 *, size_t);
 int nilfs_dat_move(struct inode *, __u64, sector_t);
 ssize_t nilfs_dat_get_vinfo(struct inode *, void *, unsigned, size_t);
 
+int nilfs_dat_read(struct inode *dat, struct nilfs_inode *raw_inode);
+struct inode *nilfs_dat_new(struct the_nilfs *nilfs, size_t entry_size);
+
 #endif	/* _NILFS_DAT_H */
diff --git a/fs/nilfs2/dir.c b/fs/nilfs2/dir.c
index e097099bfc8f..76d803e060a9 100644
--- a/fs/nilfs2/dir.c
+++ b/fs/nilfs2/dir.c
@@ -99,9 +99,9 @@ static int nilfs_prepare_chunk(struct page *page,
 				 NULL, nilfs_get_block);
 }
 
-static int nilfs_commit_chunk(struct page *page,
-			      struct address_space *mapping,
-			      unsigned from, unsigned to)
+static void nilfs_commit_chunk(struct page *page,
+			       struct address_space *mapping,
+			       unsigned from, unsigned to)
 {
 	struct inode *dir = mapping->host;
 	struct nilfs_sb_info *sbi = NILFS_SB(dir->i_sb);
@@ -112,15 +112,13 @@ static int nilfs_commit_chunk(struct page *page,
 
 	nr_dirty = nilfs_page_count_clean_buffers(page, from, to);
 	copied = block_write_end(NULL, mapping, pos, len, len, page, NULL);
-	if (pos + copied > dir->i_size) {
+	if (pos + copied > dir->i_size)
 		i_size_write(dir, pos + copied);
-		mark_inode_dirty(dir);
-	}
 	if (IS_DIRSYNC(dir))
 		nilfs_set_transaction_flag(NILFS_TI_SYNC);
 	err = nilfs_set_file_dirty(sbi, dir, nr_dirty);
+	WARN_ON(err); /* do not happen */
 	unlock_page(page);
-	return err;
 }
 
 static void nilfs_check_page(struct page *page)
@@ -455,11 +453,10 @@ void nilfs_set_link(struct inode *dir, struct nilfs_dir_entry *de,
 	BUG_ON(err);
 	de->inode = cpu_to_le64(inode->i_ino);
 	nilfs_set_de_type(de, inode);
-	err = nilfs_commit_chunk(page, mapping, from, to);
+	nilfs_commit_chunk(page, mapping, from, to);
 	nilfs_put_page(page);
 	dir->i_mtime = dir->i_ctime = CURRENT_TIME;
 /*	NILFS_I(dir)->i_flags &= ~NILFS_BTREE_FL; */
-	mark_inode_dirty(dir);
 }
 
 /*
@@ -548,10 +545,10 @@ got_it:
 	memcpy(de->name, name, namelen);
 	de->inode = cpu_to_le64(inode->i_ino);
 	nilfs_set_de_type(de, inode);
-	err = nilfs_commit_chunk(page, page->mapping, from, to);
+	nilfs_commit_chunk(page, page->mapping, from, to);
 	dir->i_mtime = dir->i_ctime = CURRENT_TIME;
 /*	NILFS_I(dir)->i_flags &= ~NILFS_BTREE_FL; */
-	mark_inode_dirty(dir);
+	nilfs_mark_inode_dirty(dir);
 	/* OFFSET_CACHE */
 out_put:
 	nilfs_put_page(page);
@@ -595,10 +592,9 @@ int nilfs_delete_entry(struct nilfs_dir_entry *dir, struct page *page)
 	if (pde)
 		pde->rec_len = cpu_to_le16(to - from);
 	dir->inode = 0;
-	err = nilfs_commit_chunk(page, mapping, from, to);
+	nilfs_commit_chunk(page, mapping, from, to);
 	inode->i_ctime = inode->i_mtime = CURRENT_TIME;
 /*	NILFS_I(inode)->i_flags &= ~NILFS_BTREE_FL; */
-	mark_inode_dirty(inode);
 out:
 	nilfs_put_page(page);
 	return err;
@@ -640,7 +636,7 @@ int nilfs_make_empty(struct inode *inode, struct inode *parent)
 	memcpy(de->name, "..\0", 4);
 	nilfs_set_de_type(de, inode);
 	kunmap_atomic(kaddr, KM_USER0);
-	err = nilfs_commit_chunk(page, mapping, 0, chunk_size);
+	nilfs_commit_chunk(page, mapping, 0, chunk_size);
 fail:
 	page_cache_release(page);
 	return err;
diff --git a/fs/nilfs2/direct.c b/fs/nilfs2/direct.c
index d369ac718277..236753df5cdf 100644
--- a/fs/nilfs2/direct.c
+++ b/fs/nilfs2/direct.c
@@ -51,11 +51,11 @@ static int nilfs_direct_lookup(const struct nilfs_bmap *bmap,
 	struct nilfs_direct *direct;
 	__u64 ptr;
 
-	direct = (struct nilfs_direct *)bmap;
-	if ((key > NILFS_DIRECT_KEY_MAX) ||
-	    (level != 1) ||	/* XXX: use macro for level 1 */
-	    ((ptr = nilfs_direct_get_ptr(direct, key)) ==
-	     NILFS_BMAP_INVALID_PTR))
+	direct = (struct nilfs_direct *)bmap;  /* XXX: use macro for level 1 */
+	if (key > NILFS_DIRECT_KEY_MAX || level != 1)
+		return -ENOENT;
+	ptr = nilfs_direct_get_ptr(direct, key);
+	if (ptr == NILFS_BMAP_INVALID_PTR)
 		return -ENOENT;
 
 	if (ptrp != NULL)
@@ -73,9 +73,10 @@ static int nilfs_direct_lookup_contig(const struct nilfs_bmap *bmap,
 	sector_t blocknr;
 	int ret, cnt;
 
-	if (key > NILFS_DIRECT_KEY_MAX ||
-	    (ptr = nilfs_direct_get_ptr(direct, key)) ==
-	    NILFS_BMAP_INVALID_PTR)
+	if (key > NILFS_DIRECT_KEY_MAX)
+		return -ENOENT;
+	ptr = nilfs_direct_get_ptr(direct, key);
+	if (ptr == NILFS_BMAP_INVALID_PTR)
 		return -ENOENT;
 
 	if (NILFS_BMAP_USE_VBN(bmap)) {
diff --git a/fs/nilfs2/gcdat.c b/fs/nilfs2/gcdat.c
index 93383c5cee90..dd5f7e0a95f6 100644
--- a/fs/nilfs2/gcdat.c
+++ b/fs/nilfs2/gcdat.c
@@ -61,6 +61,8 @@ void nilfs_commit_gcdat_inode(struct the_nilfs *nilfs)
 
 	nilfs_bmap_commit_gcdat(gii->i_bmap, dii->i_bmap);
 
+	nilfs_palloc_clear_cache(dat);
+	nilfs_palloc_clear_cache(gcdat);
 	nilfs_clear_dirty_pages(mapping);
 	nilfs_copy_back_pages(mapping, gmapping);
 	/* note: mdt dirty flags should be cleared by segctor. */
@@ -79,6 +81,7 @@ void nilfs_clear_gcdat_inode(struct the_nilfs *nilfs)
 	gcdat->i_state = I_CLEAR;
 	gii->i_flags = 0;
 
+	nilfs_palloc_clear_cache(gcdat);
 	truncate_inode_pages(gcdat->i_mapping, 0);
 	truncate_inode_pages(&gii->i_btnode_cache, 0);
 }
diff --git a/fs/nilfs2/gcinode.c b/fs/nilfs2/gcinode.c
index e6de0a27ab5d..e16a6664dfa2 100644
--- a/fs/nilfs2/gcinode.c
+++ b/fs/nilfs2/gcinode.c
@@ -149,7 +149,7 @@ int nilfs_gccache_submit_read_node(struct inode *inode, sector_t pbn,
 				   __u64 vbn, struct buffer_head **out_bh)
 {
 	int ret = nilfs_btnode_submit_block(&NILFS_I(inode)->i_btnode_cache,
-					    vbn ? : pbn, pbn, out_bh, 0);
+					    vbn ? : pbn, pbn, out_bh);
 	if (ret == -EEXIST) /* internal code (cache hit) */
 		ret = 0;
 	return ret;
@@ -212,9 +212,10 @@ void nilfs_destroy_gccache(struct the_nilfs *nilfs)
 static struct inode *alloc_gcinode(struct the_nilfs *nilfs, ino_t ino,
 				   __u64 cno)
 {
-	struct inode *inode = nilfs_mdt_new_common(nilfs, NULL, ino, GFP_NOFS);
+	struct inode *inode;
 	struct nilfs_inode_info *ii;
 
+	inode = nilfs_mdt_new_common(nilfs, NULL, ino, GFP_NOFS, 0);
 	if (!inode)
 		return NULL;
 
@@ -265,7 +266,6 @@ struct inode *nilfs_gc_iget(struct the_nilfs *nilfs, ino_t ino, __u64 cno)
  */
 void nilfs_clear_gcinode(struct inode *inode)
 {
-	nilfs_mdt_clear(inode);
 	nilfs_mdt_destroy(inode);
 }
 
diff --git a/fs/nilfs2/ifile.c b/fs/nilfs2/ifile.c
index de86401f209f..922d9dd42c8f 100644
--- a/fs/nilfs2/ifile.c
+++ b/fs/nilfs2/ifile.c
@@ -29,6 +29,17 @@
 #include "alloc.h"
 #include "ifile.h"
 
+
+struct nilfs_ifile_info {
+	struct nilfs_mdt_info mi;
+	struct nilfs_palloc_cache palloc_cache;
+};
+
+static inline struct nilfs_ifile_info *NILFS_IFILE_I(struct inode *ifile)
+{
+	return (struct nilfs_ifile_info *)NILFS_MDT(ifile);
+}
+
 /**
  * nilfs_ifile_create_inode - create a new disk inode
  * @ifile: ifile inode
@@ -148,3 +159,27 @@ int nilfs_ifile_get_inode_block(struct inode *ifile, ino_t ino,
 	}
 	return err;
 }
+
+/**
+ * nilfs_ifile_new - create inode file
+ * @sbi: nilfs_sb_info struct
+ * @inode_size: size of an inode
+ */
+struct inode *nilfs_ifile_new(struct nilfs_sb_info *sbi, size_t inode_size)
+{
+	struct inode *ifile;
+	int err;
+
+	ifile = nilfs_mdt_new(sbi->s_nilfs, sbi->s_super, NILFS_IFILE_INO,
+			      sizeof(struct nilfs_ifile_info));
+	if (ifile) {
+		err = nilfs_palloc_init_blockgroup(ifile, inode_size);
+		if (unlikely(err)) {
+			nilfs_mdt_destroy(ifile);
+			return NULL;
+		}
+		nilfs_palloc_setup_cache(ifile,
+					 &NILFS_IFILE_I(ifile)->palloc_cache);
+	}
+	return ifile;
+}
diff --git a/fs/nilfs2/ifile.h b/fs/nilfs2/ifile.h
index ecc3ba76db47..cbca32e498f2 100644
--- a/fs/nilfs2/ifile.h
+++ b/fs/nilfs2/ifile.h
@@ -49,4 +49,6 @@ int nilfs_ifile_create_inode(struct inode *, ino_t *, struct buffer_head **);
 int nilfs_ifile_delete_inode(struct inode *, ino_t);
 int nilfs_ifile_get_inode_block(struct inode *, ino_t, struct buffer_head **);
 
+struct inode *nilfs_ifile_new(struct nilfs_sb_info *sbi, size_t inode_size);
+
 #endif	/* _NILFS_IFILE_H */
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 2a0a5a3ac134..7868cc122ac7 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -97,6 +97,7 @@ int nilfs_get_block(struct inode *inode, sector_t blkoff,
 			nilfs_transaction_abort(inode->i_sb);
 			goto out;
 		}
+		nilfs_mark_inode_dirty(inode);
 		nilfs_transaction_commit(inode->i_sb); /* never fails */
 		/* Error handling should be detailed */
 		set_buffer_new(bh_result);
@@ -322,7 +323,6 @@ struct inode *nilfs_new_inode(struct inode *dir, int mode)
 				    nilfs_init_acl(), proper cancellation of
 				    above jobs should be considered */
 
-	mark_inode_dirty(inode);
 	return inode;
 
  failed_acl:
@@ -525,7 +525,6 @@ void nilfs_update_inode(struct inode *inode, struct buffer_head *ibh)
 
 	raw_inode = nilfs_ifile_map_inode(sbi->s_ifile, ino, ibh);
 
-	/* The buffer is guarded with lock_buffer() by the caller */
 	if (test_and_clear_bit(NILFS_I_NEW, &ii->i_state))
 		memset(raw_inode, 0, NILFS_MDT(sbi->s_ifile)->mi_entry_size);
 	set_bit(NILFS_I_INODE_DIRTY, &ii->i_state);
@@ -599,6 +598,7 @@ void nilfs_truncate(struct inode *inode)
 	if (IS_SYNC(inode))
 		nilfs_set_transaction_flag(NILFS_TI_SYNC);
 
+	nilfs_mark_inode_dirty(inode);
 	nilfs_set_file_dirty(NILFS_SB(sb), inode, 0);
 	nilfs_transaction_commit(sb);
 	/* May construct a logical segment and may fail in sync mode.
@@ -623,6 +623,7 @@ void nilfs_delete_inode(struct inode *inode)
 		truncate_inode_pages(&inode->i_data, 0);
 
 	nilfs_truncate_bmap(ii, 0);
+	nilfs_mark_inode_dirty(inode);
 	nilfs_free_inode(inode);
 	/* nilfs_free_inode() marks inode buffer dirty */
 	if (IS_SYNC(inode))
@@ -745,9 +746,7 @@ int nilfs_mark_inode_dirty(struct inode *inode)
 			      "failed to reget inode block.\n");
 		return err;
 	}
-	lock_buffer(ibh);
 	nilfs_update_inode(inode, ibh);
-	unlock_buffer(ibh);
 	nilfs_mdt_mark_buffer_dirty(ibh);
 	nilfs_mdt_mark_dirty(sbi->s_ifile);
 	brelse(ibh);
diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c
index f6af76042d80..d6b2b83de363 100644
--- a/fs/nilfs2/ioctl.c
+++ b/fs/nilfs2/ioctl.c
@@ -480,7 +480,7 @@ static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp,
 				      unsigned int cmd, void __user *argp)
 {
 	struct nilfs_argv argv[5];
-	const static size_t argsz[5] = {
+	static const size_t argsz[5] = {
 		sizeof(struct nilfs_vdesc),
 		sizeof(struct nilfs_period),
 		sizeof(__u64),
diff --git a/fs/nilfs2/mdt.c b/fs/nilfs2/mdt.c
index f6326112d647..06713ffcc7f2 100644
--- a/fs/nilfs2/mdt.c
+++ b/fs/nilfs2/mdt.c
@@ -186,7 +186,7 @@ nilfs_mdt_submit_block(struct inode *inode, unsigned long blkoff,
 }
 
 static int nilfs_mdt_read_block(struct inode *inode, unsigned long block,
-				struct buffer_head **out_bh)
+				int readahead, struct buffer_head **out_bh)
 {
 	struct buffer_head *first_bh, *bh;
 	unsigned long blkoff;
@@ -200,16 +200,18 @@ static int nilfs_mdt_read_block(struct inode *inode, unsigned long block,
 	if (unlikely(err))
 		goto failed;
 
-	blkoff = block + 1;
-	for (i = 0; i < nr_ra_blocks; i++, blkoff++) {
-		err = nilfs_mdt_submit_block(inode, blkoff, READA, &bh);
-		if (likely(!err || err == -EEXIST))
-			brelse(bh);
-		else if (err != -EBUSY)
-			break; /* abort readahead if bmap lookup failed */
-
-		if (!buffer_locked(first_bh))
-			goto out_no_wait;
+	if (readahead) {
+		blkoff = block + 1;
+		for (i = 0; i < nr_ra_blocks; i++, blkoff++) {
+			err = nilfs_mdt_submit_block(inode, blkoff, READA, &bh);
+			if (likely(!err || err == -EEXIST))
+				brelse(bh);
+			else if (err != -EBUSY)
+				break;
+				/* abort readahead if bmap lookup failed */
+			if (!buffer_locked(first_bh))
+				goto out_no_wait;
+		}
 	}
 
 	wait_on_buffer(first_bh);
@@ -263,7 +265,7 @@ int nilfs_mdt_get_block(struct inode *inode, unsigned long blkoff, int create,
 
 	/* Should be rewritten with merging nilfs_mdt_read_block() */
  retry:
-	ret = nilfs_mdt_read_block(inode, blkoff, out_bh);
+	ret = nilfs_mdt_read_block(inode, blkoff, !create, out_bh);
 	if (!create || ret != -ENOENT)
 		return ret;
 
@@ -371,7 +373,7 @@ int nilfs_mdt_mark_block_dirty(struct inode *inode, unsigned long block)
 	struct buffer_head *bh;
 	int err;
 
-	err = nilfs_mdt_read_block(inode, block, &bh);
+	err = nilfs_mdt_read_block(inode, block, 0, &bh);
 	if (unlikely(err))
 		return err;
 	nilfs_mark_buffer_dirty(bh);
@@ -445,9 +447,17 @@ static const struct file_operations def_mdt_fops;
  * longer than those of the super block structs; they may continue for
  * several consecutive mounts/umounts.  This would need discussions.
  */
+/**
+ * nilfs_mdt_new_common - allocate a pseudo inode for metadata file
+ * @nilfs: nilfs object
+ * @sb: super block instance the metadata file belongs to
+ * @ino: inode number
+ * @gfp_mask: gfp mask for data pages
+ * @objsz: size of the private object attached to inode->i_private
+ */
 struct inode *
 nilfs_mdt_new_common(struct the_nilfs *nilfs, struct super_block *sb,
-		     ino_t ino, gfp_t gfp_mask)
+		     ino_t ino, gfp_t gfp_mask, size_t objsz)
 {
 	struct inode *inode = nilfs_alloc_inode_common(nilfs);
 
@@ -455,8 +465,9 @@ nilfs_mdt_new_common(struct the_nilfs *nilfs, struct super_block *sb,
 		return NULL;
 	else {
 		struct address_space * const mapping = &inode->i_data;
-		struct nilfs_mdt_info *mi = kzalloc(sizeof(*mi), GFP_NOFS);
+		struct nilfs_mdt_info *mi;
 
+		mi = kzalloc(max(sizeof(*mi), objsz), GFP_NOFS);
 		if (!mi) {
 			nilfs_destroy_inode(inode);
 			return NULL;
@@ -513,11 +524,11 @@ nilfs_mdt_new_common(struct the_nilfs *nilfs, struct super_block *sb,
 }
 
 struct inode *nilfs_mdt_new(struct the_nilfs *nilfs, struct super_block *sb,
-			    ino_t ino)
+			    ino_t ino, size_t objsz)
 {
-	struct inode *inode = nilfs_mdt_new_common(nilfs, sb, ino,
-						   NILFS_MDT_GFP);
+	struct inode *inode;
 
+	inode = nilfs_mdt_new_common(nilfs, sb, ino, NILFS_MDT_GFP, objsz);
 	if (!inode)
 		return NULL;
 
@@ -544,14 +555,15 @@ void nilfs_mdt_set_shadow(struct inode *orig, struct inode *shadow)
 		&NILFS_I(orig)->i_btnode_cache;
 }
 
-void nilfs_mdt_clear(struct inode *inode)
+static void nilfs_mdt_clear(struct inode *inode)
 {
 	struct nilfs_inode_info *ii = NILFS_I(inode);
 
 	invalidate_mapping_pages(inode->i_mapping, 0, -1);
 	truncate_inode_pages(inode->i_mapping, 0);
 
-	nilfs_bmap_clear(ii->i_bmap);
+	if (test_bit(NILFS_I_BMAP, &ii->i_state))
+		nilfs_bmap_clear(ii->i_bmap);
 	nilfs_btnode_cache_clear(&ii->i_btnode_cache);
 }
 
@@ -559,6 +571,10 @@ void nilfs_mdt_destroy(struct inode *inode)
 {
 	struct nilfs_mdt_info *mdi = NILFS_MDT(inode);
 
+	if (mdi->mi_palloc_cache)
+		nilfs_palloc_destroy_cache(inode);
+	nilfs_mdt_clear(inode);
+
 	kfree(mdi->mi_bgl); /* kfree(NULL) is safe */
 	kfree(mdi);
 	nilfs_destroy_inode(inode);
diff --git a/fs/nilfs2/mdt.h b/fs/nilfs2/mdt.h
index 431599733c9b..6c4bbb0470fc 100644
--- a/fs/nilfs2/mdt.h
+++ b/fs/nilfs2/mdt.h
@@ -36,6 +36,7 @@
  * @mi_entry_size: size of an entry
  * @mi_first_entry_offset: offset to the first entry
  * @mi_entries_per_block: number of entries in a block
+ * @mi_palloc_cache: persistent object allocator cache
  * @mi_blocks_per_group: number of blocks in a group
  * @mi_blocks_per_desc_block: number of blocks per descriptor block
  */
@@ -46,6 +47,7 @@ struct nilfs_mdt_info {
 	unsigned		mi_entry_size;
 	unsigned		mi_first_entry_offset;
 	unsigned long		mi_entries_per_block;
+	struct nilfs_palloc_cache *mi_palloc_cache;
 	unsigned long		mi_blocks_per_group;
 	unsigned long		mi_blocks_per_desc_block;
 };
@@ -74,11 +76,11 @@ int nilfs_mdt_forget_block(struct inode *, unsigned long);
 int nilfs_mdt_mark_block_dirty(struct inode *, unsigned long);
 int nilfs_mdt_fetch_dirty(struct inode *);
 
-struct inode *nilfs_mdt_new(struct the_nilfs *, struct super_block *, ino_t);
+struct inode *nilfs_mdt_new(struct the_nilfs *, struct super_block *, ino_t,
+			    size_t);
 struct inode *nilfs_mdt_new_common(struct the_nilfs *, struct super_block *,
-				   ino_t, gfp_t);
+				   ino_t, gfp_t, size_t);
 void nilfs_mdt_destroy(struct inode *);
-void nilfs_mdt_clear(struct inode *);
 void nilfs_mdt_set_entry_size(struct inode *, unsigned, unsigned);
 void nilfs_mdt_set_shadow(struct inode *, struct inode *);
 
@@ -104,21 +106,4 @@ static inline __u64 nilfs_mdt_cno(struct inode *inode)
 #define nilfs_mdt_bgl_lock(inode, bg) \
 	(&NILFS_MDT(inode)->mi_bgl->locks[(bg) & (NR_BG_LOCKS-1)].lock)
 
-
-static inline int
-nilfs_mdt_read_inode_direct(struct inode *inode, struct buffer_head *bh,
-			    unsigned n)
-{
-	return nilfs_read_inode_common(
-		inode, (struct nilfs_inode *)(bh->b_data + n));
-}
-
-static inline void
-nilfs_mdt_write_inode_direct(struct inode *inode, struct buffer_head *bh,
-			     unsigned n)
-{
-	nilfs_write_inode_common(
-		inode, (struct nilfs_inode *)(bh->b_data + n), 1);
-}
-
 #endif /* _NILFS_MDT_H */
diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c
index ed02e886fa79..07ba838ef089 100644
--- a/fs/nilfs2/namei.c
+++ b/fs/nilfs2/namei.c
@@ -120,7 +120,7 @@ static int nilfs_create(struct inode *dir, struct dentry *dentry, int mode,
 		inode->i_op = &nilfs_file_inode_operations;
 		inode->i_fop = &nilfs_file_operations;
 		inode->i_mapping->a_ops = &nilfs_aops;
-		mark_inode_dirty(inode);
+		nilfs_mark_inode_dirty(inode);
 		err = nilfs_add_nondir(dentry, inode);
 	}
 	if (!err)
@@ -148,7 +148,7 @@ nilfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
 	err = PTR_ERR(inode);
 	if (!IS_ERR(inode)) {
 		init_special_inode(inode, inode->i_mode, rdev);
-		mark_inode_dirty(inode);
+		nilfs_mark_inode_dirty(inode);
 		err = nilfs_add_nondir(dentry, inode);
 	}
 	if (!err)
@@ -188,7 +188,7 @@ static int nilfs_symlink(struct inode *dir, struct dentry *dentry,
 		goto out_fail;
 
 	/* mark_inode_dirty(inode); */
-	/* nilfs_new_inode() and page_symlink() do this */
+	/* page_symlink() do this */
 
 	err = nilfs_add_nondir(dentry, inode);
 out:
@@ -200,7 +200,8 @@ out:
 	return err;
 
 out_fail:
-	inode_dec_link_count(inode);
+	drop_nlink(inode);
+	nilfs_mark_inode_dirty(inode);
 	iput(inode);
 	goto out;
 }
@@ -245,7 +246,7 @@ static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (err)
 		return err;
 
-	inode_inc_link_count(dir);
+	inc_nlink(dir);
 
 	inode = nilfs_new_inode(dir, S_IFDIR | mode);
 	err = PTR_ERR(inode);
@@ -256,7 +257,7 @@ static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	inode->i_fop = &nilfs_dir_operations;
 	inode->i_mapping->a_ops = &nilfs_aops;
 
-	inode_inc_link_count(inode);
+	inc_nlink(inode);
 
 	err = nilfs_make_empty(inode, dir);
 	if (err)
@@ -266,6 +267,7 @@ static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (err)
 		goto out_fail;
 
+	nilfs_mark_inode_dirty(inode);
 	d_instantiate(dentry, inode);
 out:
 	if (!err)
@@ -276,26 +278,23 @@ out:
 	return err;
 
 out_fail:
-	inode_dec_link_count(inode);
-	inode_dec_link_count(inode);
+	drop_nlink(inode);
+	drop_nlink(inode);
+	nilfs_mark_inode_dirty(inode);
 	iput(inode);
 out_dir:
-	inode_dec_link_count(dir);
+	drop_nlink(dir);
+	nilfs_mark_inode_dirty(dir);
 	goto out;
 }
 
-static int nilfs_unlink(struct inode *dir, struct dentry *dentry)
+static int nilfs_do_unlink(struct inode *dir, struct dentry *dentry)
 {
 	struct inode *inode;
 	struct nilfs_dir_entry *de;
 	struct page *page;
-	struct nilfs_transaction_info ti;
 	int err;
 
-	err = nilfs_transaction_begin(dir->i_sb, &ti, 0);
-	if (err)
-		return err;
-
 	err = -ENOENT;
 	de = nilfs_find_entry(dir, dentry, &page);
 	if (!de)
@@ -317,12 +316,28 @@ static int nilfs_unlink(struct inode *dir, struct dentry *dentry)
 		goto out;
 
 	inode->i_ctime = dir->i_ctime;
-	inode_dec_link_count(inode);
+	drop_nlink(inode);
 	err = 0;
 out:
-	if (!err)
+	return err;
+}
+
+static int nilfs_unlink(struct inode *dir, struct dentry *dentry)
+{
+	struct nilfs_transaction_info ti;
+	int err;
+
+	err = nilfs_transaction_begin(dir->i_sb, &ti, 0);
+	if (err)
+		return err;
+
+	err = nilfs_do_unlink(dir, dentry);
+
+	if (!err) {
+		nilfs_mark_inode_dirty(dir);
+		nilfs_mark_inode_dirty(dentry->d_inode);
 		err = nilfs_transaction_commit(dir->i_sb);
-	else
+	} else
 		nilfs_transaction_abort(dir->i_sb);
 
 	return err;
@@ -340,11 +355,13 @@ static int nilfs_rmdir(struct inode *dir, struct dentry *dentry)
 
 	err = -ENOTEMPTY;
 	if (nilfs_empty_dir(inode)) {
-		err = nilfs_unlink(dir, dentry);
+		err = nilfs_do_unlink(dir, dentry);
 		if (!err) {
 			inode->i_size = 0;
-			inode_dec_link_count(inode);
-			inode_dec_link_count(dir);
+			drop_nlink(inode);
+			nilfs_mark_inode_dirty(inode);
+			drop_nlink(dir);
+			nilfs_mark_inode_dirty(dir);
 		}
 	}
 	if (!err)
@@ -395,42 +412,48 @@ static int nilfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		new_de = nilfs_find_entry(new_dir, new_dentry, &new_page);
 		if (!new_de)
 			goto out_dir;
-		inode_inc_link_count(old_inode);
+		inc_nlink(old_inode);
 		nilfs_set_link(new_dir, new_de, new_page, old_inode);
+		nilfs_mark_inode_dirty(new_dir);
 		new_inode->i_ctime = CURRENT_TIME;
 		if (dir_de)
 			drop_nlink(new_inode);
-		inode_dec_link_count(new_inode);
+		drop_nlink(new_inode);
+		nilfs_mark_inode_dirty(new_inode);
 	} else {
 		if (dir_de) {
 			err = -EMLINK;
 			if (new_dir->i_nlink >= NILFS_LINK_MAX)
 				goto out_dir;
 		}
-		inode_inc_link_count(old_inode);
+		inc_nlink(old_inode);
 		err = nilfs_add_link(new_dentry, old_inode);
 		if (err) {
-			inode_dec_link_count(old_inode);
+			drop_nlink(old_inode);
+			nilfs_mark_inode_dirty(old_inode);
 			goto out_dir;
 		}
-		if (dir_de)
-			inode_inc_link_count(new_dir);
+		if (dir_de) {
+			inc_nlink(new_dir);
+			nilfs_mark_inode_dirty(new_dir);
+		}
 	}
 
 	/*
 	 * Like most other Unix systems, set the ctime for inodes on a
 	 * rename.
-	 * inode_dec_link_count() will mark the inode dirty.
 	 */
 	old_inode->i_ctime = CURRENT_TIME;
 
 	nilfs_delete_entry(old_de, old_page);
-	inode_dec_link_count(old_inode);
+	drop_nlink(old_inode);
 
 	if (dir_de) {
 		nilfs_set_link(old_inode, dir_de, dir_page, new_dir);
-		inode_dec_link_count(old_dir);
+		drop_nlink(old_dir);
 	}
+	nilfs_mark_inode_dirty(old_dir);
+	nilfs_mark_inode_dirty(old_inode);
 
 	err = nilfs_transaction_commit(old_dir->i_sb);
 	return err;
diff --git a/fs/nilfs2/recovery.c b/fs/nilfs2/recovery.c
index 6dc83591d118..c9c96c7825dc 100644
--- a/fs/nilfs2/recovery.c
+++ b/fs/nilfs2/recovery.c
@@ -770,14 +770,8 @@ int nilfs_recover_logical_segments(struct the_nilfs *nilfs,
 		nilfs_finish_roll_forward(nilfs, sbi, ri);
 	}
 
-	nilfs_detach_checkpoint(sbi);
-	return 0;
-
  failed:
 	nilfs_detach_checkpoint(sbi);
-	nilfs_mdt_clear(nilfs->ns_cpfile);
-	nilfs_mdt_clear(nilfs->ns_sufile);
-	nilfs_mdt_clear(nilfs->ns_dat);
 	return err;
 }
 
@@ -804,6 +798,7 @@ int nilfs_search_super_root(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi,
 	struct nilfs_segsum_info ssi;
 	sector_t pseg_start, pseg_end, sr_pseg_start = 0;
 	sector_t seg_start, seg_end; /* range of full segment (block number) */
+	sector_t b, end;
 	u64 seg_seq;
 	__u64 segnum, nextnum = 0;
 	__u64 cno;
@@ -819,6 +814,11 @@ int nilfs_search_super_root(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi,
 	/* Calculate range of segment */
 	nilfs_get_segment_range(nilfs, segnum, &seg_start, &seg_end);
 
+	/* Read ahead segment */
+	b = seg_start;
+	while (b <= seg_end)
+		sb_breadahead(sbi->s_super, b++);
+
 	for (;;) {
 		/* Load segment summary */
 		ret = load_segment_summary(sbi, pseg_start, seg_seq, &ssi, 1);
@@ -841,14 +841,20 @@ int nilfs_search_super_root(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi,
 		ri->ri_nextnum = nextnum;
 		empty_seg = 0;
 
+		if (!NILFS_SEG_HAS_SR(&ssi) && !scan_newer) {
+			/* This will never happen because a superblock
+			   (last_segment) always points to a pseg
+			   having a super root. */
+			ret = NILFS_SEG_FAIL_CONSISTENCY;
+			goto failed;
+		}
+
+		if (pseg_start == seg_start) {
+			nilfs_get_segment_range(nilfs, nextnum, &b, &end);
+			while (b <= end)
+				sb_breadahead(sbi->s_super, b++);
+		}
 		if (!NILFS_SEG_HAS_SR(&ssi)) {
-			if (!scan_newer) {
-				/* This will never happen because a superblock
-				   (last_segment) always points to a pseg
-				   having a super root. */
-				ret = NILFS_SEG_FAIL_CONSISTENCY;
-				goto failed;
-			}
 			if (!ri->ri_lsegs_start && NILFS_SEG_LOGBGN(&ssi)) {
 				ri->ri_lsegs_start = pseg_start;
 				ri->ri_lsegs_start_seq = seg_seq;
@@ -919,7 +925,7 @@ int nilfs_search_super_root(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi,
 
  super_root_found:
 	/* Updating pointers relating to the latest checkpoint */
-	list_splice(&segments, ri->ri_used_segments.prev);
+	list_splice_tail(&segments, &ri->ri_used_segments);
 	nilfs->ns_last_pseg = sr_pseg_start;
 	nilfs->ns_last_seq = nilfs->ns_seg_seq;
 	nilfs->ns_last_cno = ri->ri_cno;
diff --git a/fs/nilfs2/segbuf.c b/fs/nilfs2/segbuf.c
index e6d9e37fa241..645c78656aa0 100644
--- a/fs/nilfs2/segbuf.c
+++ b/fs/nilfs2/segbuf.c
@@ -24,10 +24,22 @@
 #include <linux/buffer_head.h>
 #include <linux/writeback.h>
 #include <linux/crc32.h>
+#include <linux/backing-dev.h>
 #include "page.h"
 #include "segbuf.h"
 
 
+struct nilfs_write_info {
+	struct the_nilfs       *nilfs;
+	struct bio	       *bio;
+	int 			start, end; /* The region to be submitted */
+	int			rest_blocks;
+	int			max_pages;
+	int			nr_vecs;
+	sector_t		blocknr;
+};
+
+
 static struct kmem_cache *nilfs_segbuf_cachep;
 
 static void nilfs_segbuf_init_once(void *obj)
@@ -63,6 +75,11 @@ struct nilfs_segment_buffer *nilfs_segbuf_new(struct super_block *sb)
 	INIT_LIST_HEAD(&segbuf->sb_list);
 	INIT_LIST_HEAD(&segbuf->sb_segsum_buffers);
 	INIT_LIST_HEAD(&segbuf->sb_payload_buffers);
+
+	init_completion(&segbuf->sb_bio_event);
+	atomic_set(&segbuf->sb_err, 0);
+	segbuf->sb_nbio = 0;
+
 	return segbuf;
 }
 
@@ -83,6 +100,22 @@ void nilfs_segbuf_map(struct nilfs_segment_buffer *segbuf, __u64 segnum,
 		segbuf->sb_fseg_end - segbuf->sb_pseg_start + 1;
 }
 
+/**
+ * nilfs_segbuf_map_cont - map a new log behind a given log
+ * @segbuf: new segment buffer
+ * @prev: segment buffer containing a log to be continued
+ */
+void nilfs_segbuf_map_cont(struct nilfs_segment_buffer *segbuf,
+			   struct nilfs_segment_buffer *prev)
+{
+	segbuf->sb_segnum = prev->sb_segnum;
+	segbuf->sb_fseg_start = prev->sb_fseg_start;
+	segbuf->sb_fseg_end = prev->sb_fseg_end;
+	segbuf->sb_pseg_start = prev->sb_pseg_start + prev->sb_sum.nblocks;
+	segbuf->sb_rest_blocks =
+		segbuf->sb_fseg_end - segbuf->sb_pseg_start + 1;
+}
+
 void nilfs_segbuf_set_next_segnum(struct nilfs_segment_buffer *segbuf,
 				  __u64 nextnum, struct the_nilfs *nilfs)
 {
@@ -132,8 +165,6 @@ int nilfs_segbuf_reset(struct nilfs_segment_buffer *segbuf, unsigned flags,
 	segbuf->sb_sum.sumbytes = sizeof(struct nilfs_segment_summary);
 	segbuf->sb_sum.nfinfo = segbuf->sb_sum.nfileblk = 0;
 	segbuf->sb_sum.ctime = ctime;
-
-	segbuf->sb_io_error = 0;
 	return 0;
 }
 
@@ -219,7 +250,7 @@ void nilfs_segbuf_fill_in_data_crc(struct nilfs_segment_buffer *segbuf,
 	raw_sum->ss_datasum = cpu_to_le32(crc);
 }
 
-void nilfs_release_buffers(struct list_head *list)
+static void nilfs_release_buffers(struct list_head *list)
 {
 	struct buffer_head *bh, *n;
 
@@ -241,13 +272,56 @@ void nilfs_release_buffers(struct list_head *list)
 	}
 }
 
+static void nilfs_segbuf_clear(struct nilfs_segment_buffer *segbuf)
+{
+	nilfs_release_buffers(&segbuf->sb_segsum_buffers);
+	nilfs_release_buffers(&segbuf->sb_payload_buffers);
+}
+
+/*
+ * Iterators for segment buffers
+ */
+void nilfs_clear_logs(struct list_head *logs)
+{
+	struct nilfs_segment_buffer *segbuf;
+
+	list_for_each_entry(segbuf, logs, sb_list)
+		nilfs_segbuf_clear(segbuf);
+}
+
+void nilfs_truncate_logs(struct list_head *logs,
+			 struct nilfs_segment_buffer *last)
+{
+	struct nilfs_segment_buffer *n, *segbuf;
+
+	segbuf = list_prepare_entry(last, logs, sb_list);
+	list_for_each_entry_safe_continue(segbuf, n, logs, sb_list) {
+		list_del_init(&segbuf->sb_list);
+		nilfs_segbuf_clear(segbuf);
+		nilfs_segbuf_free(segbuf);
+	}
+}
+
+int nilfs_wait_on_logs(struct list_head *logs)
+{
+	struct nilfs_segment_buffer *segbuf;
+	int err;
+
+	list_for_each_entry(segbuf, logs, sb_list) {
+		err = nilfs_segbuf_wait(segbuf);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
 /*
  * BIO operations
  */
 static void nilfs_end_bio_write(struct bio *bio, int err)
 {
 	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
-	struct nilfs_write_info *wi = bio->bi_private;
+	struct nilfs_segment_buffer *segbuf = bio->bi_private;
 
 	if (err == -EOPNOTSUPP) {
 		set_bit(BIO_EOPNOTSUPP, &bio->bi_flags);
@@ -256,21 +330,22 @@ static void nilfs_end_bio_write(struct bio *bio, int err)
 	}
 
 	if (!uptodate)
-		atomic_inc(&wi->err);
+		atomic_inc(&segbuf->sb_err);
 
 	bio_put(bio);
-	complete(&wi->bio_event);
+	complete(&segbuf->sb_bio_event);
 }
 
-static int nilfs_submit_seg_bio(struct nilfs_write_info *wi, int mode)
+static int nilfs_segbuf_submit_bio(struct nilfs_segment_buffer *segbuf,
+				   struct nilfs_write_info *wi, int mode)
 {
 	struct bio *bio = wi->bio;
 	int err;
 
-	if (wi->nbio > 0 && bdi_write_congested(wi->bdi)) {
-		wait_for_completion(&wi->bio_event);
-		wi->nbio--;
-		if (unlikely(atomic_read(&wi->err))) {
+	if (segbuf->sb_nbio > 0 && bdi_write_congested(wi->nilfs->ns_bdi)) {
+		wait_for_completion(&segbuf->sb_bio_event);
+		segbuf->sb_nbio--;
+		if (unlikely(atomic_read(&segbuf->sb_err))) {
 			bio_put(bio);
 			err = -EIO;
 			goto failed;
@@ -278,7 +353,7 @@ static int nilfs_submit_seg_bio(struct nilfs_write_info *wi, int mode)
 	}
 
 	bio->bi_end_io = nilfs_end_bio_write;
-	bio->bi_private = wi;
+	bio->bi_private = segbuf;
 	bio_get(bio);
 	submit_bio(mode, bio);
 	if (bio_flagged(bio, BIO_EOPNOTSUPP)) {
@@ -286,7 +361,7 @@ static int nilfs_submit_seg_bio(struct nilfs_write_info *wi, int mode)
 		err = -EOPNOTSUPP;
 		goto failed;
 	}
-	wi->nbio++;
+	segbuf->sb_nbio++;
 	bio_put(bio);
 
 	wi->bio = NULL;
@@ -301,17 +376,15 @@ static int nilfs_submit_seg_bio(struct nilfs_write_info *wi, int mode)
 }
 
 /**
- * nilfs_alloc_seg_bio - allocate a bio for writing segment.
- * @sb: super block
- * @start: beginning disk block number of this BIO.
+ * nilfs_alloc_seg_bio - allocate a new bio for writing log
+ * @nilfs: nilfs object
+ * @start: start block number of the bio
  * @nr_vecs: request size of page vector.
  *
- * alloc_seg_bio() allocates a new BIO structure and initialize it.
- *
  * Return Value: On success, pointer to the struct bio is returned.
  * On error, NULL is returned.
  */
-static struct bio *nilfs_alloc_seg_bio(struct super_block *sb, sector_t start,
+static struct bio *nilfs_alloc_seg_bio(struct the_nilfs *nilfs, sector_t start,
 				       int nr_vecs)
 {
 	struct bio *bio;
@@ -322,36 +395,33 @@ static struct bio *nilfs_alloc_seg_bio(struct super_block *sb, sector_t start,
 			bio = bio_alloc(GFP_NOIO, nr_vecs);
 	}
 	if (likely(bio)) {
-		bio->bi_bdev = sb->s_bdev;
-		bio->bi_sector = (sector_t)start << (sb->s_blocksize_bits - 9);
+		bio->bi_bdev = nilfs->ns_bdev;
+		bio->bi_sector = start << (nilfs->ns_blocksize_bits - 9);
 	}
 	return bio;
 }
 
-void nilfs_segbuf_prepare_write(struct nilfs_segment_buffer *segbuf,
-				struct nilfs_write_info *wi)
+static void nilfs_segbuf_prepare_write(struct nilfs_segment_buffer *segbuf,
+				       struct nilfs_write_info *wi)
 {
 	wi->bio = NULL;
 	wi->rest_blocks = segbuf->sb_sum.nblocks;
-	wi->max_pages = bio_get_nr_vecs(wi->sb->s_bdev);
+	wi->max_pages = bio_get_nr_vecs(wi->nilfs->ns_bdev);
 	wi->nr_vecs = min(wi->max_pages, wi->rest_blocks);
 	wi->start = wi->end = 0;
-	wi->nbio = 0;
 	wi->blocknr = segbuf->sb_pseg_start;
-
-	atomic_set(&wi->err, 0);
-	init_completion(&wi->bio_event);
 }
 
-static int nilfs_submit_bh(struct nilfs_write_info *wi, struct buffer_head *bh,
-			   int mode)
+static int nilfs_segbuf_submit_bh(struct nilfs_segment_buffer *segbuf,
+				  struct nilfs_write_info *wi,
+				  struct buffer_head *bh, int mode)
 {
 	int len, err;
 
 	BUG_ON(wi->nr_vecs <= 0);
  repeat:
 	if (!wi->bio) {
-		wi->bio = nilfs_alloc_seg_bio(wi->sb, wi->blocknr + wi->end,
+		wi->bio = nilfs_alloc_seg_bio(wi->nilfs, wi->blocknr + wi->end,
 					      wi->nr_vecs);
 		if (unlikely(!wi->bio))
 			return -ENOMEM;
@@ -363,76 +433,83 @@ static int nilfs_submit_bh(struct nilfs_write_info *wi, struct buffer_head *bh,
 		return 0;
 	}
 	/* bio is FULL */
-	err = nilfs_submit_seg_bio(wi, mode);
+	err = nilfs_segbuf_submit_bio(segbuf, wi, mode);
 	/* never submit current bh */
 	if (likely(!err))
 		goto repeat;
 	return err;
 }
 
+/**
+ * nilfs_segbuf_write - submit write requests of a log
+ * @segbuf: buffer storing a log to be written
+ * @nilfs: nilfs object
+ *
+ * Return Value: On Success, 0 is returned. On Error, one of the following
+ * negative error code is returned.
+ *
+ * %-EIO - I/O error
+ *
+ * %-ENOMEM - Insufficient memory available.
+ */
 int nilfs_segbuf_write(struct nilfs_segment_buffer *segbuf,
-		       struct nilfs_write_info *wi)
+		       struct the_nilfs *nilfs)
 {
+	struct nilfs_write_info wi;
 	struct buffer_head *bh;
-	int res, rw = WRITE;
+	int res = 0, rw = WRITE;
+
+	wi.nilfs = nilfs;
+	nilfs_segbuf_prepare_write(segbuf, &wi);
 
 	list_for_each_entry(bh, &segbuf->sb_segsum_buffers, b_assoc_buffers) {
-		res = nilfs_submit_bh(wi, bh, rw);
+		res = nilfs_segbuf_submit_bh(segbuf, &wi, bh, rw);
 		if (unlikely(res))
 			goto failed_bio;
 	}
 
 	list_for_each_entry(bh, &segbuf->sb_payload_buffers, b_assoc_buffers) {
-		res = nilfs_submit_bh(wi, bh, rw);
+		res = nilfs_segbuf_submit_bh(segbuf, &wi, bh, rw);
 		if (unlikely(res))
 			goto failed_bio;
 	}
 
-	if (wi->bio) {
+	if (wi.bio) {
 		/*
 		 * Last BIO is always sent through the following
 		 * submission.
 		 */
 		rw |= (1 << BIO_RW_SYNCIO) | (1 << BIO_RW_UNPLUG);
-		res = nilfs_submit_seg_bio(wi, rw);
-		if (unlikely(res))
-			goto failed_bio;
+		res = nilfs_segbuf_submit_bio(segbuf, &wi, rw);
 	}
 
-	res = 0;
- out:
-	return res;
-
  failed_bio:
-	atomic_inc(&wi->err);
-	goto out;
+	return res;
 }
 
 /**
  * nilfs_segbuf_wait - wait for completion of requested BIOs
- * @wi: nilfs_write_info
+ * @segbuf: segment buffer
  *
  * Return Value: On Success, 0 is returned. On Error, one of the following
  * negative error code is returned.
  *
  * %-EIO - I/O error
  */
-int nilfs_segbuf_wait(struct nilfs_segment_buffer *segbuf,
-		      struct nilfs_write_info *wi)
+int nilfs_segbuf_wait(struct nilfs_segment_buffer *segbuf)
 {
 	int err = 0;
 
-	if (!wi->nbio)
+	if (!segbuf->sb_nbio)
 		return 0;
 
 	do {
-		wait_for_completion(&wi->bio_event);
-	} while (--wi->nbio > 0);
+		wait_for_completion(&segbuf->sb_bio_event);
+	} while (--segbuf->sb_nbio > 0);
 
-	if (unlikely(atomic_read(&wi->err) > 0)) {
+	if (unlikely(atomic_read(&segbuf->sb_err) > 0)) {
 		printk(KERN_ERR "NILFS: IO error writing segment\n");
 		err = -EIO;
-		segbuf->sb_io_error = 1;
 	}
 	return err;
 }
diff --git a/fs/nilfs2/segbuf.h b/fs/nilfs2/segbuf.h
index 0c3076f4e592..6af1630fb401 100644
--- a/fs/nilfs2/segbuf.h
+++ b/fs/nilfs2/segbuf.h
@@ -27,7 +27,6 @@
 #include <linux/buffer_head.h>
 #include <linux/bio.h>
 #include <linux/completion.h>
-#include <linux/backing-dev.h>
 
 /**
  * struct nilfs_segsum_info - On-memory segment summary
@@ -77,7 +76,9 @@ struct nilfs_segsum_info {
  * @sb_rest_blocks: Number of residual blocks in the current segment
  * @sb_segsum_buffers: List of buffers for segment summaries
  * @sb_payload_buffers: List of buffers for segment payload
- * @sb_io_error: I/O error status
+ * @sb_nbio: Number of flying bio requests
+ * @sb_err: I/O error status
+ * @sb_bio_event: Completion event of log writing
  */
 struct nilfs_segment_buffer {
 	struct super_block     *sb_super;
@@ -96,7 +97,9 @@ struct nilfs_segment_buffer {
 	struct list_head	sb_payload_buffers; /* including super root */
 
 	/* io status */
-	int			sb_io_error;
+	int			sb_nbio;
+	atomic_t		sb_err;
+	struct completion	sb_bio_event;
 };
 
 #define NILFS_LIST_SEGBUF(head)  \
@@ -125,6 +128,8 @@ struct nilfs_segment_buffer *nilfs_segbuf_new(struct super_block *);
 void nilfs_segbuf_free(struct nilfs_segment_buffer *);
 void nilfs_segbuf_map(struct nilfs_segment_buffer *, __u64, unsigned long,
 		      struct the_nilfs *);
+void nilfs_segbuf_map_cont(struct nilfs_segment_buffer *segbuf,
+			   struct nilfs_segment_buffer *prev);
 void nilfs_segbuf_set_next_segnum(struct nilfs_segment_buffer *, __u64,
 				  struct the_nilfs *);
 int nilfs_segbuf_reset(struct nilfs_segment_buffer *, unsigned, time_t);
@@ -161,41 +166,18 @@ nilfs_segbuf_add_file_buffer(struct nilfs_segment_buffer *segbuf,
 	segbuf->sb_sum.nfileblk++;
 }
 
-void nilfs_release_buffers(struct list_head *);
+int nilfs_segbuf_write(struct nilfs_segment_buffer *segbuf,
+		       struct the_nilfs *nilfs);
+int nilfs_segbuf_wait(struct nilfs_segment_buffer *segbuf);
 
-static inline void nilfs_segbuf_clear(struct nilfs_segment_buffer *segbuf)
+void nilfs_clear_logs(struct list_head *logs);
+void nilfs_truncate_logs(struct list_head *logs,
+			 struct nilfs_segment_buffer *last);
+int nilfs_wait_on_logs(struct list_head *logs);
+
+static inline void nilfs_destroy_logs(struct list_head *logs)
 {
-	nilfs_release_buffers(&segbuf->sb_segsum_buffers);
-	nilfs_release_buffers(&segbuf->sb_payload_buffers);
+	nilfs_truncate_logs(logs, NULL);
 }
 
-struct nilfs_write_info {
-	struct bio	       *bio;
-	int 			start, end; /* The region to be submitted */
-	int			rest_blocks;
-	int			max_pages;
-	int			nr_vecs;
-	sector_t		blocknr;
-
-	int			nbio;
-	atomic_t		err;
-	struct completion	bio_event;
-				/* completion event of segment write */
-
-	/*
-	 * The following fields must be set explicitly
-	 */
-	struct super_block     *sb;
-	struct backing_dev_info *bdi; /* backing dev info */
-	struct buffer_head     *bh_sr;
-};
-
-
-void nilfs_segbuf_prepare_write(struct nilfs_segment_buffer *,
-				struct nilfs_write_info *);
-int nilfs_segbuf_write(struct nilfs_segment_buffer *,
-		       struct nilfs_write_info *);
-int nilfs_segbuf_wait(struct nilfs_segment_buffer *,
-		      struct nilfs_write_info *);
-
 #endif /* _NILFS_SEGBUF_H */
diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c
index 6eff66a070d5..17584c524486 100644
--- a/fs/nilfs2/segment.c
+++ b/fs/nilfs2/segment.c
@@ -974,12 +974,12 @@ static void nilfs_segctor_fill_in_super_root(struct nilfs_sc_info *sci,
 			      nilfs->ns_nongc_ctime : sci->sc_seg_ctime);
 	raw_sr->sr_flags = 0;
 
-	nilfs_mdt_write_inode_direct(
-		nilfs_dat_inode(nilfs), bh_sr, NILFS_SR_DAT_OFFSET(isz));
-	nilfs_mdt_write_inode_direct(
-		nilfs->ns_cpfile, bh_sr, NILFS_SR_CPFILE_OFFSET(isz));
-	nilfs_mdt_write_inode_direct(
-		nilfs->ns_sufile, bh_sr, NILFS_SR_SUFILE_OFFSET(isz));
+	nilfs_write_inode_common(nilfs_dat_inode(nilfs), (void *)raw_sr +
+				 NILFS_SR_DAT_OFFSET(isz), 1);
+	nilfs_write_inode_common(nilfs->ns_cpfile, (void *)raw_sr +
+				 NILFS_SR_CPFILE_OFFSET(isz), 1);
+	nilfs_write_inode_common(nilfs->ns_sufile, (void *)raw_sr +
+				 NILFS_SR_SUFILE_OFFSET(isz), 1);
 }
 
 static void nilfs_redirty_inodes(struct list_head *head)
@@ -1273,73 +1273,75 @@ static int nilfs_segctor_collect_blocks(struct nilfs_sc_info *sci, int mode)
 	return err;
 }
 
-static int nilfs_touch_segusage(struct inode *sufile, __u64 segnum)
-{
-	struct buffer_head *bh_su;
-	struct nilfs_segment_usage *raw_su;
-	int err;
-
-	err = nilfs_sufile_get_segment_usage(sufile, segnum, &raw_su, &bh_su);
-	if (unlikely(err))
-		return err;
-	nilfs_mdt_mark_buffer_dirty(bh_su);
-	nilfs_mdt_mark_dirty(sufile);
-	nilfs_sufile_put_segment_usage(sufile, segnum, bh_su);
-	return 0;
-}
-
+/**
+ * nilfs_segctor_begin_construction - setup segment buffer to make a new log
+ * @sci: nilfs_sc_info
+ * @nilfs: nilfs object
+ */
 static int nilfs_segctor_begin_construction(struct nilfs_sc_info *sci,
 					    struct the_nilfs *nilfs)
 {
-	struct nilfs_segment_buffer *segbuf, *n;
+	struct nilfs_segment_buffer *segbuf, *prev;
 	__u64 nextnum;
-	int err;
+	int err, alloc = 0;
 
-	if (list_empty(&sci->sc_segbufs)) {
-		segbuf = nilfs_segbuf_new(sci->sc_super);
-		if (unlikely(!segbuf))
-			return -ENOMEM;
-		list_add(&segbuf->sb_list, &sci->sc_segbufs);
-	} else
-		segbuf = NILFS_FIRST_SEGBUF(&sci->sc_segbufs);
+	segbuf = nilfs_segbuf_new(sci->sc_super);
+	if (unlikely(!segbuf))
+		return -ENOMEM;
+
+	if (list_empty(&sci->sc_write_logs)) {
+		nilfs_segbuf_map(segbuf, nilfs->ns_segnum,
+				 nilfs->ns_pseg_offset, nilfs);
+		if (segbuf->sb_rest_blocks < NILFS_PSEG_MIN_BLOCKS) {
+			nilfs_shift_to_next_segment(nilfs);
+			nilfs_segbuf_map(segbuf, nilfs->ns_segnum, 0, nilfs);
+		}
 
-	nilfs_segbuf_map(segbuf, nilfs->ns_segnum, nilfs->ns_pseg_offset,
-			 nilfs);
+		segbuf->sb_sum.seg_seq = nilfs->ns_seg_seq;
+		nextnum = nilfs->ns_nextnum;
 
-	if (segbuf->sb_rest_blocks < NILFS_PSEG_MIN_BLOCKS) {
-		nilfs_shift_to_next_segment(nilfs);
-		nilfs_segbuf_map(segbuf, nilfs->ns_segnum, 0, nilfs);
+		if (nilfs->ns_segnum == nilfs->ns_nextnum)
+			/* Start from the head of a new full segment */
+			alloc++;
+	} else {
+		/* Continue logs */
+		prev = NILFS_LAST_SEGBUF(&sci->sc_write_logs);
+		nilfs_segbuf_map_cont(segbuf, prev);
+		segbuf->sb_sum.seg_seq = prev->sb_sum.seg_seq;
+		nextnum = prev->sb_nextnum;
+
+		if (segbuf->sb_rest_blocks < NILFS_PSEG_MIN_BLOCKS) {
+			nilfs_segbuf_map(segbuf, prev->sb_nextnum, 0, nilfs);
+			segbuf->sb_sum.seg_seq++;
+			alloc++;
+		}
 	}
-	sci->sc_segbuf_nblocks = segbuf->sb_rest_blocks;
 
-	err = nilfs_touch_segusage(nilfs->ns_sufile, segbuf->sb_segnum);
-	if (unlikely(err))
-		return err;
+	err = nilfs_sufile_mark_dirty(nilfs->ns_sufile, segbuf->sb_segnum);
+	if (err)
+		goto failed;
 
-	if (nilfs->ns_segnum == nilfs->ns_nextnum) {
-		/* Start from the head of a new full segment */
+	if (alloc) {
 		err = nilfs_sufile_alloc(nilfs->ns_sufile, &nextnum);
-		if (unlikely(err))
-			return err;
-	} else
-		nextnum = nilfs->ns_nextnum;
-
-	segbuf->sb_sum.seg_seq = nilfs->ns_seg_seq;
+		if (err)
+			goto failed;
+	}
 	nilfs_segbuf_set_next_segnum(segbuf, nextnum, nilfs);
 
-	/* truncating segment buffers */
-	list_for_each_entry_safe_continue(segbuf, n, &sci->sc_segbufs,
-					  sb_list) {
-		list_del_init(&segbuf->sb_list);
-		nilfs_segbuf_free(segbuf);
-	}
+	BUG_ON(!list_empty(&sci->sc_segbufs));
+	list_add_tail(&segbuf->sb_list, &sci->sc_segbufs);
+	sci->sc_segbuf_nblocks = segbuf->sb_rest_blocks;
 	return 0;
+
+ failed:
+	nilfs_segbuf_free(segbuf);
+	return err;
 }
 
 static int nilfs_segctor_extend_segments(struct nilfs_sc_info *sci,
 					 struct the_nilfs *nilfs, int nadd)
 {
-	struct nilfs_segment_buffer *segbuf, *prev, *n;
+	struct nilfs_segment_buffer *segbuf, *prev;
 	struct inode *sufile = nilfs->ns_sufile;
 	__u64 nextnextnum;
 	LIST_HEAD(list);
@@ -1352,7 +1354,7 @@ static int nilfs_segctor_extend_segments(struct nilfs_sc_info *sci,
 	 * not be dirty.  The following call ensures that the buffer is dirty
 	 * and will pin the buffer on memory until the sufile is written.
 	 */
-	err = nilfs_touch_segusage(sufile, prev->sb_nextnum);
+	err = nilfs_sufile_mark_dirty(sufile, prev->sb_nextnum);
 	if (unlikely(err))
 		return err;
 
@@ -1378,33 +1380,33 @@ static int nilfs_segctor_extend_segments(struct nilfs_sc_info *sci,
 		list_add_tail(&segbuf->sb_list, &list);
 		prev = segbuf;
 	}
-	list_splice(&list, sci->sc_segbufs.prev);
+	list_splice_tail(&list, &sci->sc_segbufs);
 	return 0;
 
  failed_segbuf:
 	nilfs_segbuf_free(segbuf);
  failed:
-	list_for_each_entry_safe(segbuf, n, &list, sb_list) {
+	list_for_each_entry(segbuf, &list, sb_list) {
 		ret = nilfs_sufile_free(sufile, segbuf->sb_nextnum);
 		WARN_ON(ret); /* never fails */
-		list_del_init(&segbuf->sb_list);
-		nilfs_segbuf_free(segbuf);
 	}
+	nilfs_destroy_logs(&list);
 	return err;
 }
 
-static void nilfs_segctor_free_incomplete_segments(struct nilfs_sc_info *sci,
-						   struct the_nilfs *nilfs)
+static void nilfs_free_incomplete_logs(struct list_head *logs,
+				       struct the_nilfs *nilfs)
 {
-	struct nilfs_segment_buffer *segbuf;
-	int ret, done = 0;
+	struct nilfs_segment_buffer *segbuf, *prev;
+	struct inode *sufile = nilfs->ns_sufile;
+	int ret;
 
-	segbuf = NILFS_FIRST_SEGBUF(&sci->sc_segbufs);
+	segbuf = NILFS_FIRST_SEGBUF(logs);
 	if (nilfs->ns_nextnum != segbuf->sb_nextnum) {
-		ret = nilfs_sufile_free(nilfs->ns_sufile, segbuf->sb_nextnum);
+		ret = nilfs_sufile_free(sufile, segbuf->sb_nextnum);
 		WARN_ON(ret); /* never fails */
 	}
-	if (segbuf->sb_io_error) {
+	if (atomic_read(&segbuf->sb_err)) {
 		/* Case 1: The first segment failed */
 		if (segbuf->sb_pseg_start != segbuf->sb_fseg_start)
 			/* Case 1a:  Partial segment appended into an existing
@@ -1413,106 +1415,54 @@ static void nilfs_segctor_free_incomplete_segments(struct nilfs_sc_info *sci,
 						segbuf->sb_fseg_end);
 		else /* Case 1b:  New full segment */
 			set_nilfs_discontinued(nilfs);
-		done++;
 	}
 
-	list_for_each_entry_continue(segbuf, &sci->sc_segbufs, sb_list) {
-		ret = nilfs_sufile_free(nilfs->ns_sufile, segbuf->sb_nextnum);
-		WARN_ON(ret); /* never fails */
-		if (!done && segbuf->sb_io_error) {
-			if (segbuf->sb_segnum != nilfs->ns_nextnum)
-				/* Case 2: extended segment (!= next) failed */
-				nilfs_sufile_set_error(nilfs->ns_sufile,
-						       segbuf->sb_segnum);
-			done++;
-		}
-	}
-}
-
-static void nilfs_segctor_clear_segment_buffers(struct nilfs_sc_info *sci)
-{
-	struct nilfs_segment_buffer *segbuf;
-
-	list_for_each_entry(segbuf, &sci->sc_segbufs, sb_list)
-		nilfs_segbuf_clear(segbuf);
-	sci->sc_super_root = NULL;
-}
-
-static void nilfs_segctor_destroy_segment_buffers(struct nilfs_sc_info *sci)
-{
-	struct nilfs_segment_buffer *segbuf;
-
-	while (!list_empty(&sci->sc_segbufs)) {
-		segbuf = NILFS_FIRST_SEGBUF(&sci->sc_segbufs);
-		list_del_init(&segbuf->sb_list);
-		nilfs_segbuf_free(segbuf);
-	}
-	/* sci->sc_curseg = NULL; */
-}
-
-static void nilfs_segctor_end_construction(struct nilfs_sc_info *sci,
-					   struct the_nilfs *nilfs, int err)
-{
-	if (unlikely(err)) {
-		nilfs_segctor_free_incomplete_segments(sci, nilfs);
-		if (sci->sc_stage.flags & NILFS_CF_SUFREED) {
-			int ret;
-
-			ret = nilfs_sufile_cancel_freev(nilfs->ns_sufile,
-							sci->sc_freesegs,
-							sci->sc_nfreesegs,
-							NULL);
-			WARN_ON(ret); /* do not happen */
+	prev = segbuf;
+	list_for_each_entry_continue(segbuf, logs, sb_list) {
+		if (prev->sb_nextnum != segbuf->sb_nextnum) {
+			ret = nilfs_sufile_free(sufile, segbuf->sb_nextnum);
+			WARN_ON(ret); /* never fails */
 		}
+		if (atomic_read(&segbuf->sb_err) &&
+		    segbuf->sb_segnum != nilfs->ns_nextnum)
+			/* Case 2: extended segment (!= next) failed */
+			nilfs_sufile_set_error(sufile, segbuf->sb_segnum);
+		prev = segbuf;
 	}
-	nilfs_segctor_clear_segment_buffers(sci);
 }
 
 static void nilfs_segctor_update_segusage(struct nilfs_sc_info *sci,
 					  struct inode *sufile)
 {
 	struct nilfs_segment_buffer *segbuf;
-	struct buffer_head *bh_su;
-	struct nilfs_segment_usage *raw_su;
 	unsigned long live_blocks;
 	int ret;
 
 	list_for_each_entry(segbuf, &sci->sc_segbufs, sb_list) {
-		ret = nilfs_sufile_get_segment_usage(sufile, segbuf->sb_segnum,
-						     &raw_su, &bh_su);
-		WARN_ON(ret); /* always succeed because bh_su is dirty */
 		live_blocks = segbuf->sb_sum.nblocks +
 			(segbuf->sb_pseg_start - segbuf->sb_fseg_start);
-		raw_su->su_lastmod = cpu_to_le64(sci->sc_seg_ctime);
-		raw_su->su_nblocks = cpu_to_le32(live_blocks);
-		nilfs_sufile_put_segment_usage(sufile, segbuf->sb_segnum,
-					       bh_su);
+		ret = nilfs_sufile_set_segment_usage(sufile, segbuf->sb_segnum,
+						     live_blocks,
+						     sci->sc_seg_ctime);
+		WARN_ON(ret); /* always succeed because the segusage is dirty */
 	}
 }
 
-static void nilfs_segctor_cancel_segusage(struct nilfs_sc_info *sci,
-					  struct inode *sufile)
+static void nilfs_cancel_segusage(struct list_head *logs, struct inode *sufile)
 {
 	struct nilfs_segment_buffer *segbuf;
-	struct buffer_head *bh_su;
-	struct nilfs_segment_usage *raw_su;
 	int ret;
 
-	segbuf = NILFS_FIRST_SEGBUF(&sci->sc_segbufs);
-	ret = nilfs_sufile_get_segment_usage(sufile, segbuf->sb_segnum,
-					     &raw_su, &bh_su);
-	WARN_ON(ret); /* always succeed because bh_su is dirty */
-	raw_su->su_nblocks = cpu_to_le32(segbuf->sb_pseg_start -
-					 segbuf->sb_fseg_start);
-	nilfs_sufile_put_segment_usage(sufile, segbuf->sb_segnum, bh_su);
+	segbuf = NILFS_FIRST_SEGBUF(logs);
+	ret = nilfs_sufile_set_segment_usage(sufile, segbuf->sb_segnum,
+					     segbuf->sb_pseg_start -
+					     segbuf->sb_fseg_start, 0);
+	WARN_ON(ret); /* always succeed because the segusage is dirty */
 
-	list_for_each_entry_continue(segbuf, &sci->sc_segbufs, sb_list) {
-		ret = nilfs_sufile_get_segment_usage(sufile, segbuf->sb_segnum,
-						     &raw_su, &bh_su);
+	list_for_each_entry_continue(segbuf, logs, sb_list) {
+		ret = nilfs_sufile_set_segment_usage(sufile, segbuf->sb_segnum,
+						     0, 0);
 		WARN_ON(ret); /* always succeed */
-		raw_su->su_nblocks = 0;
-		nilfs_sufile_put_segment_usage(sufile, segbuf->sb_segnum,
-					       bh_su);
 	}
 }
 
@@ -1520,17 +1470,15 @@ static void nilfs_segctor_truncate_segments(struct nilfs_sc_info *sci,
 					    struct nilfs_segment_buffer *last,
 					    struct inode *sufile)
 {
-	struct nilfs_segment_buffer *segbuf = last, *n;
+	struct nilfs_segment_buffer *segbuf = last;
 	int ret;
 
-	list_for_each_entry_safe_continue(segbuf, n, &sci->sc_segbufs,
-					  sb_list) {
-		list_del_init(&segbuf->sb_list);
+	list_for_each_entry_continue(segbuf, &sci->sc_segbufs, sb_list) {
 		sci->sc_segbuf_nblocks -= segbuf->sb_rest_blocks;
 		ret = nilfs_sufile_free(sufile, segbuf->sb_nextnum);
 		WARN_ON(ret);
-		nilfs_segbuf_free(segbuf);
 	}
+	nilfs_truncate_logs(&sci->sc_segbufs, last);
 }
 
 
@@ -1569,7 +1517,7 @@ static int nilfs_segctor_collect(struct nilfs_sc_info *sci,
 							NULL);
 			WARN_ON(err); /* do not happen */
 		}
-		nilfs_segctor_clear_segment_buffers(sci);
+		nilfs_clear_logs(&sci->sc_segbufs);
 
 		err = nilfs_segctor_extend_segments(sci, nilfs, nadd);
 		if (unlikely(err))
@@ -1814,26 +1762,18 @@ static int nilfs_segctor_prepare_write(struct nilfs_sc_info *sci,
 }
 
 static int nilfs_segctor_write(struct nilfs_sc_info *sci,
-			       struct backing_dev_info *bdi)
+			       struct the_nilfs *nilfs)
 {
 	struct nilfs_segment_buffer *segbuf;
-	struct nilfs_write_info wi;
-	int err, res;
-
-	wi.sb = sci->sc_super;
-	wi.bh_sr = sci->sc_super_root;
-	wi.bdi = bdi;
+	int ret = 0;
 
 	list_for_each_entry(segbuf, &sci->sc_segbufs, sb_list) {
-		nilfs_segbuf_prepare_write(segbuf, &wi);
-		err = nilfs_segbuf_write(segbuf, &wi);
-
-		res = nilfs_segbuf_wait(segbuf, &wi);
-		err = err ? : res;
-		if (err)
-			return err;
+		ret = nilfs_segbuf_write(segbuf, nilfs);
+		if (ret)
+			break;
 	}
-	return 0;
+	list_splice_tail_init(&sci->sc_segbufs, &sci->sc_write_logs);
+	return ret;
 }
 
 static void __nilfs_end_page_io(struct page *page, int err)
@@ -1911,15 +1851,17 @@ static void nilfs_clear_copied_buffers(struct list_head *list, int err)
 	}
 }
 
-static void nilfs_segctor_abort_write(struct nilfs_sc_info *sci,
-				      struct page *failed_page, int err)
+static void nilfs_abort_logs(struct list_head *logs, struct page *failed_page,
+			     struct buffer_head *bh_sr, int err)
 {
 	struct nilfs_segment_buffer *segbuf;
 	struct page *bd_page = NULL, *fs_page = NULL;
+	struct buffer_head *bh;
 
-	list_for_each_entry(segbuf, &sci->sc_segbufs, sb_list) {
-		struct buffer_head *bh;
+	if (list_empty(logs))
+		return;
 
+	list_for_each_entry(segbuf, logs, sb_list) {
 		list_for_each_entry(bh, &segbuf->sb_segsum_buffers,
 				    b_assoc_buffers) {
 			if (bh->b_page != bd_page) {
@@ -1931,7 +1873,7 @@ static void nilfs_segctor_abort_write(struct nilfs_sc_info *sci,
 
 		list_for_each_entry(bh, &segbuf->sb_payload_buffers,
 				    b_assoc_buffers) {
-			if (bh == sci->sc_super_root) {
+			if (bh == bh_sr) {
 				if (bh->b_page != bd_page) {
 					end_page_writeback(bd_page);
 					bd_page = bh->b_page;
@@ -1941,7 +1883,7 @@ static void nilfs_segctor_abort_write(struct nilfs_sc_info *sci,
 			if (bh->b_page != fs_page) {
 				nilfs_end_page_io(fs_page, err);
 				if (fs_page && fs_page == failed_page)
-					goto done;
+					return;
 				fs_page = bh->b_page;
 			}
 		}
@@ -1950,8 +1892,34 @@ static void nilfs_segctor_abort_write(struct nilfs_sc_info *sci,
 		end_page_writeback(bd_page);
 
 	nilfs_end_page_io(fs_page, err);
- done:
+}
+
+static void nilfs_segctor_abort_construction(struct nilfs_sc_info *sci,
+					     struct the_nilfs *nilfs, int err)
+{
+	LIST_HEAD(logs);
+	int ret;
+
+	list_splice_tail_init(&sci->sc_write_logs, &logs);
+	ret = nilfs_wait_on_logs(&logs);
+	if (ret)
+		nilfs_abort_logs(&logs, NULL, sci->sc_super_root, ret);
+
+	list_splice_tail_init(&sci->sc_segbufs, &logs);
+	nilfs_cancel_segusage(&logs, nilfs->ns_sufile);
+	nilfs_free_incomplete_logs(&logs, nilfs);
 	nilfs_clear_copied_buffers(&sci->sc_copied_buffers, err);
+
+	if (sci->sc_stage.flags & NILFS_CF_SUFREED) {
+		ret = nilfs_sufile_cancel_freev(nilfs->ns_sufile,
+						sci->sc_freesegs,
+						sci->sc_nfreesegs,
+						NULL);
+		WARN_ON(ret); /* do not happen */
+	}
+
+	nilfs_destroy_logs(&logs);
+	sci->sc_super_root = NULL;
 }
 
 static void nilfs_set_next_segment(struct the_nilfs *nilfs,
@@ -1973,7 +1941,7 @@ static void nilfs_segctor_complete_write(struct nilfs_sc_info *sci)
 	struct the_nilfs *nilfs = sbi->s_nilfs;
 	int update_sr = (sci->sc_super_root != NULL);
 
-	list_for_each_entry(segbuf, &sci->sc_segbufs, sb_list) {
+	list_for_each_entry(segbuf, &sci->sc_write_logs, sb_list) {
 		struct buffer_head *bh;
 
 		list_for_each_entry(bh, &segbuf->sb_segsum_buffers,
@@ -2046,7 +2014,7 @@ static void nilfs_segctor_complete_write(struct nilfs_sc_info *sci)
 
 	sci->sc_nblk_inc += sci->sc_nblk_this_inc;
 
-	segbuf = NILFS_LAST_SEGBUF(&sci->sc_segbufs);
+	segbuf = NILFS_LAST_SEGBUF(&sci->sc_write_logs);
 	nilfs_set_next_segment(nilfs, segbuf);
 
 	if (update_sr) {
@@ -2057,10 +2025,23 @@ static void nilfs_segctor_complete_write(struct nilfs_sc_info *sci)
 		clear_bit(NILFS_SC_HAVE_DELTA, &sci->sc_flags);
 		clear_bit(NILFS_SC_DIRTY, &sci->sc_flags);
 		set_bit(NILFS_SC_SUPER_ROOT, &sci->sc_flags);
+		nilfs_segctor_clear_metadata_dirty(sci);
 	} else
 		clear_bit(NILFS_SC_SUPER_ROOT, &sci->sc_flags);
 }
 
+static int nilfs_segctor_wait(struct nilfs_sc_info *sci)
+{
+	int ret;
+
+	ret = nilfs_wait_on_logs(&sci->sc_write_logs);
+	if (!ret) {
+		nilfs_segctor_complete_write(sci);
+		nilfs_destroy_logs(&sci->sc_write_logs);
+	}
+	return ret;
+}
+
 static int nilfs_segctor_check_in_files(struct nilfs_sc_info *sci,
 					struct nilfs_sb_info *sbi)
 {
@@ -2173,7 +2154,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 		/* Avoid empty segment */
 		if (sci->sc_stage.scnt == NILFS_ST_DONE &&
 		    NILFS_SEG_EMPTY(&sci->sc_curseg->sb_sum)) {
-			nilfs_segctor_end_construction(sci, nilfs, 1);
+			nilfs_segctor_abort_construction(sci, nilfs, 1);
 			goto out;
 		}
 
@@ -2187,7 +2168,7 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 		if (has_sr) {
 			err = nilfs_segctor_fill_in_checkpoint(sci);
 			if (unlikely(err))
-				goto failed_to_make_up;
+				goto failed_to_write;
 
 			nilfs_segctor_fill_in_super_root(sci, nilfs);
 		}
@@ -2195,42 +2176,46 @@ static int nilfs_segctor_do_construct(struct nilfs_sc_info *sci, int mode)
 
 		/* Write partial segments */
 		err = nilfs_segctor_prepare_write(sci, &failed_page);
-		if (unlikely(err))
+		if (err) {
+			nilfs_abort_logs(&sci->sc_segbufs, failed_page,
+					 sci->sc_super_root, err);
 			goto failed_to_write;
-
+		}
 		nilfs_segctor_fill_in_checksums(sci, nilfs->ns_crc_seed);
 
-		err = nilfs_segctor_write(sci, nilfs->ns_bdi);
+		err = nilfs_segctor_write(sci, nilfs);
 		if (unlikely(err))
 			goto failed_to_write;
 
-		nilfs_segctor_complete_write(sci);
-
-		/* Commit segments */
-		if (has_sr)
-			nilfs_segctor_clear_metadata_dirty(sci);
-
-		nilfs_segctor_end_construction(sci, nilfs, 0);
-
+		if (sci->sc_stage.scnt == NILFS_ST_DONE ||
+		    nilfs->ns_blocksize_bits != PAGE_CACHE_SHIFT) {
+			/*
+			 * At this point, we avoid double buffering
+			 * for blocksize < pagesize because page dirty
+			 * flag is turned off during write and dirty
+			 * buffers are not properly collected for
+			 * pages crossing over segments.
+			 */
+			err = nilfs_segctor_wait(sci);
+			if (err)
+				goto failed_to_write;
+		}
 	} while (sci->sc_stage.scnt != NILFS_ST_DONE);
 
+	sci->sc_super_root = NULL;
+
  out:
-	nilfs_segctor_destroy_segment_buffers(sci);
 	nilfs_segctor_check_out_files(sci, sbi);
 	return err;
 
  failed_to_write:
-	nilfs_segctor_abort_write(sci, failed_page, err);
-	nilfs_segctor_cancel_segusage(sci, nilfs->ns_sufile);
-
- failed_to_make_up:
 	if (sci->sc_stage.flags & NILFS_CF_IFILE_STARTED)
 		nilfs_redirty_inodes(&sci->sc_dirty_files);
 
  failed:
 	if (nilfs_doing_gc())
 		nilfs_redirty_inodes(&sci->sc_gc_inodes);
-	nilfs_segctor_end_construction(sci, nilfs, err);
+	nilfs_segctor_abort_construction(sci, nilfs, err);
 	goto out;
 }
 
@@ -2559,7 +2544,7 @@ int nilfs_clean_segments(struct super_block *sb, struct nilfs_argv *argv,
 
 	sci->sc_freesegs = kbufs[4];
 	sci->sc_nfreesegs = argv[4].v_nmembs;
-	list_splice_init(&nilfs->ns_gc_inodes, sci->sc_gc_inodes.prev);
+	list_splice_tail_init(&nilfs->ns_gc_inodes, &sci->sc_gc_inodes);
 
 	for (;;) {
 		nilfs_segctor_accept(sci, &req);
@@ -2788,6 +2773,7 @@ static struct nilfs_sc_info *nilfs_segctor_new(struct nilfs_sb_info *sbi)
 	spin_lock_init(&sci->sc_state_lock);
 	INIT_LIST_HEAD(&sci->sc_dirty_files);
 	INIT_LIST_HEAD(&sci->sc_segbufs);
+	INIT_LIST_HEAD(&sci->sc_write_logs);
 	INIT_LIST_HEAD(&sci->sc_gc_inodes);
 	INIT_LIST_HEAD(&sci->sc_copied_buffers);
 
@@ -2855,6 +2841,7 @@ static void nilfs_segctor_destroy(struct nilfs_sc_info *sci)
 	}
 
 	WARN_ON(!list_empty(&sci->sc_segbufs));
+	WARN_ON(!list_empty(&sci->sc_write_logs));
 
 	down_write(&sbi->s_nilfs->ns_segctor_sem);
 
diff --git a/fs/nilfs2/segment.h b/fs/nilfs2/segment.h
index 0d2a475a741b..3d3ab2f9864c 100644
--- a/fs/nilfs2/segment.h
+++ b/fs/nilfs2/segment.h
@@ -97,6 +97,7 @@ struct nilfs_segsum_pointer {
  * @sc_dsync_start: start byte offset of data pages
  * @sc_dsync_end: end byte offset of data pages (inclusive)
  * @sc_segbufs: List of segment buffers
+ * @sc_write_logs: List of segment buffers to hold logs under writing
  * @sc_segbuf_nblocks: Number of available blocks in segment buffers.
  * @sc_curseg: Current segment buffer
  * @sc_super_root: Pointer to the super root buffer
@@ -143,6 +144,7 @@ struct nilfs_sc_info {
 
 	/* Segment buffers */
 	struct list_head	sc_segbufs;
+	struct list_head	sc_write_logs;
 	unsigned long		sc_segbuf_nblocks;
 	struct nilfs_segment_buffer *sc_curseg;
 	struct buffer_head     *sc_super_root;
diff --git a/fs/nilfs2/sufile.c b/fs/nilfs2/sufile.c
index 37994d4a59cc..b6c36d0cc331 100644
--- a/fs/nilfs2/sufile.c
+++ b/fs/nilfs2/sufile.c
@@ -31,6 +31,16 @@
 #include "sufile.h"
 
 
+struct nilfs_sufile_info {
+	struct nilfs_mdt_info mi;
+	unsigned long ncleansegs;
+};
+
+static inline struct nilfs_sufile_info *NILFS_SUI(struct inode *sufile)
+{
+	return (struct nilfs_sufile_info *)NILFS_MDT(sufile);
+}
+
 static inline unsigned long
 nilfs_sufile_segment_usages_per_block(const struct inode *sufile)
 {
@@ -62,14 +72,6 @@ nilfs_sufile_segment_usages_in_block(const struct inode *sufile, __u64 curr,
 		     max - curr + 1);
 }
 
-static inline struct nilfs_sufile_header *
-nilfs_sufile_block_get_header(const struct inode *sufile,
-			      struct buffer_head *bh,
-			      void *kaddr)
-{
-	return kaddr + bh_offset(bh);
-}
-
 static struct nilfs_segment_usage *
 nilfs_sufile_block_get_segment_usage(const struct inode *sufile, __u64 segnum,
 				     struct buffer_head *bh, void *kaddr)
@@ -110,6 +112,15 @@ static void nilfs_sufile_mod_counter(struct buffer_head *header_bh,
 }
 
 /**
+ * nilfs_sufile_get_ncleansegs - return the number of clean segments
+ * @sufile: inode of segment usage file
+ */
+unsigned long nilfs_sufile_get_ncleansegs(struct inode *sufile)
+{
+	return NILFS_SUI(sufile)->ncleansegs;
+}
+
+/**
  * nilfs_sufile_updatev - modify multiple segment usages at a time
  * @sufile: inode of segment usage file
  * @segnumv: array of segment numbers
@@ -270,7 +281,7 @@ int nilfs_sufile_alloc(struct inode *sufile, __u64 *segnump)
 	if (ret < 0)
 		goto out_sem;
 	kaddr = kmap_atomic(header_bh->b_page, KM_USER0);
-	header = nilfs_sufile_block_get_header(sufile, header_bh, kaddr);
+	header = kaddr + bh_offset(header_bh);
 	ncleansegs = le64_to_cpu(header->sh_ncleansegs);
 	last_alloc = le64_to_cpu(header->sh_last_alloc);
 	kunmap_atomic(kaddr, KM_USER0);
@@ -302,13 +313,13 @@ int nilfs_sufile_alloc(struct inode *sufile, __u64 *segnump)
 			kunmap_atomic(kaddr, KM_USER0);
 
 			kaddr = kmap_atomic(header_bh->b_page, KM_USER0);
-			header = nilfs_sufile_block_get_header(
-				sufile, header_bh, kaddr);
+			header = kaddr + bh_offset(header_bh);
 			le64_add_cpu(&header->sh_ncleansegs, -1);
 			le64_add_cpu(&header->sh_ndirtysegs, 1);
 			header->sh_last_alloc = cpu_to_le64(segnum);
 			kunmap_atomic(kaddr, KM_USER0);
 
+			NILFS_SUI(sufile)->ncleansegs--;
 			nilfs_mdt_mark_buffer_dirty(header_bh);
 			nilfs_mdt_mark_buffer_dirty(su_bh);
 			nilfs_mdt_mark_dirty(sufile);
@@ -351,6 +362,8 @@ void nilfs_sufile_do_cancel_free(struct inode *sufile, __u64 segnum,
 	kunmap_atomic(kaddr, KM_USER0);
 
 	nilfs_sufile_mod_counter(header_bh, -1, 1);
+	NILFS_SUI(sufile)->ncleansegs--;
+
 	nilfs_mdt_mark_buffer_dirty(su_bh);
 	nilfs_mdt_mark_dirty(sufile);
 }
@@ -380,6 +393,8 @@ void nilfs_sufile_do_scrap(struct inode *sufile, __u64 segnum,
 	kunmap_atomic(kaddr, KM_USER0);
 
 	nilfs_sufile_mod_counter(header_bh, clean ? (u64)-1 : 0, dirty ? 0 : 1);
+	NILFS_SUI(sufile)->ncleansegs -= clean;
+
 	nilfs_mdt_mark_buffer_dirty(su_bh);
 	nilfs_mdt_mark_dirty(sufile);
 }
@@ -409,79 +424,65 @@ void nilfs_sufile_do_free(struct inode *sufile, __u64 segnum,
 	nilfs_mdt_mark_buffer_dirty(su_bh);
 
 	nilfs_sufile_mod_counter(header_bh, 1, sudirty ? (u64)-1 : 0);
+	NILFS_SUI(sufile)->ncleansegs++;
+
 	nilfs_mdt_mark_dirty(sufile);
 }
 
 /**
- * nilfs_sufile_get_segment_usage - get a segment usage
+ * nilfs_sufile_mark_dirty - mark the buffer having a segment usage dirty
  * @sufile: inode of segment usage file
  * @segnum: segment number
- * @sup: pointer to segment usage
- * @bhp: pointer to buffer head
- *
- * Description: nilfs_sufile_get_segment_usage() acquires the segment usage
- * specified by @segnum.
- *
- * Return Value: On success, 0 is returned, and the segment usage and the
- * buffer head of the buffer on which the segment usage is located are stored
- * in the place pointed by @sup and @bhp, respectively. On error, one of the
- * following negative error codes is returned.
- *
- * %-EIO - I/O error.
- *
- * %-ENOMEM - Insufficient amount of memory available.
- *
- * %-EINVAL - Invalid segment usage number.
  */
-int nilfs_sufile_get_segment_usage(struct inode *sufile, __u64 segnum,
-				   struct nilfs_segment_usage **sup,
-				   struct buffer_head **bhp)
+int nilfs_sufile_mark_dirty(struct inode *sufile, __u64 segnum)
 {
 	struct buffer_head *bh;
-	struct nilfs_segment_usage *su;
-	void *kaddr;
 	int ret;
 
-	/* segnum is 0 origin */
-	if (segnum >= nilfs_sufile_get_nsegments(sufile))
-		return -EINVAL;
-	down_write(&NILFS_MDT(sufile)->mi_sem);
-	ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, 1, &bh);
-	if (ret < 0)
-		goto out_sem;
-	kaddr = kmap(bh->b_page);
-	su = nilfs_sufile_block_get_segment_usage(sufile, segnum, bh, kaddr);
-	if (nilfs_segment_usage_error(su)) {
-		kunmap(bh->b_page);
+	ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, 0, &bh);
+	if (!ret) {
+		nilfs_mdt_mark_buffer_dirty(bh);
+		nilfs_mdt_mark_dirty(sufile);
 		brelse(bh);
-		ret = -EINVAL;
-		goto out_sem;
 	}
-
-	if (sup != NULL)
-		*sup = su;
-	*bhp = bh;
-
- out_sem:
-	up_write(&NILFS_MDT(sufile)->mi_sem);
 	return ret;
 }
 
 /**
- * nilfs_sufile_put_segment_usage - put a segment usage
+ * nilfs_sufile_set_segment_usage - set usage of a segment
  * @sufile: inode of segment usage file
  * @segnum: segment number
- * @bh: buffer head
- *
- * Description: nilfs_sufile_put_segment_usage() releases the segment usage
- * specified by @segnum. @bh must be the buffer head which have been returned
- * by a previous call to nilfs_sufile_get_segment_usage() with @segnum.
+ * @nblocks: number of live blocks in the segment
+ * @modtime: modification time (option)
  */
-void nilfs_sufile_put_segment_usage(struct inode *sufile, __u64 segnum,
-				    struct buffer_head *bh)
+int nilfs_sufile_set_segment_usage(struct inode *sufile, __u64 segnum,
+				   unsigned long nblocks, time_t modtime)
 {
-	kunmap(bh->b_page);
+	struct buffer_head *bh;
+	struct nilfs_segment_usage *su;
+	void *kaddr;
+	int ret;
+
+	down_write(&NILFS_MDT(sufile)->mi_sem);
+	ret = nilfs_sufile_get_segment_usage_block(sufile, segnum, 0, &bh);
+	if (ret < 0)
+		goto out_sem;
+
+	kaddr = kmap_atomic(bh->b_page, KM_USER0);
+	su = nilfs_sufile_block_get_segment_usage(sufile, segnum, bh, kaddr);
+	WARN_ON(nilfs_segment_usage_error(su));
+	if (modtime)
+		su->su_lastmod = cpu_to_le64(modtime);
+	su->su_nblocks = cpu_to_le32(nblocks);
+	kunmap_atomic(kaddr, KM_USER0);
+
+	nilfs_mdt_mark_buffer_dirty(bh);
+	nilfs_mdt_mark_dirty(sufile);
 	brelse(bh);
+
+ out_sem:
+	up_write(&NILFS_MDT(sufile)->mi_sem);
+	return ret;
 }
 
 /**
@@ -515,7 +516,7 @@ int nilfs_sufile_get_stat(struct inode *sufile, struct nilfs_sustat *sustat)
 		goto out_sem;
 
 	kaddr = kmap_atomic(header_bh->b_page, KM_USER0);
-	header = nilfs_sufile_block_get_header(sufile, header_bh, kaddr);
+	header = kaddr + bh_offset(header_bh);
 	sustat->ss_nsegs = nilfs_sufile_get_nsegments(sufile);
 	sustat->ss_ncleansegs = le64_to_cpu(header->sh_ncleansegs);
 	sustat->ss_ndirtysegs = le64_to_cpu(header->sh_ndirtysegs);
@@ -532,33 +533,6 @@ int nilfs_sufile_get_stat(struct inode *sufile, struct nilfs_sustat *sustat)
 	return ret;
 }
 
-/**
- * nilfs_sufile_get_ncleansegs - get the number of clean segments
- * @sufile: inode of segment usage file
- * @nsegsp: pointer to the number of clean segments
- *
- * Description: nilfs_sufile_get_ncleansegs() acquires the number of clean
- * segments.
- *
- * Return Value: On success, 0 is returned and the number of clean segments is
- * stored in the place pointed by @nsegsp. On error, one of the following
- * negative error codes is returned.
- *
- * %-EIO - I/O error.
- *
- * %-ENOMEM - Insufficient amount of memory available.
- */
-int nilfs_sufile_get_ncleansegs(struct inode *sufile, unsigned long *nsegsp)
-{
-	struct nilfs_sustat sustat;
-	int ret;
-
-	ret = nilfs_sufile_get_stat(sufile, &sustat);
-	if (ret == 0)
-		*nsegsp = sustat.ss_ncleansegs;
-	return ret;
-}
-
 void nilfs_sufile_do_set_error(struct inode *sufile, __u64 segnum,
 			       struct buffer_head *header_bh,
 			       struct buffer_head *su_bh)
@@ -577,8 +551,10 @@ void nilfs_sufile_do_set_error(struct inode *sufile, __u64 segnum,
 	nilfs_segment_usage_set_error(su);
 	kunmap_atomic(kaddr, KM_USER0);
 
-	if (suclean)
+	if (suclean) {
 		nilfs_sufile_mod_counter(header_bh, -1, 0);
+		NILFS_SUI(sufile)->ncleansegs--;
+	}
 	nilfs_mdt_mark_buffer_dirty(su_bh);
 	nilfs_mdt_mark_dirty(sufile);
 }
@@ -657,3 +633,48 @@ ssize_t nilfs_sufile_get_suinfo(struct inode *sufile, __u64 segnum, void *buf,
 	up_read(&NILFS_MDT(sufile)->mi_sem);
 	return ret;
 }
+
+/**
+ * nilfs_sufile_read - read sufile inode
+ * @sufile: sufile inode
+ * @raw_inode: on-disk sufile inode
+ */
+int nilfs_sufile_read(struct inode *sufile, struct nilfs_inode *raw_inode)
+{
+	struct nilfs_sufile_info *sui = NILFS_SUI(sufile);
+	struct buffer_head *header_bh;
+	struct nilfs_sufile_header *header;
+	void *kaddr;
+	int ret;
+
+	ret = nilfs_read_inode_common(sufile, raw_inode);
+	if (ret < 0)
+		return ret;
+
+	ret = nilfs_sufile_get_header_block(sufile, &header_bh);
+	if (!ret) {
+		kaddr = kmap_atomic(header_bh->b_page, KM_USER0);
+		header = kaddr + bh_offset(header_bh);
+		sui->ncleansegs = le64_to_cpu(header->sh_ncleansegs);
+		kunmap_atomic(kaddr, KM_USER0);
+		brelse(header_bh);
+	}
+	return ret;
+}
+
+/**
+ * nilfs_sufile_new - create sufile
+ * @nilfs: nilfs object
+ * @susize: size of a segment usage entry
+ */
+struct inode *nilfs_sufile_new(struct the_nilfs *nilfs, size_t susize)
+{
+	struct inode *sufile;
+
+	sufile = nilfs_mdt_new(nilfs, NULL, NILFS_SUFILE_INO,
+			       sizeof(struct nilfs_sufile_info));
+	if (sufile)
+		nilfs_mdt_set_entry_size(sufile, susize,
+					 sizeof(struct nilfs_sufile_header));
+	return sufile;
+}
diff --git a/fs/nilfs2/sufile.h b/fs/nilfs2/sufile.h
index 0e99e5c0bd0f..15163b8aff7d 100644
--- a/fs/nilfs2/sufile.h
+++ b/fs/nilfs2/sufile.h
@@ -34,14 +34,13 @@ static inline unsigned long nilfs_sufile_get_nsegments(struct inode *sufile)
 	return NILFS_MDT(sufile)->mi_nilfs->ns_nsegments;
 }
 
+unsigned long nilfs_sufile_get_ncleansegs(struct inode *sufile);
+
 int nilfs_sufile_alloc(struct inode *, __u64 *);
-int nilfs_sufile_get_segment_usage(struct inode *, __u64,
-				   struct nilfs_segment_usage **,
-				   struct buffer_head **);
-void nilfs_sufile_put_segment_usage(struct inode *, __u64,
-				    struct buffer_head *);
+int nilfs_sufile_mark_dirty(struct inode *sufile, __u64 segnum);
+int nilfs_sufile_set_segment_usage(struct inode *sufile, __u64 segnum,
+				   unsigned long nblocks, time_t modtime);
 int nilfs_sufile_get_stat(struct inode *, struct nilfs_sustat *);
-int nilfs_sufile_get_ncleansegs(struct inode *, unsigned long *);
 ssize_t nilfs_sufile_get_suinfo(struct inode *, __u64, void *, unsigned,
 				size_t);
 
@@ -62,6 +61,9 @@ void nilfs_sufile_do_cancel_free(struct inode *, __u64, struct buffer_head *,
 void nilfs_sufile_do_set_error(struct inode *, __u64, struct buffer_head *,
 			       struct buffer_head *);
 
+int nilfs_sufile_read(struct inode *sufile, struct nilfs_inode *raw_inode);
+struct inode *nilfs_sufile_new(struct the_nilfs *nilfs, size_t susize);
+
 /**
  * nilfs_sufile_scrap - make a segment garbage
  * @sufile: inode of segment usage file
diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c
index 644e66727dd0..8173faee31e6 100644
--- a/fs/nilfs2/super.c
+++ b/fs/nilfs2/super.c
@@ -363,14 +363,10 @@ int nilfs_attach_checkpoint(struct nilfs_sb_info *sbi, __u64 cno)
 	list_add(&sbi->s_list, &nilfs->ns_supers);
 	up_write(&nilfs->ns_super_sem);
 
-	sbi->s_ifile = nilfs_mdt_new(nilfs, sbi->s_super, NILFS_IFILE_INO);
+	sbi->s_ifile = nilfs_ifile_new(sbi, nilfs->ns_inode_size);
 	if (!sbi->s_ifile)
 		return -ENOMEM;
 
-	err = nilfs_palloc_init_blockgroup(sbi->s_ifile, nilfs->ns_inode_size);
-	if (unlikely(err))
-		goto failed;
-
 	down_read(&nilfs->ns_segctor_sem);
 	err = nilfs_cpfile_get_checkpoint(nilfs->ns_cpfile, cno, 0, &raw_cp,
 					  &bh_cp);
@@ -411,7 +407,6 @@ void nilfs_detach_checkpoint(struct nilfs_sb_info *sbi)
 {
 	struct the_nilfs *nilfs = sbi->s_nilfs;
 
-	nilfs_mdt_clear(sbi->s_ifile);
 	nilfs_mdt_destroy(sbi->s_ifile);
 	sbi->s_ifile = NULL;
 	down_write(&nilfs->ns_super_sem);
@@ -419,22 +414,6 @@ void nilfs_detach_checkpoint(struct nilfs_sb_info *sbi)
 	up_write(&nilfs->ns_super_sem);
 }
 
-static int nilfs_mark_recovery_complete(struct nilfs_sb_info *sbi)
-{
-	struct the_nilfs *nilfs = sbi->s_nilfs;
-	int err = 0;
-
-	down_write(&nilfs->ns_sem);
-	if (!(nilfs->ns_mount_state & NILFS_VALID_FS)) {
-		nilfs->ns_mount_state |= NILFS_VALID_FS;
-		err = nilfs_commit_super(sbi, 1);
-		if (likely(!err))
-			printk(KERN_INFO "NILFS: recovery complete.\n");
-	}
-	up_write(&nilfs->ns_sem);
-	return err;
-}
-
 static int nilfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct super_block *sb = dentry->d_sb;
@@ -490,7 +469,7 @@ static int nilfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
 	struct nilfs_sb_info *sbi = NILFS_SB(sb);
 
 	if (!nilfs_test_opt(sbi, BARRIER))
-		seq_printf(seq, ",barrier=off");
+		seq_printf(seq, ",nobarrier");
 	if (nilfs_test_opt(sbi, SNAPSHOT))
 		seq_printf(seq, ",cp=%llu",
 			   (unsigned long long int)sbi->s_snapshot_cno);
@@ -500,6 +479,8 @@ static int nilfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
 		seq_printf(seq, ",errors=panic");
 	if (nilfs_test_opt(sbi, STRICT_ORDER))
 		seq_printf(seq, ",order=strict");
+	if (nilfs_test_opt(sbi, NORECOVERY))
+		seq_printf(seq, ",norecovery");
 
 	return 0;
 }
@@ -568,7 +549,7 @@ static const struct export_operations nilfs_export_ops = {
 
 enum {
 	Opt_err_cont, Opt_err_panic, Opt_err_ro,
-	Opt_barrier, Opt_snapshot, Opt_order,
+	Opt_nobarrier, Opt_snapshot, Opt_order, Opt_norecovery,
 	Opt_err,
 };
 
@@ -576,25 +557,13 @@ static match_table_t tokens = {
 	{Opt_err_cont, "errors=continue"},
 	{Opt_err_panic, "errors=panic"},
 	{Opt_err_ro, "errors=remount-ro"},
-	{Opt_barrier, "barrier=%s"},
+	{Opt_nobarrier, "nobarrier"},
 	{Opt_snapshot, "cp=%u"},
 	{Opt_order, "order=%s"},
+	{Opt_norecovery, "norecovery"},
 	{Opt_err, NULL}
 };
 
-static int match_bool(substring_t *s, int *result)
-{
-	int len = s->to - s->from;
-
-	if (strncmp(s->from, "on", len) == 0)
-		*result = 1;
-	else if (strncmp(s->from, "off", len) == 0)
-		*result = 0;
-	else
-		return 1;
-	return 0;
-}
-
 static int parse_options(char *options, struct super_block *sb)
 {
 	struct nilfs_sb_info *sbi = NILFS_SB(sb);
@@ -612,13 +581,8 @@ static int parse_options(char *options, struct super_block *sb)
 
 		token = match_token(p, tokens, args);
 		switch (token) {
-		case Opt_barrier:
-			if (match_bool(&args[0], &option))
-				return 0;
-			if (option)
-				nilfs_set_opt(sbi, BARRIER);
-			else
-				nilfs_clear_opt(sbi, BARRIER);
+		case Opt_nobarrier:
+			nilfs_clear_opt(sbi, BARRIER);
 			break;
 		case Opt_order:
 			if (strcmp(args[0].from, "relaxed") == 0)
@@ -647,6 +611,9 @@ static int parse_options(char *options, struct super_block *sb)
 			sbi->s_snapshot_cno = option;
 			nilfs_set_opt(sbi, SNAPSHOT);
 			break;
+		case Opt_norecovery:
+			nilfs_set_opt(sbi, NORECOVERY);
+			break;
 		default:
 			printk(KERN_ERR
 			       "NILFS: Unrecognized mount option \"%s\"\n", p);
@@ -672,9 +639,7 @@ static int nilfs_setup_super(struct nilfs_sb_info *sbi)
 	int mnt_count = le16_to_cpu(sbp->s_mnt_count);
 
 	/* nilfs->sem must be locked by the caller. */
-	if (!(nilfs->ns_mount_state & NILFS_VALID_FS)) {
-		printk(KERN_WARNING "NILFS warning: mounting unchecked fs\n");
-	} else if (nilfs->ns_mount_state & NILFS_ERROR_FS) {
+	if (nilfs->ns_mount_state & NILFS_ERROR_FS) {
 		printk(KERN_WARNING
 		       "NILFS warning: mounting fs with errors\n");
 #if 0
@@ -782,11 +747,10 @@ nilfs_fill_super(struct super_block *sb, void *data, int silent,
 	sb->s_root = NULL;
 	sb->s_time_gran = 1;
 
-	if (!nilfs_loaded(nilfs)) {
-		err = load_nilfs(nilfs, sbi);
-		if (err)
-			goto failed_sbi;
-	}
+	err = load_nilfs(nilfs, sbi);
+	if (err)
+		goto failed_sbi;
+
 	cno = nilfs_last_cno(nilfs);
 
 	if (sb->s_flags & MS_RDONLY) {
@@ -854,12 +818,6 @@ nilfs_fill_super(struct super_block *sb, void *data, int silent,
 		up_write(&nilfs->ns_sem);
 	}
 
-	err = nilfs_mark_recovery_complete(sbi);
-	if (unlikely(err)) {
-		printk(KERN_ERR "NILFS: recovery failed.\n");
-		goto failed_root;
-	}
-
 	down_write(&nilfs->ns_super_sem);
 	if (!nilfs_test_opt(sbi, SNAPSHOT))
 		nilfs->ns_current = sbi;
@@ -867,10 +825,6 @@ nilfs_fill_super(struct super_block *sb, void *data, int silent,
 
 	return 0;
 
- failed_root:
-	dput(sb->s_root);
-	sb->s_root = NULL;
-
  failed_segctor:
 	nilfs_detach_segment_constructor(sbi);
 
@@ -915,6 +869,14 @@ static int nilfs_remount(struct super_block *sb, int *flags, char *data)
 		goto restore_opts;
 	}
 
+	if (!nilfs_valid_fs(nilfs)) {
+		printk(KERN_WARNING "NILFS (device %s): couldn't "
+		       "remount because the filesystem is in an "
+		       "incomplete recovery state.\n", sb->s_id);
+		err = -EINVAL;
+		goto restore_opts;
+	}
+
 	if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY))
 		goto out;
 	if (*flags & MS_RDONLY) {
@@ -1156,8 +1118,7 @@ nilfs_get_sb(struct file_system_type *fs_type, int flags,
 	/* Abandoning the newly allocated superblock */
 	mutex_unlock(&nilfs->ns_mount_mutex);
 	put_nilfs(nilfs);
-	up_write(&s->s_umount);
-	deactivate_super(s);
+	deactivate_locked_super(s);
 	/*
 	 * deactivate_super() invokes close_bdev_exclusive().
 	 * We must finish all post-cleaning before this call;
diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c
index ad391a8c3e7e..6241e1722efc 100644
--- a/fs/nilfs2/the_nilfs.c
+++ b/fs/nilfs2/the_nilfs.c
@@ -146,13 +146,9 @@ void put_nilfs(struct the_nilfs *nilfs)
 
 	might_sleep();
 	if (nilfs_loaded(nilfs)) {
-		nilfs_mdt_clear(nilfs->ns_sufile);
 		nilfs_mdt_destroy(nilfs->ns_sufile);
-		nilfs_mdt_clear(nilfs->ns_cpfile);
 		nilfs_mdt_destroy(nilfs->ns_cpfile);
-		nilfs_mdt_clear(nilfs->ns_dat);
 		nilfs_mdt_destroy(nilfs->ns_dat);
-		/* XXX: how and when to clear nilfs->ns_gc_dat? */
 		nilfs_mdt_destroy(nilfs->ns_gc_dat);
 	}
 	if (nilfs_init(nilfs)) {
@@ -166,7 +162,6 @@ void put_nilfs(struct the_nilfs *nilfs)
 static int nilfs_load_super_root(struct the_nilfs *nilfs,
 				 struct nilfs_sb_info *sbi, sector_t sr_block)
 {
-	static struct lock_class_key dat_lock_key;
 	struct buffer_head *bh_sr;
 	struct nilfs_super_root *raw_sr;
 	struct nilfs_super_block **sbp = nilfs->ns_sbp;
@@ -187,51 +182,36 @@ static int nilfs_load_super_root(struct the_nilfs *nilfs,
 	inode_size = nilfs->ns_inode_size;
 
 	err = -ENOMEM;
-	nilfs->ns_dat = nilfs_mdt_new(nilfs, NULL, NILFS_DAT_INO);
+	nilfs->ns_dat = nilfs_dat_new(nilfs, dat_entry_size);
 	if (unlikely(!nilfs->ns_dat))
 		goto failed;
 
-	nilfs->ns_gc_dat = nilfs_mdt_new(nilfs, NULL, NILFS_DAT_INO);
+	nilfs->ns_gc_dat = nilfs_dat_new(nilfs, dat_entry_size);
 	if (unlikely(!nilfs->ns_gc_dat))
 		goto failed_dat;
 
-	nilfs->ns_cpfile = nilfs_mdt_new(nilfs, NULL, NILFS_CPFILE_INO);
+	nilfs->ns_cpfile = nilfs_cpfile_new(nilfs, checkpoint_size);
 	if (unlikely(!nilfs->ns_cpfile))
 		goto failed_gc_dat;
 
-	nilfs->ns_sufile = nilfs_mdt_new(nilfs, NULL, NILFS_SUFILE_INO);
+	nilfs->ns_sufile = nilfs_sufile_new(nilfs, segment_usage_size);
 	if (unlikely(!nilfs->ns_sufile))
 		goto failed_cpfile;
 
-	err = nilfs_palloc_init_blockgroup(nilfs->ns_dat, dat_entry_size);
-	if (unlikely(err))
-		goto failed_sufile;
-
-	err = nilfs_palloc_init_blockgroup(nilfs->ns_gc_dat, dat_entry_size);
-	if (unlikely(err))
-		goto failed_sufile;
-
-	lockdep_set_class(&NILFS_MDT(nilfs->ns_dat)->mi_sem, &dat_lock_key);
-	lockdep_set_class(&NILFS_MDT(nilfs->ns_gc_dat)->mi_sem, &dat_lock_key);
-
 	nilfs_mdt_set_shadow(nilfs->ns_dat, nilfs->ns_gc_dat);
-	nilfs_mdt_set_entry_size(nilfs->ns_cpfile, checkpoint_size,
-				 sizeof(struct nilfs_cpfile_header));
-	nilfs_mdt_set_entry_size(nilfs->ns_sufile, segment_usage_size,
-				 sizeof(struct nilfs_sufile_header));
 
-	err = nilfs_mdt_read_inode_direct(
-		nilfs->ns_dat, bh_sr, NILFS_SR_DAT_OFFSET(inode_size));
+	err = nilfs_dat_read(nilfs->ns_dat, (void *)bh_sr->b_data +
+			     NILFS_SR_DAT_OFFSET(inode_size));
 	if (unlikely(err))
 		goto failed_sufile;
 
-	err = nilfs_mdt_read_inode_direct(
-		nilfs->ns_cpfile, bh_sr, NILFS_SR_CPFILE_OFFSET(inode_size));
+	err = nilfs_cpfile_read(nilfs->ns_cpfile, (void *)bh_sr->b_data +
+				NILFS_SR_CPFILE_OFFSET(inode_size));
 	if (unlikely(err))
 		goto failed_sufile;
 
-	err = nilfs_mdt_read_inode_direct(
-		nilfs->ns_sufile, bh_sr, NILFS_SR_SUFILE_OFFSET(inode_size));
+	err = nilfs_sufile_read(nilfs->ns_sufile, (void *)bh_sr->b_data +
+				NILFS_SR_SUFILE_OFFSET(inode_size));
 	if (unlikely(err))
 		goto failed_sufile;
 
@@ -281,29 +261,30 @@ int load_nilfs(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi)
 	struct nilfs_recovery_info ri;
 	unsigned int s_flags = sbi->s_super->s_flags;
 	int really_read_only = bdev_read_only(nilfs->ns_bdev);
-	unsigned valid_fs;
-	int err = 0;
-
-	nilfs_init_recovery_info(&ri);
+	int valid_fs = nilfs_valid_fs(nilfs);
+	int err;
 
-	down_write(&nilfs->ns_sem);
-	valid_fs = (nilfs->ns_mount_state & NILFS_VALID_FS);
-	up_write(&nilfs->ns_sem);
+	if (nilfs_loaded(nilfs)) {
+		if (valid_fs ||
+		    ((s_flags & MS_RDONLY) && nilfs_test_opt(sbi, NORECOVERY)))
+			return 0;
+		printk(KERN_ERR "NILFS: the filesystem is in an incomplete "
+		       "recovery state.\n");
+		return -EINVAL;
+	}
 
-	if (!valid_fs && (s_flags & MS_RDONLY)) {
-		printk(KERN_INFO "NILFS: INFO: recovery "
-		       "required for readonly filesystem.\n");
-		if (really_read_only) {
-			printk(KERN_ERR "NILFS: write access "
-			       "unavailable, cannot proceed.\n");
-			err = -EROFS;
-			goto failed;
+	if (!valid_fs) {
+		printk(KERN_WARNING "NILFS warning: mounting unchecked fs\n");
+		if (s_flags & MS_RDONLY) {
+			printk(KERN_INFO "NILFS: INFO: recovery "
+			       "required for readonly filesystem.\n");
+			printk(KERN_INFO "NILFS: write access will "
+			       "be enabled during recovery.\n");
 		}
-		printk(KERN_INFO "NILFS: write access will "
-		       "be enabled during recovery.\n");
-		sbi->s_super->s_flags &= ~MS_RDONLY;
 	}
 
+	nilfs_init_recovery_info(&ri);
+
 	err = nilfs_search_super_root(nilfs, sbi, &ri);
 	if (unlikely(err)) {
 		printk(KERN_ERR "NILFS: error searching super root.\n");
@@ -316,19 +297,56 @@ int load_nilfs(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi)
 		goto failed;
 	}
 
-	if (!valid_fs) {
-		err = nilfs_recover_logical_segments(nilfs, sbi, &ri);
-		if (unlikely(err)) {
-			nilfs_mdt_destroy(nilfs->ns_cpfile);
-			nilfs_mdt_destroy(nilfs->ns_sufile);
-			nilfs_mdt_destroy(nilfs->ns_dat);
-			goto failed;
+	if (valid_fs)
+		goto skip_recovery;
+
+	if (s_flags & MS_RDONLY) {
+		if (nilfs_test_opt(sbi, NORECOVERY)) {
+			printk(KERN_INFO "NILFS: norecovery option specified. "
+			       "skipping roll-forward recovery\n");
+			goto skip_recovery;
 		}
-		if (ri.ri_need_recovery == NILFS_RECOVERY_SR_UPDATED)
-			sbi->s_super->s_dirt = 1;
+		if (really_read_only) {
+			printk(KERN_ERR "NILFS: write access "
+			       "unavailable, cannot proceed.\n");
+			err = -EROFS;
+			goto failed_unload;
+		}
+		sbi->s_super->s_flags &= ~MS_RDONLY;
+	} else if (nilfs_test_opt(sbi, NORECOVERY)) {
+		printk(KERN_ERR "NILFS: recovery cancelled because norecovery "
+		       "option was specified for a read/write mount\n");
+		err = -EINVAL;
+		goto failed_unload;
 	}
 
+	err = nilfs_recover_logical_segments(nilfs, sbi, &ri);
+	if (err)
+		goto failed_unload;
+
+	down_write(&nilfs->ns_sem);
+	nilfs->ns_mount_state |= NILFS_VALID_FS;
+	nilfs->ns_sbp[0]->s_state = cpu_to_le16(nilfs->ns_mount_state);
+	err = nilfs_commit_super(sbi, 1);
+	up_write(&nilfs->ns_sem);
+
+	if (err) {
+		printk(KERN_ERR "NILFS: failed to update super block. "
+		       "recovery unfinished.\n");
+		goto failed_unload;
+	}
+	printk(KERN_INFO "NILFS: recovery complete.\n");
+
+ skip_recovery:
 	set_nilfs_loaded(nilfs);
+	nilfs_clear_recovery_info(&ri);
+	sbi->s_super->s_flags = s_flags;
+	return 0;
+
+ failed_unload:
+	nilfs_mdt_destroy(nilfs->ns_cpfile);
+	nilfs_mdt_destroy(nilfs->ns_sufile);
+	nilfs_mdt_destroy(nilfs->ns_dat);
 
  failed:
 	nilfs_clear_recovery_info(&ri);
@@ -632,30 +650,23 @@ int nilfs_count_free_blocks(struct the_nilfs *nilfs, sector_t *nblocks)
 {
 	struct inode *dat = nilfs_dat_inode(nilfs);
 	unsigned long ncleansegs;
-	int err;
 
 	down_read(&NILFS_MDT(dat)->mi_sem);	/* XXX */
-	err = nilfs_sufile_get_ncleansegs(nilfs->ns_sufile, &ncleansegs);
+	ncleansegs = nilfs_sufile_get_ncleansegs(nilfs->ns_sufile);
 	up_read(&NILFS_MDT(dat)->mi_sem);	/* XXX */
-	if (likely(!err))
-		*nblocks = (sector_t)ncleansegs * nilfs->ns_blocks_per_segment;
-	return err;
+	*nblocks = (sector_t)ncleansegs * nilfs->ns_blocks_per_segment;
+	return 0;
 }
 
 int nilfs_near_disk_full(struct the_nilfs *nilfs)
 {
-	struct inode *sufile = nilfs->ns_sufile;
 	unsigned long ncleansegs, nincsegs;
-	int ret;
 
-	ret = nilfs_sufile_get_ncleansegs(sufile, &ncleansegs);
-	if (likely(!ret)) {
-		nincsegs = atomic_read(&nilfs->ns_ndirtyblks) /
-			nilfs->ns_blocks_per_segment + 1;
-		if (ncleansegs <= nilfs->ns_nrsvsegs + nincsegs)
-			ret++;
-	}
-	return ret;
+	ncleansegs = nilfs_sufile_get_ncleansegs(nilfs->ns_sufile);
+	nincsegs = atomic_read(&nilfs->ns_ndirtyblks) /
+		nilfs->ns_blocks_per_segment + 1;
+
+	return ncleansegs <= nilfs->ns_nrsvsegs + nincsegs;
 }
 
 /**
diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h
index 20abd55881e0..589786e33464 100644
--- a/fs/nilfs2/the_nilfs.h
+++ b/fs/nilfs2/the_nilfs.h
@@ -258,6 +258,16 @@ static inline void nilfs_put_sbinfo(struct nilfs_sb_info *sbi)
 		kfree(sbi);
 }
 
+static inline int nilfs_valid_fs(struct the_nilfs *nilfs)
+{
+	unsigned valid_fs;
+
+	down_read(&nilfs->ns_sem);
+	valid_fs = (nilfs->ns_mount_state & NILFS_VALID_FS);
+	up_read(&nilfs->ns_sem);
+	return valid_fs;
+}
+
 static inline void
 nilfs_get_segment_range(struct the_nilfs *nilfs, __u64 segnum,
 			sector_t *seg_start, sector_t *seg_end)
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
index c9ee67b442e1..1afb0a10229f 100644
--- a/fs/notify/inotify/inotify_fsnotify.c
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -121,7 +121,7 @@ static int idr_callback(int id, void *p, void *data)
 	if (warned)
 		return 0;
 
-	warned = false;
+	warned = true;
 	entry = p;
 	ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
 
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 5ef5f365a5c8..a94e8bd8eb1f 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -552,7 +552,7 @@ retry:
 
 	spin_lock(&group->inotify_data.idr_lock);
 	ret = idr_get_new_above(&group->inotify_data.idr, &tmp_ientry->fsn_entry,
-				group->inotify_data.last_wd,
+				group->inotify_data.last_wd+1,
 				&tmp_ientry->wd);
 	spin_unlock(&group->inotify_data.idr_lock);
 	if (ret) {
@@ -632,7 +632,7 @@ static struct fsnotify_group *inotify_new_group(struct user_struct *user, unsign
 
 	spin_lock_init(&group->inotify_data.idr_lock);
 	idr_init(&group->inotify_data.idr);
-	group->inotify_data.last_wd = 1;
+	group->inotify_data.last_wd = 0;
 	group->inotify_data.user = user;
 	group->inotify_data.fa = NULL;
 
@@ -646,6 +646,7 @@ SYSCALL_DEFINE1(inotify_init1, int, flags)
 	struct fsnotify_group *group;
 	struct user_struct *user;
 	struct file *filp;
+	struct path path;
 	int fd, ret;
 
 	/* Check the IN_* constants for consistency.  */
@@ -659,12 +660,6 @@ SYSCALL_DEFINE1(inotify_init1, int, flags)
 	if (fd < 0)
 		return fd;
 
-	filp = get_empty_filp();
-	if (!filp) {
-		ret = -ENFILE;
-		goto out_put_fd;
-	}
-
 	user = get_current_user();
 	if (unlikely(atomic_read(&user->inotify_devs) >=
 			inotify_max_user_instances)) {
@@ -679,24 +674,28 @@ SYSCALL_DEFINE1(inotify_init1, int, flags)
 		goto out_free_uid;
 	}
 
-	filp->f_op = &inotify_fops;
-	filp->f_path.mnt = mntget(inotify_mnt);
-	filp->f_path.dentry = dget(inotify_mnt->mnt_root);
-	filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping;
-	filp->f_mode = FMODE_READ;
+	atomic_inc(&user->inotify_devs);
+
+	path.mnt = inotify_mnt;
+	path.dentry = inotify_mnt->mnt_root;
+	path_get(&path);
+	filp = alloc_file(&path, FMODE_READ, &inotify_fops);
+	if (!filp)
+		goto Enfile;
+
 	filp->f_flags = O_RDONLY | (flags & O_NONBLOCK);
 	filp->private_data = group;
 
-	atomic_inc(&user->inotify_devs);
-
 	fd_install(fd, filp);
 
 	return fd;
 
+Enfile:
+	ret = -ENFILE;
+	path_put(&path);
+	atomic_dec(&user->inotify_devs);
 out_free_uid:
 	free_uid(user);
-	put_filp(filp);
-out_put_fd:
 	put_unused_fd(fd);
 	return ret;
 }
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 9938034762cc..dc2505abb6d7 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -530,7 +530,7 @@ err_corrupt_attr:
  * the ntfs inode.
  *
  * Q: What locks are held when the function is called?
- * A: i_state has I_LOCK set, hence the inode is locked, also
+ * A: i_state has I_NEW set, hence the inode is locked, also
  *    i_count is set to 1, so it is not going to go away
  *    i_flags is set to 0 and we have no business touching it.  Only an ioctl()
  *    is allowed to write to them. We should of course be honouring them but
@@ -1207,7 +1207,7 @@ err_out:
  * necessary fields in @vi as well as initializing the ntfs inode.
  *
  * Q: What locks are held when the function is called?
- * A: i_state has I_LOCK set, hence the inode is locked, also
+ * A: i_state has I_NEW set, hence the inode is locked, also
  *    i_count is set to 1, so it is not going to go away
  *
  * Return 0 on success and -errno on error.  In the error case, the inode will
@@ -1474,7 +1474,7 @@ err_out:
  * normal directory inodes.
  *
  * Q: What locks are held when the function is called?
- * A: i_state has I_LOCK set, hence the inode is locked, also
+ * A: i_state has I_NEW set, hence the inode is locked, also
  *    i_count is set to 1, so it is not going to go away
  *
  * Return 0 on success and -errno on error.  In the error case, the inode will
diff --git a/fs/ocfs2/Kconfig b/fs/ocfs2/Kconfig
index 701b7a3a872e..0d840669698e 100644
--- a/fs/ocfs2/Kconfig
+++ b/fs/ocfs2/Kconfig
@@ -6,6 +6,7 @@ config OCFS2_FS
 	select CRC32
 	select QUOTA
 	select QUOTA_TREE
+	select FS_POSIX_ACL
 	help
 	  OCFS2 is a general purpose extent based shared disk cluster file
 	  system with many similarities to ext3. It supports 64 bit inode
@@ -74,12 +75,3 @@ config OCFS2_DEBUG_FS
 	  This option will enable expensive consistency checks. Enable
 	  this option for debugging only as it is likely to decrease
 	  performance of the filesystem.
-
-config OCFS2_FS_POSIX_ACL
-	bool "OCFS2 POSIX Access Control Lists"
-	depends on OCFS2_FS
-	select FS_POSIX_ACL
-	default n
-	help
-	  Posix Access Control Lists (ACLs) support permissions for users and
-	  groups beyond the owner/group/world scheme.
diff --git a/fs/ocfs2/Makefile b/fs/ocfs2/Makefile
index 31f25ce32c97..600d2d2ade11 100644
--- a/fs/ocfs2/Makefile
+++ b/fs/ocfs2/Makefile
@@ -39,11 +39,8 @@ ocfs2-objs := \
 	ver.o			\
 	quota_local.o		\
 	quota_global.o		\
-	xattr.o
-
-ifeq ($(CONFIG_OCFS2_FS_POSIX_ACL),y)
-ocfs2-objs += acl.o
-endif
+	xattr.o			\
+	acl.o
 
 ocfs2_stackglue-objs := stackglue.o
 ocfs2_stack_o2cb-objs := stack_o2cb.o
diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c
index fbeaec762103..0501974bedd0 100644
--- a/fs/ocfs2/acl.c
+++ b/fs/ocfs2/acl.c
@@ -98,15 +98,11 @@ static struct posix_acl *ocfs2_get_acl_nolock(struct inode *inode,
 					      int type,
 					      struct buffer_head *di_bh)
 {
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	int name_index;
 	char *value = NULL;
 	struct posix_acl *acl;
 	int retval;
 
-	if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
-		return NULL;
-
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name_index = OCFS2_XATTR_INDEX_POSIX_ACL_ACCESS;
@@ -331,13 +327,14 @@ cleanup:
 	return ret;
 }
 
-static size_t ocfs2_xattr_list_acl_access(struct inode *inode,
+static size_t ocfs2_xattr_list_acl_access(struct dentry *dentry,
 					  char *list,
 					  size_t list_len,
 					  const char *name,
-					  size_t name_len)
+					  size_t name_len,
+					  int type)
 {
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
 	const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS);
 
 	if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
@@ -348,13 +345,14 @@ static size_t ocfs2_xattr_list_acl_access(struct inode *inode,
 	return size;
 }
 
-static size_t ocfs2_xattr_list_acl_default(struct inode *inode,
+static size_t ocfs2_xattr_list_acl_default(struct dentry *dentry,
 					   char *list,
 					   size_t list_len,
 					   const char *name,
-					   size_t name_len)
+					   size_t name_len,
+					   int type)
 {
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
 	const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT);
 
 	if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
@@ -365,19 +363,19 @@ static size_t ocfs2_xattr_list_acl_default(struct inode *inode,
 	return size;
 }
 
-static int ocfs2_xattr_get_acl(struct inode *inode,
-			       int type,
-			       void *buffer,
-			       size_t size)
+static int ocfs2_xattr_get_acl(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
 	struct posix_acl *acl;
 	int ret;
 
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
 	if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
 		return -EOPNOTSUPP;
 
-	acl = ocfs2_get_acl(inode, type);
+	acl = ocfs2_get_acl(dentry->d_inode, type);
 	if (IS_ERR(acl))
 		return PTR_ERR(acl);
 	if (acl == NULL)
@@ -388,35 +386,16 @@ static int ocfs2_xattr_get_acl(struct inode *inode,
 	return ret;
 }
 
-static int ocfs2_xattr_get_acl_access(struct inode *inode,
-				      const char *name,
-				      void *buffer,
-				      size_t size)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ocfs2_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size);
-}
-
-static int ocfs2_xattr_get_acl_default(struct inode *inode,
-				       const char *name,
-				       void *buffer,
-				       size_t size)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ocfs2_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size);
-}
-
-static int ocfs2_xattr_set_acl(struct inode *inode,
-			       int type,
-			       const void *value,
-			       size_t size)
+static int ocfs2_xattr_set_acl(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
+	struct inode *inode = dentry->d_inode;
 	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 	struct posix_acl *acl;
 	int ret = 0;
 
+	if (strcmp(name, "") != 0)
+		return -EINVAL;
 	if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL))
 		return -EOPNOTSUPP;
 
@@ -442,38 +421,18 @@ cleanup:
 	return ret;
 }
 
-static int ocfs2_xattr_set_acl_access(struct inode *inode,
-				      const char *name,
-				      const void *value,
-				      size_t size,
-				      int flags)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ocfs2_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size);
-}
-
-static int ocfs2_xattr_set_acl_default(struct inode *inode,
-				       const char *name,
-				       const void *value,
-				       size_t size,
-				       int flags)
-{
-	if (strcmp(name, "") != 0)
-		return -EINVAL;
-	return ocfs2_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size);
-}
-
 struct xattr_handler ocfs2_xattr_acl_access_handler = {
 	.prefix	= POSIX_ACL_XATTR_ACCESS,
+	.flags	= ACL_TYPE_ACCESS,
 	.list	= ocfs2_xattr_list_acl_access,
-	.get	= ocfs2_xattr_get_acl_access,
-	.set	= ocfs2_xattr_set_acl_access,
+	.get	= ocfs2_xattr_get_acl,
+	.set	= ocfs2_xattr_set_acl,
 };
 
 struct xattr_handler ocfs2_xattr_acl_default_handler = {
 	.prefix	= POSIX_ACL_XATTR_DEFAULT,
+	.flags	= ACL_TYPE_DEFAULT,
 	.list	= ocfs2_xattr_list_acl_default,
-	.get	= ocfs2_xattr_get_acl_default,
-	.set	= ocfs2_xattr_set_acl_default,
+	.get	= ocfs2_xattr_get_acl,
+	.set	= ocfs2_xattr_set_acl,
 };
diff --git a/fs/ocfs2/acl.h b/fs/ocfs2/acl.h
index 8f6389ed4da5..5c5d31f05853 100644
--- a/fs/ocfs2/acl.h
+++ b/fs/ocfs2/acl.h
@@ -26,8 +26,6 @@ struct ocfs2_acl_entry {
 	__le32 e_id;
 };
 
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
-
 extern int ocfs2_check_acl(struct inode *, int);
 extern int ocfs2_acl_chmod(struct inode *);
 extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *,
@@ -35,24 +33,4 @@ extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *,
 			  struct ocfs2_alloc_context *,
 			  struct ocfs2_alloc_context *);
 
-#else /* CONFIG_OCFS2_FS_POSIX_ACL*/
-
-#define ocfs2_check_acl NULL
-static inline int ocfs2_acl_chmod(struct inode *inode)
-{
-	return 0;
-}
-static inline int ocfs2_init_acl(handle_t *handle,
-				 struct inode *inode,
-				 struct inode *dir,
-				 struct buffer_head *di_bh,
-				 struct buffer_head *dir_bh,
-				 struct ocfs2_alloc_context *meta_ac,
-				 struct ocfs2_alloc_context *data_ac)
-{
-	return 0;
-}
-
-#endif /* CONFIG_OCFS2_FS_POSIX_ACL*/
-
 #endif /* OCFS2_ACL_H */
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 7c7198a5bc90..d17bdc718f74 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -1765,9 +1765,9 @@ set_and_inc:
  *
  * The array index of the subtree root is passed back.
  */
-static int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
-				   struct ocfs2_path *left,
-				   struct ocfs2_path *right)
+int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
+			    struct ocfs2_path *left,
+			    struct ocfs2_path *right)
 {
 	int i = 0;
 
@@ -2872,8 +2872,8 @@ out:
  * This looks similar, but is subtly different to
  * ocfs2_find_cpos_for_left_leaf().
  */
-static int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
-					  struct ocfs2_path *path, u32 *cpos)
+int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
+				   struct ocfs2_path *path, u32 *cpos)
 {
 	int i, j, ret = 0;
 	u64 blkno;
@@ -7190,8 +7190,8 @@ int ocfs2_zero_range_for_truncate(struct inode *inode, handle_t *handle,
 	 * wait on them - the truncate_inode_pages() call later will
 	 * do that for us.
 	 */
-	ret = do_sync_mapping_range(inode->i_mapping, range_start,
-				    range_end - 1, SYNC_FILE_RANGE_WRITE);
+	ret = filemap_fdatawrite_range(inode->i_mapping, range_start,
+				       range_end - 1);
 	if (ret)
 		mlog_errno(ret);
 
diff --git a/fs/ocfs2/alloc.h b/fs/ocfs2/alloc.h
index 9c122d574464..1db4359ccb90 100644
--- a/fs/ocfs2/alloc.h
+++ b/fs/ocfs2/alloc.h
@@ -317,4 +317,9 @@ int ocfs2_path_bh_journal_access(handle_t *handle,
 int ocfs2_journal_access_path(struct ocfs2_caching_info *ci,
 			      handle_t *handle,
 			      struct ocfs2_path *path);
+int ocfs2_find_cpos_for_right_leaf(struct super_block *sb,
+				   struct ocfs2_path *path, u32 *cpos);
+int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et,
+			    struct ocfs2_path *left,
+			    struct ocfs2_path *right);
 #endif /* OCFS2_ALLOC_H */
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index deb2b132ae5e..3dae4a13f6e4 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -547,6 +547,9 @@ bail:
  *
  * called like this: dio->get_blocks(dio->inode, fs_startblk,
  * 					fs_count, map_bh, dio->rw == WRITE);
+ *
+ * Note that we never bother to allocate blocks here, and thus ignore the
+ * create argument.
  */
 static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
 				     struct buffer_head *bh_result, int create)
@@ -563,14 +566,6 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
 
 	inode_blocks = ocfs2_blocks_for_bytes(inode->i_sb, i_size_read(inode));
 
-	/*
-	 * Any write past EOF is not allowed because we'd be extending.
-	 */
-	if (create && (iblock + max_blocks) > inode_blocks) {
-		ret = -EIO;
-		goto bail;
-	}
-
 	/* This figures out the size of the next contiguous block, and
 	 * our logical offset */
 	ret = ocfs2_extent_map_get_blocks(inode, iblock, &p_blkno,
@@ -582,15 +577,6 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
 		goto bail;
 	}
 
-	if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)) && !p_blkno && create) {
-		ocfs2_error(inode->i_sb,
-			    "Inode %llu has a hole at block %llu\n",
-			    (unsigned long long)OCFS2_I(inode)->ip_blkno,
-			    (unsigned long long)iblock);
-		ret = -EROFS;
-		goto bail;
-	}
-
 	/* We should already CoW the refcounted extent. */
 	BUG_ON(ext_flags & OCFS2_EXT_REFCOUNTED);
 	/*
@@ -601,20 +587,8 @@ static int ocfs2_direct_IO_get_blocks(struct inode *inode, sector_t iblock,
 	 */
 	if (p_blkno && !(ext_flags & OCFS2_EXT_UNWRITTEN))
 		map_bh(bh_result, inode->i_sb, p_blkno);
-	else {
-		/*
-		 * ocfs2_prepare_inode_for_write() should have caught
-		 * the case where we'd be filling a hole and triggered
-		 * a buffered write instead.
-		 */
-		if (create) {
-			ret = -EIO;
-			mlog_errno(ret);
-			goto bail;
-		}
-
+	else
 		clear_buffer_mapped(bh_result);
-	}
 
 	/* make sure we don't map more than max_blocks blocks here as
 	   that's all the kernel will handle at this point. */
diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c
index c452d116b892..eda5b8bcddd5 100644
--- a/fs/ocfs2/cluster/heartbeat.c
+++ b/fs/ocfs2/cluster/heartbeat.c
@@ -176,7 +176,8 @@ static void o2hb_write_timeout(struct work_struct *work)
 
 static void o2hb_arm_write_timeout(struct o2hb_region *reg)
 {
-	mlog(0, "Queue write timeout for %u ms\n", O2HB_MAX_WRITE_TIMEOUT_MS);
+	mlog(ML_HEARTBEAT, "Queue write timeout for %u ms\n",
+	     O2HB_MAX_WRITE_TIMEOUT_MS);
 
 	cancel_delayed_work(&reg->hr_write_timeout_work);
 	reg->hr_last_timeout_start = jiffies;
@@ -874,7 +875,8 @@ static int o2hb_thread(void *data)
 		do_gettimeofday(&after_hb);
 		elapsed_msec = o2hb_elapsed_msecs(&before_hb, &after_hb);
 
-		mlog(0, "start = %lu.%lu, end = %lu.%lu, msec = %u\n",
+		mlog(ML_HEARTBEAT,
+		     "start = %lu.%lu, end = %lu.%lu, msec = %u\n",
 		     before_hb.tv_sec, (unsigned long) before_hb.tv_usec,
 		     after_hb.tv_sec, (unsigned long) after_hb.tv_usec,
 		     elapsed_msec);
diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c
index 7ee6188bc79a..c81142e3ef84 100644
--- a/fs/ocfs2/cluster/nodemanager.c
+++ b/fs/ocfs2/cluster/nodemanager.c
@@ -35,6 +35,10 @@
  * cluster references throughout where nodes are looked up */
 struct o2nm_cluster *o2nm_single_cluster = NULL;
 
+char *o2nm_fence_method_desc[O2NM_FENCE_METHODS] = {
+		"reset",	/* O2NM_FENCE_RESET */
+		"panic",	/* O2NM_FENCE_PANIC */
+};
 
 struct o2nm_node *o2nm_get_node_by_num(u8 node_num)
 {
@@ -579,6 +583,43 @@ static ssize_t o2nm_cluster_attr_reconnect_delay_ms_write(
 	return o2nm_cluster_attr_write(page, count,
 	                               &cluster->cl_reconnect_delay_ms);
 }
+
+static ssize_t o2nm_cluster_attr_fence_method_read(
+	struct o2nm_cluster *cluster, char *page)
+{
+	ssize_t ret = 0;
+
+	if (cluster)
+		ret = sprintf(page, "%s\n",
+			      o2nm_fence_method_desc[cluster->cl_fence_method]);
+	return ret;
+}
+
+static ssize_t o2nm_cluster_attr_fence_method_write(
+	struct o2nm_cluster *cluster, const char *page, size_t count)
+{
+	unsigned int i;
+
+	if (page[count - 1] != '\n')
+		goto bail;
+
+	for (i = 0; i < O2NM_FENCE_METHODS; ++i) {
+		if (count != strlen(o2nm_fence_method_desc[i]) + 1)
+			continue;
+		if (strncasecmp(page, o2nm_fence_method_desc[i], count - 1))
+			continue;
+		if (cluster->cl_fence_method != i) {
+			printk(KERN_INFO "ocfs2: Changing fence method to %s\n",
+			       o2nm_fence_method_desc[i]);
+			cluster->cl_fence_method = i;
+		}
+		return count;
+	}
+
+bail:
+	return -EINVAL;
+}
+
 static struct o2nm_cluster_attribute o2nm_cluster_attr_idle_timeout_ms = {
 	.attr	= { .ca_owner = THIS_MODULE,
 		    .ca_name = "idle_timeout_ms",
@@ -603,10 +644,19 @@ static struct o2nm_cluster_attribute o2nm_cluster_attr_reconnect_delay_ms = {
 	.store	= o2nm_cluster_attr_reconnect_delay_ms_write,
 };
 
+static struct o2nm_cluster_attribute o2nm_cluster_attr_fence_method = {
+	.attr	= { .ca_owner = THIS_MODULE,
+		    .ca_name = "fence_method",
+		    .ca_mode = S_IRUGO | S_IWUSR },
+	.show	= o2nm_cluster_attr_fence_method_read,
+	.store	= o2nm_cluster_attr_fence_method_write,
+};
+
 static struct configfs_attribute *o2nm_cluster_attrs[] = {
 	&o2nm_cluster_attr_idle_timeout_ms.attr,
 	&o2nm_cluster_attr_keepalive_delay_ms.attr,
 	&o2nm_cluster_attr_reconnect_delay_ms.attr,
+	&o2nm_cluster_attr_fence_method.attr,
 	NULL,
 };
 static ssize_t o2nm_cluster_show(struct config_item *item,
@@ -778,6 +828,7 @@ static struct config_group *o2nm_cluster_group_make_group(struct config_group *g
 	cluster->cl_reconnect_delay_ms = O2NET_RECONNECT_DELAY_MS_DEFAULT;
 	cluster->cl_idle_timeout_ms    = O2NET_IDLE_TIMEOUT_MS_DEFAULT;
 	cluster->cl_keepalive_delay_ms = O2NET_KEEPALIVE_DELAY_MS_DEFAULT;
+	cluster->cl_fence_method       = O2NM_FENCE_RESET;
 
 	ret = &cluster->cl_group;
 	o2nm_single_cluster = cluster;
diff --git a/fs/ocfs2/cluster/nodemanager.h b/fs/ocfs2/cluster/nodemanager.h
index c992ea0da4ad..09ea2d388bbb 100644
--- a/fs/ocfs2/cluster/nodemanager.h
+++ b/fs/ocfs2/cluster/nodemanager.h
@@ -33,6 +33,12 @@
 #include <linux/configfs.h>
 #include <linux/rbtree.h>
 
+enum o2nm_fence_method {
+	O2NM_FENCE_RESET	= 0,
+	O2NM_FENCE_PANIC,
+	O2NM_FENCE_METHODS,	/* Number of fence methods */
+};
+
 struct o2nm_node {
 	spinlock_t		nd_lock;
 	struct config_item	nd_item;
@@ -58,6 +64,7 @@ struct o2nm_cluster {
 	unsigned int		cl_idle_timeout_ms;
 	unsigned int		cl_keepalive_delay_ms;
 	unsigned int		cl_reconnect_delay_ms;
+	enum o2nm_fence_method	cl_fence_method;
 
 	/* this bitmap is part of a hack for disk bitmap.. will go eventually. - zab */
 	unsigned long	cl_nodes_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)];
diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c
index bbacf7da48a4..639024033fce 100644
--- a/fs/ocfs2/cluster/quorum.c
+++ b/fs/ocfs2/cluster/quorum.c
@@ -74,8 +74,20 @@ static void o2quo_fence_self(void)
 	 * threads can still schedule, etc, etc */
 	o2hb_stop_all_regions();
 
-	printk("ocfs2 is very sorry to be fencing this system by restarting\n");
-	emergency_restart();
+	switch (o2nm_single_cluster->cl_fence_method) {
+	case O2NM_FENCE_PANIC:
+		panic("*** ocfs2 is very sorry to be fencing this system by "
+		      "panicing ***\n");
+		break;
+	default:
+		WARN_ON(o2nm_single_cluster->cl_fence_method >=
+			O2NM_FENCE_METHODS);
+	case O2NM_FENCE_RESET:
+		printk(KERN_ERR "*** ocfs2 is very sorry to be fencing this "
+		       "system by restarting ***\n");
+		emergency_restart();
+		break;
+	};
 }
 
 /* Indicate that a timeout occured on a hearbeat region write. The
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index d9fa3d22e17c..2f9e4e19a4f2 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -2589,6 +2589,14 @@ retry:
 			     "begin reco msg (%d)\n", dlm->name, nodenum, ret);
 			ret = 0;
 		}
+		if (ret == -EAGAIN) {
+			mlog(0, "%s: trying to start recovery of node "
+			     "%u, but node %u is waiting for last recovery "
+			     "to complete, backoff for a bit\n", dlm->name,
+			     dead_node, nodenum);
+			msleep(100);
+			goto retry;
+		}
 		if (ret < 0) {
 			struct dlm_lock_resource *res;
 			/* this is now a serious problem, possibly ENOMEM 
@@ -2608,14 +2616,6 @@ retry:
 			 * another ENOMEM */
 			msleep(100);
 			goto retry;
-		} else if (ret == EAGAIN) {
-			mlog(0, "%s: trying to start recovery of node "
-			     "%u, but node %u is waiting for last recovery "
-			     "to complete, backoff for a bit\n", dlm->name,
-			     dead_node, nodenum);
-			/* TODO Look into replacing msleep with cond_resched() */
-			msleep(100);
-			goto retry;
 		}
 	}
 
@@ -2639,7 +2639,7 @@ int dlm_begin_reco_handler(struct o2net_msg *msg, u32 len, void *data,
 		     dlm->name, br->node_idx, br->dead_node,
 		     dlm->reco.dead_node, dlm->reco.new_master);
 		spin_unlock(&dlm->spinlock);
-		return EAGAIN;
+		return -EAGAIN;
 	}
 	spin_unlock(&dlm->spinlock);
 
diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c
index 843db64e9d4a..d35a27f4523e 100644
--- a/fs/ocfs2/extent_map.c
+++ b/fs/ocfs2/extent_map.c
@@ -37,6 +37,7 @@
 #include "extent_map.h"
 #include "inode.h"
 #include "super.h"
+#include "symlink.h"
 
 #include "buffer_head_io.h"
 
@@ -703,6 +704,12 @@ out:
 	return ret;
 }
 
+/*
+ * The ocfs2_fiemap_inline() may be a little bit misleading, since
+ * it not only handles the fiemap for inlined files, but also deals
+ * with the fast symlink, cause they have no difference for extent
+ * mapping per se.
+ */
 static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh,
 			       struct fiemap_extent_info *fieinfo,
 			       u64 map_start)
@@ -715,11 +722,18 @@ static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh,
 	struct ocfs2_inode_info *oi = OCFS2_I(inode);
 
 	di = (struct ocfs2_dinode *)di_bh->b_data;
-	id_count = le16_to_cpu(di->id2.i_data.id_count);
+	if (ocfs2_inode_is_fast_symlink(inode))
+		id_count = ocfs2_fast_symlink_chars(inode->i_sb);
+	else
+		id_count = le16_to_cpu(di->id2.i_data.id_count);
 
 	if (map_start < id_count) {
 		phys = oi->ip_blkno << inode->i_sb->s_blocksize_bits;
-		phys += offsetof(struct ocfs2_dinode, id2.i_data.id_data);
+		if (ocfs2_inode_is_fast_symlink(inode))
+			phys += offsetof(struct ocfs2_dinode, id2.i_symlink);
+		else
+			phys += offsetof(struct ocfs2_dinode,
+					 id2.i_data.id_data);
 
 		ret = fiemap_fill_next_extent(fieinfo, 0, phys, id_count,
 					      flags);
@@ -756,9 +770,10 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	down_read(&OCFS2_I(inode)->ip_alloc_sem);
 
 	/*
-	 * Handle inline-data separately.
+	 * Handle inline-data and fast symlink separately.
 	 */
-	if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) {
+	if ((OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) ||
+	    ocfs2_inode_is_fast_symlink(inode)) {
 		ret = ocfs2_fiemap_inline(inode, di_bh, fieinfo, map_start);
 		goto out_unlock;
 	}
@@ -786,6 +801,8 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		fe_flags = 0;
 		if (rec.e_flags & OCFS2_EXT_UNWRITTEN)
 			fe_flags |= FIEMAP_EXTENT_UNWRITTEN;
+		if (rec.e_flags & OCFS2_EXT_REFCOUNTED)
+			fe_flags |= FIEMAP_EXTENT_SHARED;
 		if (is_last)
 			fe_flags |= FIEMAP_EXTENT_LAST;
 		len_bytes = (u64)le16_to_cpu(rec.e_leaf_clusters) << osb->s_clustersize_bits;
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index de059f490586..06ccf6a86d35 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1772,7 +1772,8 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
 					 loff_t *ppos,
 					 size_t count,
 					 int appending,
-					 int *direct_io)
+					 int *direct_io,
+					 int *has_refcount)
 {
 	int ret = 0, meta_level = 0;
 	struct inode *inode = dentry->d_inode;
@@ -1833,6 +1834,8 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
 							       saved_pos,
 							       count,
 							       &meta_level);
+			if (has_refcount)
+				*has_refcount = 1;
 		}
 
 		if (ret < 0) {
@@ -1856,6 +1859,10 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
 			break;
 		}
 
+		if (has_refcount && *has_refcount == 1) {
+			*direct_io = 0;
+			break;
+		}
 		/*
 		 * Allowing concurrent direct writes means
 		 * i_size changes wouldn't be synchronized, so
@@ -1899,7 +1906,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
 				    loff_t pos)
 {
 	int ret, direct_io, appending, rw_level, have_alloc_sem  = 0;
-	int can_do_direct;
+	int can_do_direct, has_refcount = 0;
 	ssize_t written = 0;
 	size_t ocount;		/* original count */
 	size_t count;		/* after file limit checks */
@@ -1942,7 +1949,7 @@ relock:
 	can_do_direct = direct_io;
 	ret = ocfs2_prepare_inode_for_write(file->f_path.dentry, ppos,
 					    iocb->ki_left, appending,
-					    &can_do_direct);
+					    &can_do_direct, &has_refcount);
 	if (ret < 0) {
 		mlog_errno(ret);
 		goto out;
@@ -2006,14 +2013,16 @@ out_dio:
 	/* buffered aio wouldn't have proper lock coverage today */
 	BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT));
 
-	if ((file->f_flags & O_SYNC && !direct_io) || IS_SYNC(inode)) {
+	if ((file->f_flags & O_DSYNC && !direct_io) || IS_SYNC(inode) ||
+	    (file->f_flags & O_DIRECT && has_refcount)) {
 		ret = filemap_fdatawrite_range(file->f_mapping, pos,
 					       pos + count - 1);
 		if (ret < 0)
 			written = ret;
 
 		if (!ret && (old_size != i_size_read(inode) ||
-		    old_clusters != OCFS2_I(inode)->ip_clusters)) {
+		    old_clusters != OCFS2_I(inode)->ip_clusters ||
+		    has_refcount)) {
 			ret = jbd2_journal_force_commit(osb->journal->j_journal);
 			if (ret < 0)
 				written = ret;
@@ -2062,7 +2071,7 @@ static int ocfs2_splice_to_file(struct pipe_inode_info *pipe,
 	int ret;
 
 	ret = ocfs2_prepare_inode_for_write(out->f_path.dentry,	&sd->pos,
-					    sd->total_len, 0, NULL);
+					    sd->total_len, 0, NULL, NULL);
 	if (ret < 0) {
 		mlog_errno(ret);
 		return ret;
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index f010b22b1c44..50fb26a6a5f5 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -2108,6 +2108,7 @@ int ocfs2_create_inode_in_orphan(struct inode *dir,
 	}
 	did_quota_inode = 1;
 
+	inode->i_nlink = 0;
 	/* do the real work now. */
 	status = ocfs2_mknod_locked(osb, dir, inode,
 				    0, &new_di_bh, parent_di_bh, handle,
@@ -2136,6 +2137,7 @@ int ocfs2_create_inode_in_orphan(struct inode *dir,
 	if (status < 0)
 		mlog_errno(status);
 
+	insert_inode_hash(inode);
 leave:
 	if (status < 0 && did_quota_inode)
 		vfs_dq_free_inode(inode);
@@ -2267,6 +2269,8 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
 	di = (struct ocfs2_dinode *)di_bh->b_data;
 	le32_add_cpu(&di->i_flags, -OCFS2_ORPHANED_FL);
 	di->i_orphaned_slot = 0;
+	inode->i_nlink = 1;
+	ocfs2_set_links_count(di, inode->i_nlink);
 	ocfs2_journal_dirty(handle, di_bh);
 
 	status = ocfs2_add_entry(handle, dentry, inode,
@@ -2284,7 +2288,6 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir,
 		goto out_commit;
 	}
 
-	insert_inode_hash(inode);
 	dentry->d_op = &ocfs2_dentry_ops;
 	d_instantiate(dentry, inode);
 	status = 0;
@@ -2326,4 +2329,5 @@ const struct inode_operations ocfs2_dir_iops = {
 	.getxattr	= generic_getxattr,
 	.listxattr	= ocfs2_listxattr,
 	.removexattr	= generic_removexattr,
+	.fiemap         = ocfs2_fiemap,
 };
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index d963d8638709..9362eea7424b 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -245,9 +245,11 @@ enum ocfs2_mount_options
 	OCFS2_MOUNT_LOCALFLOCKS = 1 << 5, /* No cluster aware user file locks */
 	OCFS2_MOUNT_NOUSERXATTR = 1 << 6, /* No user xattr */
 	OCFS2_MOUNT_INODE64 = 1 << 7,	/* Allow inode numbers > 2^32 */
-	OCFS2_MOUNT_POSIX_ACL = 1 << 8,	/* POSIX access control lists */
-	OCFS2_MOUNT_USRQUOTA = 1 << 9, /* We support user quotas */
-	OCFS2_MOUNT_GRPQUOTA = 1 << 10, /* We support group quotas */
+	OCFS2_MOUNT_POSIX_ACL = 1 << 8,	/* Force POSIX access control lists */
+	OCFS2_MOUNT_NO_POSIX_ACL = 1 << 9,	/* Disable POSIX access
+						   control lists */
+	OCFS2_MOUNT_USRQUOTA = 1 << 10, /* We support user quotas */
+	OCFS2_MOUNT_GRPQUOTA = 1 << 11, /* We support group quotas */
 };
 
 #define OCFS2_OSB_SOFT_RO			0x0001
diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h
index e9431e4a5e7c..1a1a679e51b5 100644
--- a/fs/ocfs2/ocfs2_fs.h
+++ b/fs/ocfs2/ocfs2_fs.h
@@ -1202,7 +1202,7 @@ struct ocfs2_local_disk_dqinfo {
 /* Header of one chunk of a quota file */
 struct ocfs2_local_disk_chunk {
 	__le32 dqc_free;	/* Number of free entries in the bitmap */
-	u8 dqc_bitmap[0];	/* Bitmap of entries in the corresponding
+	__u8 dqc_bitmap[0];	/* Bitmap of entries in the corresponding
 				 * chunk of quota file */
 };
 
diff --git a/fs/ocfs2/quota.h b/fs/ocfs2/quota.h
index e5df9d170b0c..123bc520a2c0 100644
--- a/fs/ocfs2/quota.h
+++ b/fs/ocfs2/quota.h
@@ -17,10 +17,6 @@
 
 #include "ocfs2.h"
 
-/* Common stuff */
-/* id number of quota format */
-#define QFMT_OCFS2 3
-
 /*
  * In-memory structures
  */
diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c
index 1a2c50a759fa..21f9e71223ca 100644
--- a/fs/ocfs2/quota_local.c
+++ b/fs/ocfs2/quota_local.c
@@ -1325,7 +1325,7 @@ out:
 	return status;
 }
 
-static struct quota_format_ops ocfs2_format_ops = {
+static const struct quota_format_ops ocfs2_format_ops = {
 	.check_quota_file	= ocfs2_local_check_quota_file,
 	.read_file_info		= ocfs2_local_read_info,
 	.write_file_info	= ocfs2_global_write_info,
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index 30967e3f5e43..74db2be75dd6 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -276,7 +276,7 @@ static void ocfs2_erase_refcount_tree_from_list(struct ocfs2_super *osb,
 	spin_unlock(&osb->osb_lock);
 }
 
-void ocfs2_kref_remove_refcount_tree(struct kref *kref)
+static void ocfs2_kref_remove_refcount_tree(struct kref *kref)
 {
 	struct ocfs2_refcount_tree *tree =
 		container_of(kref, struct ocfs2_refcount_tree, rf_getcnt);
@@ -524,23 +524,6 @@ out:
 	return ret;
 }
 
-int ocfs2_lock_refcount_tree_by_inode(struct inode *inode, int rw,
-				      struct ocfs2_refcount_tree **ret_tree,
-				      struct buffer_head **ref_bh)
-{
-	int ret;
-	u64 ref_blkno;
-
-	ret = ocfs2_get_refcount_block(inode, &ref_blkno);
-	if (ret) {
-		mlog_errno(ret);
-		return ret;
-	}
-
-	return ocfs2_lock_refcount_tree(OCFS2_SB(inode->i_sb), ref_blkno,
-					rw, ret_tree, ref_bh);
-}
-
 void ocfs2_unlock_refcount_tree(struct ocfs2_super *osb,
 				struct ocfs2_refcount_tree *tree, int rw)
 {
@@ -969,6 +952,103 @@ out:
 }
 
 /*
+ * Find the end range for a leaf refcount block indicated by
+ * el->l_recs[index].e_blkno.
+ */
+static int ocfs2_get_refcount_cpos_end(struct ocfs2_caching_info *ci,
+				       struct buffer_head *ref_root_bh,
+				       struct ocfs2_extent_block *eb,
+				       struct ocfs2_extent_list *el,
+				       int index,  u32 *cpos_end)
+{
+	int ret, i, subtree_root;
+	u32 cpos;
+	u64 blkno;
+	struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
+	struct ocfs2_path *left_path = NULL, *right_path = NULL;
+	struct ocfs2_extent_tree et;
+	struct ocfs2_extent_list *tmp_el;
+
+	if (index < le16_to_cpu(el->l_next_free_rec) - 1) {
+		/*
+		 * We have a extent rec after index, so just use the e_cpos
+		 * of the next extent rec.
+		 */
+		*cpos_end = le32_to_cpu(el->l_recs[index+1].e_cpos);
+		return 0;
+	}
+
+	if (!eb || (eb && !eb->h_next_leaf_blk)) {
+		/*
+		 * We are the last extent rec, so any high cpos should
+		 * be stored in this leaf refcount block.
+		 */
+		*cpos_end = UINT_MAX;
+		return 0;
+	}
+
+	/*
+	 * If the extent block isn't the last one, we have to find
+	 * the subtree root between this extent block and the next
+	 * leaf extent block and get the corresponding e_cpos from
+	 * the subroot. Otherwise we may corrupt the b-tree.
+	 */
+	ocfs2_init_refcount_extent_tree(&et, ci, ref_root_bh);
+
+	left_path = ocfs2_new_path_from_et(&et);
+	if (!left_path) {
+		ret = -ENOMEM;
+		mlog_errno(ret);
+		goto out;
+	}
+
+	cpos = le32_to_cpu(eb->h_list.l_recs[index].e_cpos);
+	ret = ocfs2_find_path(ci, left_path, cpos);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	right_path = ocfs2_new_path_from_path(left_path);
+	if (!right_path) {
+		ret = -ENOMEM;
+		mlog_errno(ret);
+		goto out;
+	}
+
+	ret = ocfs2_find_cpos_for_right_leaf(sb, left_path, &cpos);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	ret = ocfs2_find_path(ci, right_path, cpos);
+	if (ret) {
+		mlog_errno(ret);
+		goto out;
+	}
+
+	subtree_root = ocfs2_find_subtree_root(&et, left_path,
+					       right_path);
+
+	tmp_el = left_path->p_node[subtree_root].el;
+	blkno = left_path->p_node[subtree_root+1].bh->b_blocknr;
+	for (i = 0; i < le32_to_cpu(tmp_el->l_next_free_rec); i++) {
+		if (le64_to_cpu(tmp_el->l_recs[i].e_blkno) == blkno) {
+			*cpos_end = le32_to_cpu(tmp_el->l_recs[i+1].e_cpos);
+			break;
+		}
+	}
+
+	BUG_ON(i == le32_to_cpu(tmp_el->l_next_free_rec));
+
+out:
+	ocfs2_free_path(left_path);
+	ocfs2_free_path(right_path);
+	return ret;
+}
+
+/*
  * Given a cpos and len, try to find the refcount record which contains cpos.
  * 1. If cpos can be found in one refcount record, return the record.
  * 2. If cpos can't be found, return a fake record which start from cpos
@@ -983,10 +1063,10 @@ static int ocfs2_get_refcount_rec(struct ocfs2_caching_info *ci,
 				  struct buffer_head **ret_bh)
 {
 	int ret = 0, i, found;
-	u32 low_cpos;
+	u32 low_cpos, uninitialized_var(cpos_end);
 	struct ocfs2_extent_list *el;
-	struct ocfs2_extent_rec *tmp, *rec = NULL;
-	struct ocfs2_extent_block *eb;
+	struct ocfs2_extent_rec *rec = NULL;
+	struct ocfs2_extent_block *eb = NULL;
 	struct buffer_head *eb_bh = NULL, *ref_leaf_bh = NULL;
 	struct super_block *sb = ocfs2_metadata_cache_get_super(ci);
 	struct ocfs2_refcount_block *rb =
@@ -1034,12 +1114,16 @@ static int ocfs2_get_refcount_rec(struct ocfs2_caching_info *ci,
 		}
 	}
 
-	/* adjust len when we have ocfs2_extent_rec after it. */
-	if (found && i < le16_to_cpu(el->l_next_free_rec) - 1) {
-		tmp = &el->l_recs[i+1];
+	if (found) {
+		ret = ocfs2_get_refcount_cpos_end(ci, ref_root_bh,
+						  eb, el, i, &cpos_end);
+		if (ret) {
+			mlog_errno(ret);
+			goto out;
+		}
 
-		if (le32_to_cpu(tmp->e_cpos) < cpos + len)
-			len = le32_to_cpu(tmp->e_cpos) - cpos;
+		if (cpos_end < low_cpos + len)
+			len = cpos_end - low_cpos;
 	}
 
 	ret = ocfs2_read_refcount_block(ci, le64_to_cpu(rec->e_blkno),
@@ -1418,7 +1502,7 @@ static int ocfs2_divide_leaf_refcount_block(struct buffer_head *ref_leaf_bh,
 
 	/* change old and new rl_used accordingly. */
 	le16_add_cpu(&rl->rl_used, -num_moved);
-	new_rl->rl_used = cpu_to_le32(num_moved);
+	new_rl->rl_used = cpu_to_le16(num_moved);
 
 	sort(&rl->rl_recs, le16_to_cpu(rl->rl_used),
 	     sizeof(struct ocfs2_refcount_rec),
@@ -1797,7 +1881,8 @@ static int ocfs2_split_refcount_rec(handle_t *handle,
 		recs_need++;
 
 	/* If the leaf block don't have enough record, expand it. */
-	if (le16_to_cpu(rf_list->rl_used) + recs_need > rf_list->rl_count) {
+	if (le16_to_cpu(rf_list->rl_used) + recs_need >
+					 le16_to_cpu(rf_list->rl_count)) {
 		struct ocfs2_refcount_rec tmp_rec;
 		u64 cpos = le64_to_cpu(orig_rec->r_cpos);
 		len = le32_to_cpu(orig_rec->r_clusters);
@@ -1859,7 +1944,7 @@ static int ocfs2_split_refcount_rec(handle_t *handle,
 		memcpy(tail_rec, orig_rec, sizeof(struct ocfs2_refcount_rec));
 		le64_add_cpu(&tail_rec->r_cpos,
 			     le32_to_cpu(tail_rec->r_clusters) - len);
-		tail_rec->r_clusters = le32_to_cpu(len);
+		tail_rec->r_clusters = cpu_to_le32(len);
 	}
 
 	/*
@@ -3840,8 +3925,7 @@ static int ocfs2_add_refcounted_extent(struct inode *inode,
 	}
 
 	ret = ocfs2_insert_extent(handle, et, cpos,
-			cpu_to_le64(ocfs2_clusters_to_blocks(inode->i_sb,
-							     p_cluster)),
+			ocfs2_clusters_to_blocks(inode->i_sb, p_cluster),
 			num_clusters, ext_flags, meta_ac);
 	if (ret) {
 		mlog_errno(ret);
@@ -4253,8 +4337,8 @@ static int ocfs2_user_path_parent(const char __user *path,
  * @new_dentry:        target dentry
  * @preserve:  if true, preserve all file attributes
  */
-int ocfs2_vfs_reflink(struct dentry *old_dentry, struct inode *dir,
-		      struct dentry *new_dentry, bool preserve)
+static int ocfs2_vfs_reflink(struct dentry *old_dentry, struct inode *dir,
+			     struct dentry *new_dentry, bool preserve)
 {
 	struct inode *inode = old_dentry->d_inode;
 	int error;
diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c
index ff4c798a5635..da78a2a334fd 100644
--- a/fs/ocfs2/stack_user.c
+++ b/fs/ocfs2/stack_user.c
@@ -814,7 +814,7 @@ static int fs_protocol_compare(struct ocfs2_protocol_version *existing,
 static int user_cluster_connect(struct ocfs2_cluster_connection *conn)
 {
 	dlm_lockspace_t *fsdlm;
-	struct ocfs2_live_connection *control;
+	struct ocfs2_live_connection *uninitialized_var(control);
 	int rc = 0;
 
 	BUG_ON(conn == NULL);
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 14f47d2bfe02..26069917a9f5 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -100,6 +100,8 @@ struct mount_options
 static int ocfs2_parse_options(struct super_block *sb, char *options,
 			       struct mount_options *mopt,
 			       int is_remount);
+static int ocfs2_check_set_options(struct super_block *sb,
+				   struct mount_options *options);
 static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt);
 static void ocfs2_put_super(struct super_block *sb);
 static int ocfs2_mount_volume(struct super_block *sb);
@@ -600,7 +602,8 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data)
 
 	lock_kernel();
 
-	if (!ocfs2_parse_options(sb, data, &parsed_options, 1)) {
+	if (!ocfs2_parse_options(sb, data, &parsed_options, 1) ||
+	    !ocfs2_check_set_options(sb, &parsed_options)) {
 		ret = -EINVAL;
 		goto out;
 	}
@@ -691,8 +694,6 @@ unlock_osb:
 	if (!ret) {
 		/* Only save off the new mount options in case of a successful
 		 * remount. */
-		if (!(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_XATTR))
-			parsed_options.mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
 		osb->s_mount_opt = parsed_options.mount_opt;
 		osb->s_atime_quantum = parsed_options.atime_quantum;
 		osb->preferred_slot = parsed_options.slot;
@@ -701,6 +702,10 @@ unlock_osb:
 
 		if (!ocfs2_is_hard_readonly(osb))
 			ocfs2_set_journal_params(osb);
+
+		sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
+			((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ?
+							MS_POSIXACL : 0);
 	}
 out:
 	unlock_kernel();
@@ -1011,31 +1016,16 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent)
 	brelse(bh);
 	bh = NULL;
 
-	if (!(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_XATTR))
-		parsed_options.mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
-
+	if (!ocfs2_check_set_options(sb, &parsed_options)) {
+		status = -EINVAL;
+		goto read_super_error;
+	}
 	osb->s_mount_opt = parsed_options.mount_opt;
 	osb->s_atime_quantum = parsed_options.atime_quantum;
 	osb->preferred_slot = parsed_options.slot;
 	osb->osb_commit_interval = parsed_options.commit_interval;
 	osb->local_alloc_default_bits = ocfs2_megabytes_to_clusters(sb, parsed_options.localalloc_opt);
 	osb->local_alloc_bits = osb->local_alloc_default_bits;
-	if (osb->s_mount_opt & OCFS2_MOUNT_USRQUOTA &&
-	    !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
-					 OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
-		status = -EINVAL;
-		mlog(ML_ERROR, "User quotas were requested, but this "
-		     "filesystem does not have the feature enabled.\n");
-		goto read_super_error;
-	}
-	if (osb->s_mount_opt & OCFS2_MOUNT_GRPQUOTA &&
-	    !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
-					 OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
-		status = -EINVAL;
-		mlog(ML_ERROR, "Group quotas were requested, but this "
-		     "filesystem does not have the feature enabled.\n");
-		goto read_super_error;
-	}
 
 	status = ocfs2_verify_userspace_stack(osb, &parsed_options);
 	if (status)
@@ -1245,6 +1235,40 @@ static struct file_system_type ocfs2_fs_type = {
 	.next           = NULL
 };
 
+static int ocfs2_check_set_options(struct super_block *sb,
+				   struct mount_options *options)
+{
+	if (options->mount_opt & OCFS2_MOUNT_USRQUOTA &&
+	    !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
+					 OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
+		mlog(ML_ERROR, "User quotas were requested, but this "
+		     "filesystem does not have the feature enabled.\n");
+		return 0;
+	}
+	if (options->mount_opt & OCFS2_MOUNT_GRPQUOTA &&
+	    !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
+					 OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
+		mlog(ML_ERROR, "Group quotas were requested, but this "
+		     "filesystem does not have the feature enabled.\n");
+		return 0;
+	}
+	if (options->mount_opt & OCFS2_MOUNT_POSIX_ACL &&
+	    !OCFS2_HAS_INCOMPAT_FEATURE(sb, OCFS2_FEATURE_INCOMPAT_XATTR)) {
+		mlog(ML_ERROR, "ACL support requested but extended attributes "
+		     "feature is not enabled\n");
+		return 0;
+	}
+	/* No ACL setting specified? Use XATTR feature... */
+	if (!(options->mount_opt & (OCFS2_MOUNT_POSIX_ACL |
+				    OCFS2_MOUNT_NO_POSIX_ACL))) {
+		if (OCFS2_HAS_INCOMPAT_FEATURE(sb, OCFS2_FEATURE_INCOMPAT_XATTR))
+			options->mount_opt |= OCFS2_MOUNT_POSIX_ACL;
+		else
+			options->mount_opt |= OCFS2_MOUNT_NO_POSIX_ACL;
+	}
+	return 1;
+}
+
 static int ocfs2_parse_options(struct super_block *sb,
 			       char *options,
 			       struct mount_options *mopt,
@@ -1392,40 +1416,19 @@ static int ocfs2_parse_options(struct super_block *sb,
 			mopt->mount_opt |= OCFS2_MOUNT_INODE64;
 			break;
 		case Opt_usrquota:
-			/* We check only on remount, otherwise features
-			 * aren't yet initialized. */
-			if (is_remount && !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
-			    OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) {
-				mlog(ML_ERROR, "User quota requested but "
-				     "filesystem feature is not set\n");
-				status = 0;
-				goto bail;
-			}
 			mopt->mount_opt |= OCFS2_MOUNT_USRQUOTA;
 			break;
 		case Opt_grpquota:
-			if (is_remount && !OCFS2_HAS_RO_COMPAT_FEATURE(sb,
-			    OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) {
-				mlog(ML_ERROR, "Group quota requested but "
-				     "filesystem feature is not set\n");
-				status = 0;
-				goto bail;
-			}
 			mopt->mount_opt |= OCFS2_MOUNT_GRPQUOTA;
 			break;
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
 		case Opt_acl:
 			mopt->mount_opt |= OCFS2_MOUNT_POSIX_ACL;
+			mopt->mount_opt &= ~OCFS2_MOUNT_NO_POSIX_ACL;
 			break;
 		case Opt_noacl:
+			mopt->mount_opt |= OCFS2_MOUNT_NO_POSIX_ACL;
 			mopt->mount_opt &= ~OCFS2_MOUNT_POSIX_ACL;
 			break;
-#else
-		case Opt_acl:
-		case Opt_noacl:
-			printk(KERN_INFO "ocfs2 (no)acl options not supported\n");
-			break;
-#endif
 		default:
 			mlog(ML_ERROR,
 			     "Unrecognized mount option \"%s\" "
@@ -1502,12 +1505,10 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt)
 	if (opts & OCFS2_MOUNT_INODE64)
 		seq_printf(s, ",inode64");
 
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
 	if (opts & OCFS2_MOUNT_POSIX_ACL)
 		seq_printf(s, ",acl");
 	else
 		seq_printf(s, ",noacl");
-#endif
 
 	return 0;
 }
diff --git a/fs/ocfs2/symlink.c b/fs/ocfs2/symlink.c
index e3421030a69f..49b133ccbf11 100644
--- a/fs/ocfs2/symlink.c
+++ b/fs/ocfs2/symlink.c
@@ -163,6 +163,7 @@ const struct inode_operations ocfs2_symlink_inode_operations = {
 	.getxattr	= generic_getxattr,
 	.listxattr	= ocfs2_listxattr,
 	.removexattr	= generic_removexattr,
+	.fiemap		= ocfs2_fiemap,
 };
 const struct inode_operations ocfs2_fast_symlink_inode_operations = {
 	.readlink	= ocfs2_readlink,
@@ -174,4 +175,5 @@ const struct inode_operations ocfs2_fast_symlink_inode_operations = {
 	.getxattr	= generic_getxattr,
 	.listxattr	= ocfs2_listxattr,
 	.removexattr	= generic_removexattr,
+	.fiemap		= ocfs2_fiemap,
 };
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index fe3419068df2..8fc6fb071c6d 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -98,10 +98,8 @@ static struct ocfs2_xattr_def_value_root def_xv = {
 
 struct xattr_handler *ocfs2_xattr_handlers[] = {
 	&ocfs2_xattr_user_handler,
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
 	&ocfs2_xattr_acl_access_handler,
 	&ocfs2_xattr_acl_default_handler,
-#endif
 	&ocfs2_xattr_trusted_handler,
 	&ocfs2_xattr_security_handler,
 	NULL
@@ -109,12 +107,10 @@ struct xattr_handler *ocfs2_xattr_handlers[] = {
 
 static struct xattr_handler *ocfs2_xattr_handler_map[OCFS2_XATTR_MAX] = {
 	[OCFS2_XATTR_INDEX_USER]	= &ocfs2_xattr_user_handler,
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
 	[OCFS2_XATTR_INDEX_POSIX_ACL_ACCESS]
 					= &ocfs2_xattr_acl_access_handler,
 	[OCFS2_XATTR_INDEX_POSIX_ACL_DEFAULT]
 					= &ocfs2_xattr_acl_default_handler,
-#endif
 	[OCFS2_XATTR_INDEX_TRUSTED]	= &ocfs2_xattr_trusted_handler,
 	[OCFS2_XATTR_INDEX_SECURITY]	= &ocfs2_xattr_security_handler,
 };
@@ -205,8 +201,6 @@ static int ocfs2_get_xattr_tree_value_root(struct super_block *sb,
 					   int offset,
 					   struct ocfs2_xattr_value_root **xv,
 					   struct buffer_head **bh);
-static int ocfs2_xattr_security_set(struct inode *inode, const char *name,
-				    const void *value, size_t size, int flags);
 
 static inline u16 ocfs2_xattr_buckets_per_cluster(struct ocfs2_super *osb)
 {
@@ -6066,7 +6060,7 @@ static int ocfs2_value_metas_in_xattr_header(struct super_block *sb,
 		 * to the extent block, so just calculate a maximum record num.
 		 */
 		if (!xv->xr_list.l_tree_depth)
-			*num_recs += xv->xr_list.l_next_free_rec;
+			*num_recs += le16_to_cpu(xv->xr_list.l_next_free_rec);
 		else
 			*num_recs += ocfs2_clusters_for_bytes(sb,
 							      XATTR_SIZE_MAX);
@@ -6978,9 +6972,9 @@ int ocfs2_init_security_and_acl(struct inode *dir,
 
 	ret = ocfs2_init_security_get(inode, dir, &si);
 	if (!ret) {
-		ret = ocfs2_xattr_security_set(inode, si.name,
-					       si.value, si.value_len,
-					       XATTR_CREATE);
+		ret = ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_SECURITY,
+				      si.name, si.value, si.value_len,
+				      XATTR_CREATE);
 		if (ret) {
 			mlog_errno(ret);
 			goto leave;
@@ -7008,9 +7002,9 @@ leave:
 /*
  * 'security' attributes support
  */
-static size_t ocfs2_xattr_security_list(struct inode *inode, char *list,
+static size_t ocfs2_xattr_security_list(struct dentry *dentry, char *list,
 					size_t list_size, const char *name,
-					size_t name_len)
+					size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_SECURITY_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
@@ -7023,23 +7017,23 @@ static size_t ocfs2_xattr_security_list(struct inode *inode, char *list,
 	return total_len;
 }
 
-static int ocfs2_xattr_security_get(struct inode *inode, const char *name,
-				    void *buffer, size_t size)
+static int ocfs2_xattr_security_get(struct dentry *dentry, const char *name,
+				    void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ocfs2_xattr_get(inode, OCFS2_XATTR_INDEX_SECURITY, name,
-			       buffer, size);
+	return ocfs2_xattr_get(dentry->d_inode, OCFS2_XATTR_INDEX_SECURITY,
+			       name, buffer, size);
 }
 
-static int ocfs2_xattr_security_set(struct inode *inode, const char *name,
-				    const void *value, size_t size, int flags)
+static int ocfs2_xattr_security_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
 
-	return ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_SECURITY, name, value,
-			       size, flags);
+	return ocfs2_xattr_set(dentry->d_inode, OCFS2_XATTR_INDEX_SECURITY,
+			       name, value, size, flags);
 }
 
 int ocfs2_init_security_get(struct inode *inode,
@@ -7076,9 +7070,9 @@ struct xattr_handler ocfs2_xattr_security_handler = {
 /*
  * 'trusted' attributes support
  */
-static size_t ocfs2_xattr_trusted_list(struct inode *inode, char *list,
+static size_t ocfs2_xattr_trusted_list(struct dentry *dentry, char *list,
 				       size_t list_size, const char *name,
-				       size_t name_len)
+				       size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_TRUSTED_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
@@ -7091,23 +7085,23 @@ static size_t ocfs2_xattr_trusted_list(struct inode *inode, char *list,
 	return total_len;
 }
 
-static int ocfs2_xattr_trusted_get(struct inode *inode, const char *name,
-				   void *buffer, size_t size)
+static int ocfs2_xattr_trusted_get(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
-	return ocfs2_xattr_get(inode, OCFS2_XATTR_INDEX_TRUSTED, name,
-			       buffer, size);
+	return ocfs2_xattr_get(dentry->d_inode, OCFS2_XATTR_INDEX_TRUSTED,
+			       name, buffer, size);
 }
 
-static int ocfs2_xattr_trusted_set(struct inode *inode, const char *name,
-				   const void *value, size_t size, int flags)
+static int ocfs2_xattr_trusted_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
 
-	return ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_TRUSTED, name, value,
-			       size, flags);
+	return ocfs2_xattr_set(dentry->d_inode, OCFS2_XATTR_INDEX_TRUSTED,
+			       name, value, size, flags);
 }
 
 struct xattr_handler ocfs2_xattr_trusted_handler = {
@@ -7120,13 +7114,13 @@ struct xattr_handler ocfs2_xattr_trusted_handler = {
 /*
  * 'user' attributes support
  */
-static size_t ocfs2_xattr_user_list(struct inode *inode, char *list,
+static size_t ocfs2_xattr_user_list(struct dentry *dentry, char *list,
 				    size_t list_size, const char *name,
-				    size_t name_len)
+				    size_t name_len, int type)
 {
 	const size_t prefix_len = XATTR_USER_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
 
 	if (osb->s_mount_opt & OCFS2_MOUNT_NOUSERXATTR)
 		return 0;
@@ -7139,31 +7133,31 @@ static size_t ocfs2_xattr_user_list(struct inode *inode, char *list,
 	return total_len;
 }
 
-static int ocfs2_xattr_user_get(struct inode *inode, const char *name,
-				void *buffer, size_t size)
+static int ocfs2_xattr_user_get(struct dentry *dentry, const char *name,
+		void *buffer, size_t size, int type)
 {
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
 
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
 	if (osb->s_mount_opt & OCFS2_MOUNT_NOUSERXATTR)
 		return -EOPNOTSUPP;
-	return ocfs2_xattr_get(inode, OCFS2_XATTR_INDEX_USER, name,
+	return ocfs2_xattr_get(dentry->d_inode, OCFS2_XATTR_INDEX_USER, name,
 			       buffer, size);
 }
 
-static int ocfs2_xattr_user_set(struct inode *inode, const char *name,
-				const void *value, size_t size, int flags)
+static int ocfs2_xattr_user_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+	struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
 
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
 	if (osb->s_mount_opt & OCFS2_MOUNT_NOUSERXATTR)
 		return -EOPNOTSUPP;
 
-	return ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_USER, name, value,
-			       size, flags);
+	return ocfs2_xattr_set(dentry->d_inode, OCFS2_XATTR_INDEX_USER,
+			       name, value, size, flags);
 }
 
 struct xattr_handler ocfs2_xattr_user_handler = {
diff --git a/fs/ocfs2/xattr.h b/fs/ocfs2/xattr.h
index 08e36389f56d..abd72a47f520 100644
--- a/fs/ocfs2/xattr.h
+++ b/fs/ocfs2/xattr.h
@@ -40,10 +40,8 @@ struct ocfs2_security_xattr_info {
 extern struct xattr_handler ocfs2_xattr_user_handler;
 extern struct xattr_handler ocfs2_xattr_trusted_handler;
 extern struct xattr_handler ocfs2_xattr_security_handler;
-#ifdef CONFIG_OCFS2_FS_POSIX_ACL
 extern struct xattr_handler ocfs2_xattr_acl_access_handler;
 extern struct xattr_handler ocfs2_xattr_acl_default_handler;
-#endif
 extern struct xattr_handler *ocfs2_xattr_handlers[];
 
 ssize_t ocfs2_listxattr(struct dentry *, char *, size_t);
diff --git a/fs/open.c b/fs/open.c
index b4b31d277f3a..040cef72bc00 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -30,6 +30,9 @@
 #include <linux/audit.h>
 #include <linux/falloc.h>
 #include <linux/fs_struct.h>
+#include <linux/ima.h>
+
+#include "internal.h"
 
 int vfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
@@ -818,15 +821,14 @@ static inline int __get_file_write_access(struct inode *inode,
 }
 
 static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
-					int flags, struct file *f,
+					struct file *f,
 					int (*open)(struct inode *, struct file *),
 					const struct cred *cred)
 {
 	struct inode *inode;
 	int error;
 
-	f->f_flags = flags;
-	f->f_mode = (__force fmode_t)((flags+1) & O_ACCMODE) | FMODE_LSEEK |
+	f->f_mode = OPEN_FMODE(f->f_flags) | FMODE_LSEEK |
 				FMODE_PREAD | FMODE_PWRITE;
 	inode = dentry->d_inode;
 	if (f->f_mode & FMODE_WRITE) {
@@ -855,6 +857,7 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt,
 		if (error)
 			goto cleanup_all;
 	}
+	ima_counts_get(f);
 
 	f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);
 
@@ -926,7 +929,6 @@ struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry
 	if (IS_ERR(dentry))
 		goto out_err;
 	nd->intent.open.file = __dentry_open(dget(dentry), mntget(nd->path.mnt),
-					     nd->intent.open.flags - 1,
 					     nd->intent.open.file,
 					     open, cred);
 out:
@@ -945,7 +947,7 @@ EXPORT_SYMBOL_GPL(lookup_instantiate_filp);
  *
  * Note that this function destroys the original nameidata
  */
-struct file *nameidata_to_filp(struct nameidata *nd, int flags)
+struct file *nameidata_to_filp(struct nameidata *nd)
 {
 	const struct cred *cred = current_cred();
 	struct file *filp;
@@ -954,7 +956,7 @@ struct file *nameidata_to_filp(struct nameidata *nd, int flags)
 	filp = nd->intent.open.file;
 	/* Has the filesystem initialised the file for us? */
 	if (filp->f_path.dentry == NULL)
-		filp = __dentry_open(nd->path.dentry, nd->path.mnt, flags, filp,
+		filp = __dentry_open(nd->path.dentry, nd->path.mnt, filp,
 				     NULL, cred);
 	else
 		path_put(&nd->path);
@@ -993,7 +995,8 @@ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags,
 		return ERR_PTR(error);
 	}
 
-	return __dentry_open(dentry, mnt, flags, f, NULL, cred);
+	f->f_flags = flags;
+	return __dentry_open(dentry, mnt, f, NULL, cred);
 }
 EXPORT_SYMBOL(dentry_open);
 
diff --git a/fs/pipe.c b/fs/pipe.c
index ae17d026aaa3..37ba29ff3158 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -906,17 +906,6 @@ void free_pipe_info(struct inode *inode)
 }
 
 static struct vfsmount *pipe_mnt __read_mostly;
-static int pipefs_delete_dentry(struct dentry *dentry)
-{
-	/*
-	 * At creation time, we pretended this dentry was hashed
-	 * (by clearing DCACHE_UNHASHED bit in d_flags)
-	 * At delete time, we restore the truth : not hashed.
-	 * (so that dput() can proceed correctly)
-	 */
-	dentry->d_flags |= DCACHE_UNHASHED;
-	return 0;
-}
 
 /*
  * pipefs_dname() is called from d_path().
@@ -928,7 +917,6 @@ static char *pipefs_dname(struct dentry *dentry, char *buffer, int buflen)
 }
 
 static const struct dentry_operations pipefs_dentry_operations = {
-	.d_delete	= pipefs_delete_dentry,
 	.d_dname	= pipefs_dname,
 };
 
@@ -974,7 +962,7 @@ struct file *create_write_pipe(int flags)
 	int err;
 	struct inode *inode;
 	struct file *f;
-	struct dentry *dentry;
+	struct path path;
 	struct qstr name = { .name = "" };
 
 	err = -ENFILE;
@@ -983,21 +971,16 @@ struct file *create_write_pipe(int flags)
 		goto err;
 
 	err = -ENOMEM;
-	dentry = d_alloc(pipe_mnt->mnt_sb->s_root, &name);
-	if (!dentry)
+	path.dentry = d_alloc(pipe_mnt->mnt_sb->s_root, &name);
+	if (!path.dentry)
 		goto err_inode;
+	path.mnt = mntget(pipe_mnt);
 
-	dentry->d_op = &pipefs_dentry_operations;
-	/*
-	 * We dont want to publish this dentry into global dentry hash table.
-	 * We pretend dentry is already hashed, by unsetting DCACHE_UNHASHED
-	 * This permits a working /proc/$pid/fd/XXX on pipes
-	 */
-	dentry->d_flags &= ~DCACHE_UNHASHED;
-	d_instantiate(dentry, inode);
+	path.dentry->d_op = &pipefs_dentry_operations;
+	d_instantiate(path.dentry, inode);
 
 	err = -ENFILE;
-	f = alloc_file(pipe_mnt, dentry, FMODE_WRITE, &write_pipefifo_fops);
+	f = alloc_file(&path, FMODE_WRITE, &write_pipefifo_fops);
 	if (!f)
 		goto err_dentry;
 	f->f_mapping = inode->i_mapping;
@@ -1009,7 +992,7 @@ struct file *create_write_pipe(int flags)
 
  err_dentry:
 	free_pipe_info(inode);
-	dput(dentry);
+	path_put(&path);
 	return ERR_PTR(err);
 
  err_inode:
@@ -1028,20 +1011,14 @@ void free_write_pipe(struct file *f)
 
 struct file *create_read_pipe(struct file *wrf, int flags)
 {
-	struct file *f = get_empty_filp();
+	/* Grab pipe from the writer */
+	struct file *f = alloc_file(&wrf->f_path, FMODE_READ,
+				    &read_pipefifo_fops);
 	if (!f)
 		return ERR_PTR(-ENFILE);
 
-	/* Grab pipe from the writer */
-	f->f_path = wrf->f_path;
 	path_get(&wrf->f_path);
-	f->f_mapping = wrf->f_path.dentry->d_inode->i_mapping;
-
-	f->f_pos = 0;
 	f->f_flags = O_RDONLY | (flags & O_NONBLOCK);
-	f->f_op = &read_pipefifo_fops;
-	f->f_mode = FMODE_READ;
-	f->f_version = 0;
 
 	return f;
 }
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 4badde179b18..13b5d0708175 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -134,13 +134,16 @@ static inline void task_name(struct seq_file *m, struct task_struct *p)
  * simple bit tests.
  */
 static const char *task_state_array[] = {
-	"R (running)",		/*  0 */
-	"S (sleeping)",		/*  1 */
-	"D (disk sleep)",	/*  2 */
-	"T (stopped)",		/*  4 */
-	"T (tracing stop)",	/*  8 */
-	"Z (zombie)",		/* 16 */
-	"X (dead)"		/* 32 */
+	"R (running)",		/*   0 */
+	"S (sleeping)",		/*   1 */
+	"D (disk sleep)",	/*   2 */
+	"T (stopped)",		/*   4 */
+	"t (tracing stop)",	/*   8 */
+	"Z (zombie)",		/*  16 */
+	"X (dead)",		/*  32 */
+	"x (dead)",		/*  64 */
+	"K (wakekill)",		/* 128 */
+	"W (waking)",		/* 256 */
 };
 
 static inline const char *get_task_state(struct task_struct *tsk)
@@ -148,6 +151,8 @@ static inline const char *get_task_state(struct task_struct *tsk)
 	unsigned int state = (tsk->state & TASK_REPORT) | tsk->exit_state;
 	const char **p = &task_state_array[0];
 
+	BUILD_BUG_ON(1 + ilog2(TASK_STATE_MAX) != ARRAY_SIZE(task_state_array));
+
 	while (state) {
 		p++;
 		state >>= 1;
@@ -322,94 +327,6 @@ static inline void task_context_switch_counts(struct seq_file *m,
 			p->nivcsw);
 }
 
-#ifdef CONFIG_MMU
-
-struct stack_stats {
-	struct vm_area_struct *vma;
-	unsigned long	startpage;
-	unsigned long	usage;
-};
-
-static int stack_usage_pte_range(pmd_t *pmd, unsigned long addr,
-				unsigned long end, struct mm_walk *walk)
-{
-	struct stack_stats *ss = walk->private;
-	struct vm_area_struct *vma = ss->vma;
-	pte_t *pte, ptent;
-	spinlock_t *ptl;
-	int ret = 0;
-
-	pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
-	for (; addr != end; pte++, addr += PAGE_SIZE) {
-		ptent = *pte;
-
-#ifdef CONFIG_STACK_GROWSUP
-		if (pte_present(ptent) || is_swap_pte(ptent))
-			ss->usage = addr - ss->startpage + PAGE_SIZE;
-#else
-		if (pte_present(ptent) || is_swap_pte(ptent)) {
-			ss->usage = ss->startpage - addr + PAGE_SIZE;
-			pte++;
-			ret = 1;
-			break;
-		}
-#endif
-	}
-	pte_unmap_unlock(pte - 1, ptl);
-	cond_resched();
-	return ret;
-}
-
-static inline unsigned long get_stack_usage_in_bytes(struct vm_area_struct *vma,
-				struct task_struct *task)
-{
-	struct stack_stats ss;
-	struct mm_walk stack_walk = {
-		.pmd_entry = stack_usage_pte_range,
-		.mm = vma->vm_mm,
-		.private = &ss,
-	};
-
-	if (!vma->vm_mm || is_vm_hugetlb_page(vma))
-		return 0;
-
-	ss.vma = vma;
-	ss.startpage = task->stack_start & PAGE_MASK;
-	ss.usage = 0;
-
-#ifdef CONFIG_STACK_GROWSUP
-	walk_page_range(KSTK_ESP(task) & PAGE_MASK, vma->vm_end,
-		&stack_walk);
-#else
-	walk_page_range(vma->vm_start, (KSTK_ESP(task) & PAGE_MASK) + PAGE_SIZE,
-		&stack_walk);
-#endif
-	return ss.usage;
-}
-
-static inline void task_show_stack_usage(struct seq_file *m,
-						struct task_struct *task)
-{
-	struct vm_area_struct	*vma;
-	struct mm_struct	*mm = get_task_mm(task);
-
-	if (mm) {
-		down_read(&mm->mmap_sem);
-		vma = find_vma(mm, task->stack_start);
-		if (vma)
-			seq_printf(m, "Stack usage:\t%lu kB\n",
-				get_stack_usage_in_bytes(vma, task) >> 10);
-
-		up_read(&mm->mmap_sem);
-		mmput(mm);
-	}
-}
-#else
-static void task_show_stack_usage(struct seq_file *m, struct task_struct *task)
-{
-}
-#endif		/* CONFIG_MMU */
-
 static void task_cpus_allowed(struct seq_file *m, struct task_struct *task)
 {
 	seq_printf(m, "Cpus_allowed:\t");
@@ -440,7 +357,6 @@ int proc_pid_status(struct seq_file *m, struct pid_namespace *ns,
 	task_show_regs(m, task);
 #endif
 	task_context_switch_counts(m, task);
-	task_show_stack_usage(m, task);
 	return 0;
 }
 
diff --git a/fs/proc/base.c b/fs/proc/base.c
index af643b5aefe8..e42bbd843ed1 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1265,6 +1265,72 @@ static const struct file_operations proc_pid_sched_operations = {
 
 #endif
 
+static ssize_t comm_write(struct file *file, const char __user *buf,
+				size_t count, loff_t *offset)
+{
+	struct inode *inode = file->f_path.dentry->d_inode;
+	struct task_struct *p;
+	char buffer[TASK_COMM_LEN];
+
+	memset(buffer, 0, sizeof(buffer));
+	if (count > sizeof(buffer) - 1)
+		count = sizeof(buffer) - 1;
+	if (copy_from_user(buffer, buf, count))
+		return -EFAULT;
+
+	p = get_proc_task(inode);
+	if (!p)
+		return -ESRCH;
+
+	if (same_thread_group(current, p))
+		set_task_comm(p, buffer);
+	else
+		count = -EINVAL;
+
+	put_task_struct(p);
+
+	return count;
+}
+
+static int comm_show(struct seq_file *m, void *v)
+{
+	struct inode *inode = m->private;
+	struct task_struct *p;
+
+	p = get_proc_task(inode);
+	if (!p)
+		return -ESRCH;
+
+	task_lock(p);
+	seq_printf(m, "%s\n", p->comm);
+	task_unlock(p);
+
+	put_task_struct(p);
+
+	return 0;
+}
+
+static int comm_open(struct inode *inode, struct file *filp)
+{
+	int ret;
+
+	ret = single_open(filp, comm_show, NULL);
+	if (!ret) {
+		struct seq_file *m = filp->private_data;
+
+		m->private = inode;
+	}
+	return ret;
+}
+
+static const struct file_operations proc_pid_set_comm_operations = {
+	.open		= comm_open,
+	.read		= seq_read,
+	.write		= comm_write,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
 /*
  * We added or removed a vma mapping the executable. The vmas are only mapped
  * during exec and are not mapped with the mmap system call.
@@ -1353,7 +1419,6 @@ static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd)
 		goto out;
 
 	error = PROC_I(inode)->op.proc_get_link(inode, &nd->path);
-	nd->last_type = LAST_BIND;
 out:
 	return ERR_PTR(error);
 }
@@ -2200,7 +2265,7 @@ static const struct inode_operations proc_attr_dir_inode_operations = {
 
 #endif
 
-#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE)
+#ifdef CONFIG_ELF_CORE
 static ssize_t proc_coredump_filter_read(struct file *file, char __user *buf,
 					 size_t count, loff_t *ppos)
 {
@@ -2504,6 +2569,7 @@ static const struct pid_entry tgid_base_stuff[] = {
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",      S_IRUGO|S_IWUSR, proc_pid_sched_operations),
 #endif
+	REG("comm",      S_IRUGO|S_IWUSR, proc_pid_set_comm_operations),
 #ifdef CONFIG_HAVE_ARCH_TRACEHOOK
 	INF("syscall",    S_IRUSR, proc_pid_syscall),
 #endif
@@ -2556,7 +2622,7 @@ static const struct pid_entry tgid_base_stuff[] = {
 #ifdef CONFIG_FAULT_INJECTION
 	REG("make-it-fail", S_IRUGO|S_IWUSR, proc_fault_inject_operations),
 #endif
-#if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE)
+#ifdef CONFIG_ELF_CORE
 	REG("coredump_filter", S_IRUGO|S_IWUSR, proc_coredump_filter_operations),
 #endif
 #ifdef CONFIG_TASK_IO_ACCOUNTING
@@ -2838,6 +2904,7 @@ static const struct pid_entry tid_base_stuff[] = {
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",     S_IRUGO|S_IWUSR, proc_pid_sched_operations),
 #endif
+	REG("comm",      S_IRUGO|S_IWUSR, proc_pid_set_comm_operations),
 #ifdef CONFIG_HAVE_ARCH_TRACEHOOK
 	INF("syscall",   S_IRUSR, proc_pid_syscall),
 #endif
diff --git a/fs/proc/generic.c b/fs/proc/generic.c
index fa678abc9db1..480cb1065eec 100644
--- a/fs/proc/generic.c
+++ b/fs/proc/generic.c
@@ -429,7 +429,7 @@ struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *dir,
 			unsigned int ino;
 
 			ino = de->low_ino;
-			de_get(de);
+			pde_get(de);
 			spin_unlock(&proc_subdir_lock);
 			error = -EINVAL;
 			inode = proc_get_inode(dir->i_sb, ino, de);
@@ -445,7 +445,7 @@ out_unlock:
 		return NULL;
 	}
 	if (de)
-		de_put(de);
+		pde_put(de);
 	return ERR_PTR(error);
 }
 
@@ -509,17 +509,17 @@ int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent,
 				struct proc_dir_entry *next;
 
 				/* filldir passes info to user space */
-				de_get(de);
+				pde_get(de);
 				spin_unlock(&proc_subdir_lock);
 				if (filldir(dirent, de->name, de->namelen, filp->f_pos,
 					    de->low_ino, de->mode >> 12) < 0) {
-					de_put(de);
+					pde_put(de);
 					goto out;
 				}
 				spin_lock(&proc_subdir_lock);
 				filp->f_pos++;
 				next = de->next;
-				de_put(de);
+				pde_put(de);
 				de = next;
 			} while (de);
 			spin_unlock(&proc_subdir_lock);
@@ -763,7 +763,7 @@ out:
 	return NULL;
 }
 
-void free_proc_entry(struct proc_dir_entry *de)
+static void free_proc_entry(struct proc_dir_entry *de)
 {
 	unsigned int ino = de->low_ino;
 
@@ -777,6 +777,12 @@ void free_proc_entry(struct proc_dir_entry *de)
 	kfree(de);
 }
 
+void pde_put(struct proc_dir_entry *pde)
+{
+	if (atomic_dec_and_test(&pde->count))
+		free_proc_entry(pde);
+}
+
 /*
  * Remove a /proc entry and free it if it's not currently in use.
  */
@@ -845,6 +851,5 @@ continue_removing:
 	WARN(de->subdir, KERN_WARNING "%s: removing non-empty directory "
 			"'%s/%s', leaking at least '%s'\n", __func__,
 			de->parent->name, de->name, de->subdir->name);
-	if (atomic_dec_and_test(&de->count))
-		free_proc_entry(de);
+	pde_put(de);
 }
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index d78ade305541..445a02bcaab3 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -24,29 +24,6 @@
 
 #include "internal.h"
 
-struct proc_dir_entry *de_get(struct proc_dir_entry *de)
-{
-	atomic_inc(&de->count);
-	return de;
-}
-
-/*
- * Decrements the use count and checks for deferred deletion.
- */
-void de_put(struct proc_dir_entry *de)
-{
-	if (!atomic_read(&de->count)) {
-		printk("de_put: entry %s already free!\n", de->name);
-		return;
-	}
-
-	if (atomic_dec_and_test(&de->count))
-		free_proc_entry(de);
-}
-
-/*
- * Decrement the use count of the proc_dir_entry.
- */
 static void proc_delete_inode(struct inode *inode)
 {
 	struct proc_dir_entry *de;
@@ -59,7 +36,7 @@ static void proc_delete_inode(struct inode *inode)
 	/* Let go of any associated proc directory entry */
 	de = PROC_I(inode)->pde;
 	if (de)
-		de_put(de);
+		pde_put(de);
 	if (PROC_I(inode)->sysctl)
 		sysctl_head_put(PROC_I(inode)->sysctl);
 	clear_inode(inode);
@@ -480,7 +457,7 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino,
 		}
 		unlock_new_inode(inode);
 	} else
-	       de_put(de);
+	       pde_put(de);
 	return inode;
 }			
 
@@ -495,7 +472,7 @@ int proc_fill_super(struct super_block *s)
 	s->s_op = &proc_sops;
 	s->s_time_gran = 1;
 	
-	de_get(&proc_root);
+	pde_get(&proc_root);
 	root_inode = proc_get_inode(s, PROC_ROOT_INO, &proc_root);
 	if (!root_inode)
 		goto out_no_root;
@@ -509,6 +486,6 @@ int proc_fill_super(struct super_block *s)
 out_no_root:
 	printk("proc_read_super: get root inode failed\n");
 	iput(root_inode);
-	de_put(&proc_root);
+	pde_put(&proc_root);
 	return -ENOMEM;
 }
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 753ca37002c8..1f24a3eddd12 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -61,8 +61,6 @@ extern const struct file_operations proc_pagemap_operations;
 extern const struct file_operations proc_net_operations;
 extern const struct inode_operations proc_net_inode_operations;
 
-void free_proc_entry(struct proc_dir_entry *de);
-
 void proc_init_inodecache(void);
 
 static inline struct pid *proc_pid(struct inode *inode)
@@ -101,8 +99,12 @@ unsigned long task_vsize(struct mm_struct *);
 int task_statm(struct mm_struct *, int *, int *, int *, int *);
 void task_mem(struct seq_file *, struct mm_struct *);
 
-struct proc_dir_entry *de_get(struct proc_dir_entry *de);
-void de_put(struct proc_dir_entry *de);
+static inline struct proc_dir_entry *pde_get(struct proc_dir_entry *pde)
+{
+	atomic_inc(&pde->count);
+	return pde;
+}
+void pde_put(struct proc_dir_entry *pde);
 
 extern struct vfsmount *proc_mnt;
 int proc_fill_super(struct super_block *);
diff --git a/fs/proc/page.c b/fs/proc/page.c
index 5033ce0d254b..180cf5a0bd67 100644
--- a/fs/proc/page.c
+++ b/fs/proc/page.c
@@ -8,6 +8,7 @@
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/hugetlb.h>
+#include <linux/kernel-page-flags.h>
 #include <asm/uaccess.h>
 #include "internal.h"
 
@@ -71,52 +72,12 @@ static const struct file_operations proc_kpagecount_operations = {
  * physical page flags.
  */
 
-/* These macros are used to decouple internal flags from exported ones */
-
-#define KPF_LOCKED		0
-#define KPF_ERROR		1
-#define KPF_REFERENCED		2
-#define KPF_UPTODATE		3
-#define KPF_DIRTY		4
-#define KPF_LRU			5
-#define KPF_ACTIVE		6
-#define KPF_SLAB		7
-#define KPF_WRITEBACK		8
-#define KPF_RECLAIM		9
-#define KPF_BUDDY		10
-
-/* 11-20: new additions in 2.6.31 */
-#define KPF_MMAP		11
-#define KPF_ANON		12
-#define KPF_SWAPCACHE		13
-#define KPF_SWAPBACKED		14
-#define KPF_COMPOUND_HEAD	15
-#define KPF_COMPOUND_TAIL	16
-#define KPF_HUGE		17
-#define KPF_UNEVICTABLE		18
-#define KPF_HWPOISON		19
-#define KPF_NOPAGE		20
-
-#define KPF_KSM			21
-
-/* kernel hacking assistances
- * WARNING: subject to change, never rely on them!
- */
-#define KPF_RESERVED		32
-#define KPF_MLOCKED		33
-#define KPF_MAPPEDTODISK	34
-#define KPF_PRIVATE		35
-#define KPF_PRIVATE_2		36
-#define KPF_OWNER_PRIVATE	37
-#define KPF_ARCH		38
-#define KPF_UNCACHED		39
-
 static inline u64 kpf_copy_bit(u64 kflags, int ubit, int kbit)
 {
 	return ((kflags >> kbit) & 1) << ubit;
 }
 
-static u64 get_uflags(struct page *page)
+u64 stable_page_flags(struct page *page)
 {
 	u64 k;
 	u64 u;
@@ -219,7 +180,7 @@ static ssize_t kpageflags_read(struct file *file, char __user *buf,
 		else
 			ppage = NULL;
 
-		if (put_user(get_uflags(ppage), out)) {
+		if (put_user(stable_page_flags(ppage), out)) {
 			ret = -EFAULT;
 			break;
 		}
diff --git a/fs/proc/proc_devtree.c b/fs/proc/proc_devtree.c
index 7ba79a54948c..123257bb356b 100644
--- a/fs/proc/proc_devtree.c
+++ b/fs/proc/proc_devtree.c
@@ -7,6 +7,7 @@
 #include <linux/init.h>
 #include <linux/time.h>
 #include <linux/proc_fs.h>
+#include <linux/seq_file.h>
 #include <linux/stat.h>
 #include <linux/string.h>
 #include <asm/prom.h>
@@ -25,26 +26,27 @@ static struct proc_dir_entry *proc_device_tree;
 /*
  * Supply data on a read from /proc/device-tree/node/property.
  */
-static int property_read_proc(char *page, char **start, off_t off,
-			      int count, int *eof, void *data)
+static int property_proc_show(struct seq_file *m, void *v)
 {
-	struct property *pp = data;
-	int n;
+	struct property *pp = m->private;
 
-	if (off >= pp->length) {
-		*eof = 1;
-		return 0;
-	}
-	n = pp->length - off;
-	if (n > count)
-		n = count;
-	else
-		*eof = 1;
-	memcpy(page, (char *)pp->value + off, n);
-	*start = page;
-	return n;
+	seq_write(m, pp->value, pp->length);
+	return 0;
+}
+
+static int property_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, property_proc_show, PDE(inode)->data);
 }
 
+static const struct file_operations property_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= property_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
 /*
  * For a node with a name like "gc@10", we make symlinks called "gc"
  * and "@10" to it.
@@ -63,10 +65,9 @@ __proc_device_tree_add_prop(struct proc_dir_entry *de, struct property *pp,
 	 * Unfortunately proc_register puts each new entry
 	 * at the beginning of the list.  So we rearrange them.
 	 */
-	ent = create_proc_read_entry(name,
-				     strncmp(name, "security-", 9)
-				     ? S_IRUGO : S_IRUSR, de,
-				     property_read_proc, pp);
+	ent = proc_create_data(name,
+			       strncmp(name, "security-", 9) ? S_IRUGO : S_IRUSR,
+			       de, &property_proc_fops, pp);
 	if (ent == NULL)
 		return NULL;
 
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 2a1bef9203c6..f277c4a111cb 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -361,12 +361,11 @@ static int smaps_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
 		if (!pte_present(ptent))
 			continue;
 
-		mss->resident += PAGE_SIZE;
-
 		page = vm_normal_page(vma, addr, ptent);
 		if (!page)
 			continue;
 
+		mss->resident += PAGE_SIZE;
 		/* Accumulate the size in pages that have been accessed. */
 		if (pte_young(ptent) || PageReferenced(page))
 			mss->referenced += PAGE_SIZE;
@@ -650,6 +649,50 @@ static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
 	return err;
 }
 
+static u64 huge_pte_to_pagemap_entry(pte_t pte, int offset)
+{
+	u64 pme = 0;
+	if (pte_present(pte))
+		pme = PM_PFRAME(pte_pfn(pte) + offset)
+			| PM_PSHIFT(PAGE_SHIFT) | PM_PRESENT;
+	return pme;
+}
+
+static int pagemap_hugetlb_range(pte_t *pte, unsigned long addr,
+				 unsigned long end, struct mm_walk *walk)
+{
+	struct vm_area_struct *vma;
+	struct pagemapread *pm = walk->private;
+	struct hstate *hs = NULL;
+	int err = 0;
+
+	vma = find_vma(walk->mm, addr);
+	if (vma)
+		hs = hstate_vma(vma);
+	for (; addr != end; addr += PAGE_SIZE) {
+		u64 pfn = PM_NOT_PRESENT;
+
+		if (vma && (addr >= vma->vm_end)) {
+			vma = find_vma(walk->mm, addr);
+			if (vma)
+				hs = hstate_vma(vma);
+		}
+
+		if (vma && (vma->vm_start <= addr) && is_vm_hugetlb_page(vma)) {
+			/* calculate pfn of the "raw" page in the hugepage. */
+			int offset = (addr & ~huge_page_mask(hs)) >> PAGE_SHIFT;
+			pfn = huge_pte_to_pagemap_entry(*pte, offset);
+		}
+		err = add_to_pagemap(addr, pfn, pm);
+		if (err)
+			return err;
+	}
+
+	cond_resched();
+
+	return err;
+}
+
 /*
  * /proc/pid/pagemap - an array mapping virtual pages to pfns
  *
@@ -742,6 +785,7 @@ static ssize_t pagemap_read(struct file *file, char __user *buf,
 
 	pagemap_walk.pmd_entry = pagemap_pte_range;
 	pagemap_walk.pte_hole = pagemap_pte_hole;
+	pagemap_walk.hugetlb_entry = pagemap_hugetlb_range;
 	pagemap_walk.mm = mm;
 	pagemap_walk.private = &pm;
 
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index 8f5c05d3dbd3..5d9fd64ef81a 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -110,9 +110,13 @@ int task_statm(struct mm_struct *mm, int *shared, int *text,
 		}
 	}
 
-	size += (*text = mm->end_code - mm->start_code);
-	size += (*data = mm->start_stack - mm->start_data);
+	*text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK))
+		>> PAGE_SHIFT;
+	*data = (PAGE_ALIGN(mm->start_stack) - (mm->start_data & PAGE_MASK))
+		>> PAGE_SHIFT;
 	up_read(&mm->mmap_sem);
+	size >>= PAGE_SHIFT;
+	size += *text + *data;
 	*resident = size;
 	return size;
 }
diff --git a/fs/qnx4/bitmap.c b/fs/qnx4/bitmap.c
index 32f5d131a644..22e0d60e53ef 100644
--- a/fs/qnx4/bitmap.c
+++ b/fs/qnx4/bitmap.c
@@ -17,13 +17,6 @@
 #include <linux/bitops.h>
 #include "qnx4.h"
 
-#if 0
-int qnx4_new_block(struct super_block *sb)
-{
-	return 0;
-}
-#endif  /*  0  */
-
 static void count_bits(register const char *bmPart, register int size,
 		       int *const tf)
 {
@@ -35,22 +28,7 @@ static void count_bits(register const char *bmPart, register int size,
 	}
 	do {
 		b = *bmPart++;
-		if ((b & 1) == 0)
-			tot++;
-		if ((b & 2) == 0)
-			tot++;
-		if ((b & 4) == 0)
-			tot++;
-		if ((b & 8) == 0)
-			tot++;
-		if ((b & 16) == 0)
-			tot++;
-		if ((b & 32) == 0)
-			tot++;
-		if ((b & 64) == 0)
-			tot++;
-		if ((b & 128) == 0)
-			tot++;
+		tot += 8 - hweight8(b);
 		size--;
 	} while (size != 0);
 	*tf = tot;
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index 449f5a66dd34..ebf3440d28ca 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -64,25 +64,7 @@ static struct buffer_head *qnx4_getblk(struct inode *inode, int nr,
 		result = sb_getblk(inode->i_sb, nr);
 		return result;
 	}
-	if (!create) {
-		return NULL;
-	}
-#if 0
-	tmp = qnx4_new_block(inode->i_sb);
-	if (!tmp) {
-		return NULL;
-	}
-	result = sb_getblk(inode->i_sb, tmp);
-	if (tst) {
-		qnx4_free_block(inode->i_sb, tmp);
-		brelse(result);
-		goto repeat;
-	}
-	tst = tmp;
-#endif
-	inode->i_ctime = CURRENT_TIME_SEC;
-	mark_inode_dirty(inode);
-	return result;
+	return NULL;
 }
 
 struct buffer_head *qnx4_bread(struct inode *inode, int block, int create)
@@ -113,8 +95,6 @@ static int qnx4_get_block( struct inode *inode, sector_t iblock, struct buffer_h
 	if ( phys ) {
 		// logical block is before EOF
 		map_bh(bh, inode->i_sb, phys);
-	} else if ( create ) {
-		// to be done.
 	}
 	return 0;
 }
diff --git a/fs/quota/Kconfig b/fs/quota/Kconfig
index 353e78a9ebee..efc02ebb8c70 100644
--- a/fs/quota/Kconfig
+++ b/fs/quota/Kconfig
@@ -46,12 +46,14 @@ config QFMT_V1
 	  format say Y here.
 
 config QFMT_V2
-	tristate "Quota format v2 support"
+	tristate "Quota format vfsv0 and vfsv1 support"
 	depends on QUOTA
 	select QUOTA_TREE
 	help
-	  This quota format allows using quotas with 32-bit UIDs/GIDs. If you
-	  need this functionality say Y here.
+	  This config option enables kernel support for vfsv0 and vfsv1 quota
+	  formats. Both these formats support 32-bit UIDs/GIDs and vfsv1 format
+	  also supports 64-bit inode and block quota limits. If you need this
+	  functionality say Y here.
 
 config QUOTACTL
 	bool
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c
index eb5a755718f6..3fc62b097bed 100644
--- a/fs/quota/dquot.c
+++ b/fs/quota/dquot.c
@@ -323,6 +323,30 @@ int dquot_mark_dquot_dirty(struct dquot *dquot)
 }
 EXPORT_SYMBOL(dquot_mark_dquot_dirty);
 
+/* Dirtify all the dquots - this can block when journalling */
+static inline int mark_all_dquot_dirty(struct dquot * const *dquot)
+{
+	int ret, err, cnt;
+
+	ret = err = 0;
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
+		if (dquot[cnt])
+			/* Even in case of error we have to continue */
+			ret = mark_dquot_dirty(dquot[cnt]);
+		if (!err)
+			err = ret;
+	}
+	return err;
+}
+
+static inline void dqput_all(struct dquot **dquot)
+{
+	unsigned int cnt;
+
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+		dqput(dquot[cnt]);
+}
+
 /* This function needs dq_list_lock */
 static inline int clear_dquot_dirty(struct dquot *dquot)
 {
@@ -1268,8 +1292,7 @@ int dquot_initialize(struct inode *inode, int type)
 out_err:
 	up_write(&sb_dqopt(sb)->dqptr_sem);
 	/* Drop unused references */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		dqput(got[cnt]);
+	dqput_all(got);
 	return ret;
 }
 EXPORT_SYMBOL(dquot_initialize);
@@ -1288,9 +1311,7 @@ int dquot_drop(struct inode *inode)
 		inode->i_dquot[cnt] = NULL;
 	}
 	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
-
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		dqput(put[cnt]);
+	dqput_all(put);
 	return 0;
 }
 EXPORT_SYMBOL(dquot_drop);
@@ -1319,6 +1340,70 @@ void vfs_dq_drop(struct inode *inode)
 EXPORT_SYMBOL(vfs_dq_drop);
 
 /*
+ * inode_reserved_space is managed internally by quota, and protected by
+ * i_lock similar to i_blocks+i_bytes.
+ */
+static qsize_t *inode_reserved_space(struct inode * inode)
+{
+	/* Filesystem must explicitly define it's own method in order to use
+	 * quota reservation interface */
+	BUG_ON(!inode->i_sb->dq_op->get_reserved_space);
+	return inode->i_sb->dq_op->get_reserved_space(inode);
+}
+
+static void inode_add_rsv_space(struct inode *inode, qsize_t number)
+{
+	spin_lock(&inode->i_lock);
+	*inode_reserved_space(inode) += number;
+	spin_unlock(&inode->i_lock);
+}
+
+
+static void inode_claim_rsv_space(struct inode *inode, qsize_t number)
+{
+	spin_lock(&inode->i_lock);
+	*inode_reserved_space(inode) -= number;
+	__inode_add_bytes(inode, number);
+	spin_unlock(&inode->i_lock);
+}
+
+static void inode_sub_rsv_space(struct inode *inode, qsize_t number)
+{
+	spin_lock(&inode->i_lock);
+	*inode_reserved_space(inode) -= number;
+	spin_unlock(&inode->i_lock);
+}
+
+static qsize_t inode_get_rsv_space(struct inode *inode)
+{
+	qsize_t ret;
+
+	if (!inode->i_sb->dq_op->get_reserved_space)
+		return 0;
+	spin_lock(&inode->i_lock);
+	ret = *inode_reserved_space(inode);
+	spin_unlock(&inode->i_lock);
+	return ret;
+}
+
+static void inode_incr_space(struct inode *inode, qsize_t number,
+				int reserve)
+{
+	if (reserve)
+		inode_add_rsv_space(inode, number);
+	else
+		inode_add_bytes(inode, number);
+}
+
+static void inode_decr_space(struct inode *inode, qsize_t number, int reserve)
+{
+	if (reserve)
+		inode_sub_rsv_space(inode, number);
+	else
+		inode_sub_bytes(inode, number);
+}
+
+/*
  * Following four functions update i_blocks+i_bytes fields and
  * quota information (together with appropriate checks)
  * NOTE: We absolutely rely on the fact that caller dirties
@@ -1336,6 +1421,21 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number,
 	int cnt, ret = QUOTA_OK;
 	char warntype[MAXQUOTAS];
 
+	/*
+	 * First test before acquiring mutex - solves deadlocks when we
+	 * re-enter the quota code and are already holding the mutex
+	 */
+	if (IS_NOQUOTA(inode)) {
+		inode_incr_space(inode, number, reserve);
+		goto out;
+	}
+
+	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+	if (IS_NOQUOTA(inode)) {
+		inode_incr_space(inode, number, reserve);
+		goto out_unlock;
+	}
+
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
 		warntype[cnt] = QUOTA_NL_NOWARN;
 
@@ -1346,7 +1446,8 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number,
 		if (check_bdq(inode->i_dquot[cnt], number, warn, warntype+cnt)
 		    == NO_QUOTA) {
 			ret = NO_QUOTA;
-			goto out_unlock;
+			spin_unlock(&dq_data_lock);
+			goto out_flush_warn;
 		}
 	}
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
@@ -1357,64 +1458,29 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number,
 		else
 			dquot_incr_space(inode->i_dquot[cnt], number);
 	}
-	if (!reserve)
-		inode_add_bytes(inode, number);
-out_unlock:
+	inode_incr_space(inode, number, reserve);
 	spin_unlock(&dq_data_lock);
+
+	if (reserve)
+		goto out_flush_warn;
+	mark_all_dquot_dirty(inode->i_dquot);
+out_flush_warn:
 	flush_warnings(inode->i_dquot, warntype);
+out_unlock:
+	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
+out:
 	return ret;
 }
 
 int dquot_alloc_space(struct inode *inode, qsize_t number, int warn)
 {
-	int cnt, ret = QUOTA_OK;
-
-	/*
-	 * First test before acquiring mutex - solves deadlocks when we
-	 * re-enter the quota code and are already holding the mutex
-	 */
-	if (IS_NOQUOTA(inode)) {
-		inode_add_bytes(inode, number);
-		goto out;
-	}
-
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
-	if (IS_NOQUOTA(inode)) {
-		inode_add_bytes(inode, number);
-		goto out_unlock;
-	}
-
-	ret = __dquot_alloc_space(inode, number, warn, 0);
-	if (ret == NO_QUOTA)
-		goto out_unlock;
-
-	/* Dirtify all the dquots - this can block when journalling */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		if (inode->i_dquot[cnt])
-			mark_dquot_dirty(inode->i_dquot[cnt]);
-out_unlock:
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
-out:
-	return ret;
+	return __dquot_alloc_space(inode, number, warn, 0);
 }
 EXPORT_SYMBOL(dquot_alloc_space);
 
 int dquot_reserve_space(struct inode *inode, qsize_t number, int warn)
 {
-	int ret = QUOTA_OK;
-
-	if (IS_NOQUOTA(inode))
-		goto out;
-
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
-	if (IS_NOQUOTA(inode))
-		goto out_unlock;
-
-	ret = __dquot_alloc_space(inode, number, warn, 1);
-out_unlock:
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
-out:
-	return ret;
+	return __dquot_alloc_space(inode, number, warn, 1);
 }
 EXPORT_SYMBOL(dquot_reserve_space);
 
@@ -1455,10 +1521,7 @@ int dquot_alloc_inode(const struct inode *inode, qsize_t number)
 warn_put_all:
 	spin_unlock(&dq_data_lock);
 	if (ret == QUOTA_OK)
-		/* Dirtify all the dquots - this can block when journalling */
-		for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-			if (inode->i_dquot[cnt])
-				mark_dquot_dirty(inode->i_dquot[cnt]);
+		mark_all_dquot_dirty(inode->i_dquot);
 	flush_warnings(inode->i_dquot, warntype);
 	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	return ret;
@@ -1471,14 +1534,14 @@ int dquot_claim_space(struct inode *inode, qsize_t number)
 	int ret = QUOTA_OK;
 
 	if (IS_NOQUOTA(inode)) {
-		inode_add_bytes(inode, number);
+		inode_claim_rsv_space(inode, number);
 		goto out;
 	}
 
 	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	if (IS_NOQUOTA(inode))	{
 		up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
-		inode_add_bytes(inode, number);
+		inode_claim_rsv_space(inode, number);
 		goto out;
 	}
 
@@ -1490,12 +1553,9 @@ int dquot_claim_space(struct inode *inode, qsize_t number)
 							number);
 	}
 	/* Update inode bytes */
-	inode_add_bytes(inode, number);
+	inode_claim_rsv_space(inode, number);
 	spin_unlock(&dq_data_lock);
-	/* Dirtify all the dquots - this can block when journalling */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		if (inode->i_dquot[cnt])
-			mark_dquot_dirty(inode->i_dquot[cnt]);
+	mark_all_dquot_dirty(inode->i_dquot);
 	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 out:
 	return ret;
@@ -1503,38 +1563,9 @@ out:
 EXPORT_SYMBOL(dquot_claim_space);
 
 /*
- * Release reserved quota space
- */
-void dquot_release_reserved_space(struct inode *inode, qsize_t number)
-{
-	int cnt;
-
-	if (IS_NOQUOTA(inode))
-		goto out;
-
-	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
-	if (IS_NOQUOTA(inode))
-		goto out_unlock;
-
-	spin_lock(&dq_data_lock);
-	/* Release reserved dquots */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		if (inode->i_dquot[cnt])
-			dquot_free_reserved_space(inode->i_dquot[cnt], number);
-	}
-	spin_unlock(&dq_data_lock);
-
-out_unlock:
-	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
-out:
-	return;
-}
-EXPORT_SYMBOL(dquot_release_reserved_space);
-
-/*
  * This operation can block, but only after everything is updated
  */
-int dquot_free_space(struct inode *inode, qsize_t number)
+int __dquot_free_space(struct inode *inode, qsize_t number, int reserve)
 {
 	unsigned int cnt;
 	char warntype[MAXQUOTAS];
@@ -1543,7 +1574,7 @@ int dquot_free_space(struct inode *inode, qsize_t number)
          * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode)) {
 out_sub:
-		inode_sub_bytes(inode, number);
+		inode_decr_space(inode, number, reserve);
 		return QUOTA_OK;
 	}
 
@@ -1558,21 +1589,40 @@ out_sub:
 		if (!inode->i_dquot[cnt])
 			continue;
 		warntype[cnt] = info_bdq_free(inode->i_dquot[cnt], number);
-		dquot_decr_space(inode->i_dquot[cnt], number);
+		if (reserve)
+			dquot_free_reserved_space(inode->i_dquot[cnt], number);
+		else
+			dquot_decr_space(inode->i_dquot[cnt], number);
 	}
-	inode_sub_bytes(inode, number);
+	inode_decr_space(inode, number, reserve);
 	spin_unlock(&dq_data_lock);
-	/* Dirtify all the dquots - this can block when journalling */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		if (inode->i_dquot[cnt])
-			mark_dquot_dirty(inode->i_dquot[cnt]);
+
+	if (reserve)
+		goto out_unlock;
+	mark_all_dquot_dirty(inode->i_dquot);
+out_unlock:
 	flush_warnings(inode->i_dquot, warntype);
 	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	return QUOTA_OK;
 }
+
+int dquot_free_space(struct inode *inode, qsize_t number)
+{
+	return  __dquot_free_space(inode, number, 0);
+}
 EXPORT_SYMBOL(dquot_free_space);
 
 /*
+ * Release reserved quota space
+ */
+void dquot_release_reserved_space(struct inode *inode, qsize_t number)
+{
+	__dquot_free_space(inode, number, 1);
+
+}
+EXPORT_SYMBOL(dquot_release_reserved_space);
+
+/*
  * This operation can block, but only after everything is updated
  */
 int dquot_free_inode(const struct inode *inode, qsize_t number)
@@ -1599,10 +1649,7 @@ int dquot_free_inode(const struct inode *inode, qsize_t number)
 		dquot_decr_inodes(inode->i_dquot[cnt], number);
 	}
 	spin_unlock(&dq_data_lock);
-	/* Dirtify all the dquots - this can block when journalling */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-		if (inode->i_dquot[cnt])
-			mark_dquot_dirty(inode->i_dquot[cnt]);
+	mark_all_dquot_dirty(inode->i_dquot);
 	flush_warnings(inode->i_dquot, warntype);
 	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	return QUOTA_OK;
@@ -1610,19 +1657,6 @@ int dquot_free_inode(const struct inode *inode, qsize_t number)
 EXPORT_SYMBOL(dquot_free_inode);
 
 /*
- * call back function, get reserved quota space from underlying fs
- */
-qsize_t dquot_get_reserved_space(struct inode *inode)
-{
-	qsize_t reserved_space = 0;
-
-	if (sb_any_quota_active(inode->i_sb) &&
-	    inode->i_sb->dq_op->get_reserved_space)
-		reserved_space = inode->i_sb->dq_op->get_reserved_space(inode);
-	return reserved_space;
-}
-
-/*
  * Transfer the number of inode and blocks from one diskquota to an other.
  *
  * This operation can block, but only after everything is updated
@@ -1665,7 +1699,7 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 	}
 	spin_lock(&dq_data_lock);
 	cur_space = inode_get_bytes(inode);
-	rsv_space = dquot_get_reserved_space(inode);
+	rsv_space = inode_get_rsv_space(inode);
 	space = cur_space + rsv_space;
 	/* Build the transfer_from list and check the limits */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
@@ -1709,25 +1743,18 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 	spin_unlock(&dq_data_lock);
 	up_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 
-	/* Dirtify all the dquots - this can block when journalling */
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		if (transfer_from[cnt])
-			mark_dquot_dirty(transfer_from[cnt]);
-		if (transfer_to[cnt]) {
-			mark_dquot_dirty(transfer_to[cnt]);
-			/* The reference we got is transferred to the inode */
-			transfer_to[cnt] = NULL;
-		}
-	}
+	mark_all_dquot_dirty(transfer_from);
+	mark_all_dquot_dirty(transfer_to);
+	/* The reference we got is transferred to the inode */
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+		transfer_to[cnt] = NULL;
 warn_put_all:
 	flush_warnings(transfer_to, warntype_to);
 	flush_warnings(transfer_from, warntype_from_inodes);
 	flush_warnings(transfer_from, warntype_from_space);
 put_all:
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		dqput(transfer_from[cnt]);
-		dqput(transfer_to[cnt]);
-	}
+	dqput_all(transfer_from);
+	dqput_all(transfer_to);
 	return ret;
 over_quota:
 	spin_unlock(&dq_data_lock);
@@ -2164,7 +2191,9 @@ int vfs_quota_on_mount(struct super_block *sb, char *qf_name,
 	struct dentry *dentry;
 	int error;
 
+	mutex_lock(&sb->s_root->d_inode->i_mutex);
 	dentry = lookup_one_len(qf_name, sb->s_root, strlen(qf_name));
+	mutex_unlock(&sb->s_root->d_inode->i_mutex);
 	if (IS_ERR(dentry))
 		return PTR_ERR(dentry);
 
diff --git a/fs/quota/quota_v1.c b/fs/quota/quota_v1.c
index 0edcf42b1778..2ae757e9c008 100644
--- a/fs/quota/quota_v1.c
+++ b/fs/quota/quota_v1.c
@@ -204,7 +204,7 @@ out:
 	return ret;
 }
 
-static struct quota_format_ops v1_format_ops = {
+static const struct quota_format_ops v1_format_ops = {
 	.check_quota_file	= v1_check_quota_file,
 	.read_file_info		= v1_read_file_info,
 	.write_file_info	= v1_write_file_info,
diff --git a/fs/quota/quota_v2.c b/fs/quota/quota_v2.c
index a5475fb1ae44..e3da02f4986f 100644
--- a/fs/quota/quota_v2.c
+++ b/fs/quota/quota_v2.c
@@ -23,14 +23,23 @@ MODULE_LICENSE("GPL");
 
 #define __QUOTA_V2_PARANOIA
 
-static void v2_mem2diskdqb(void *dp, struct dquot *dquot);
-static void v2_disk2memdqb(struct dquot *dquot, void *dp);
-static int v2_is_id(void *dp, struct dquot *dquot);
-
-static struct qtree_fmt_operations v2_qtree_ops = {
-	.mem2disk_dqblk = v2_mem2diskdqb,
-	.disk2mem_dqblk = v2_disk2memdqb,
-	.is_id = v2_is_id,
+static void v2r0_mem2diskdqb(void *dp, struct dquot *dquot);
+static void v2r0_disk2memdqb(struct dquot *dquot, void *dp);
+static int v2r0_is_id(void *dp, struct dquot *dquot);
+static void v2r1_mem2diskdqb(void *dp, struct dquot *dquot);
+static void v2r1_disk2memdqb(struct dquot *dquot, void *dp);
+static int v2r1_is_id(void *dp, struct dquot *dquot);
+
+static struct qtree_fmt_operations v2r0_qtree_ops = {
+	.mem2disk_dqblk = v2r0_mem2diskdqb,
+	.disk2mem_dqblk = v2r0_disk2memdqb,
+	.is_id = v2r0_is_id,
+};
+
+static struct qtree_fmt_operations v2r1_qtree_ops = {
+	.mem2disk_dqblk = v2r1_mem2diskdqb,
+	.disk2mem_dqblk = v2r1_disk2memdqb,
+	.is_id = v2r1_is_id,
 };
 
 #define QUOTABLOCK_BITS 10
@@ -46,23 +55,33 @@ static inline qsize_t v2_qbtos(qsize_t blocks)
 	return blocks << QUOTABLOCK_BITS;
 }
 
+static int v2_read_header(struct super_block *sb, int type,
+			  struct v2_disk_dqheader *dqhead)
+{
+	ssize_t size;
+
+	size = sb->s_op->quota_read(sb, type, (char *)dqhead,
+				    sizeof(struct v2_disk_dqheader), 0);
+	if (size != sizeof(struct v2_disk_dqheader)) {
+		printk(KERN_WARNING "quota_v2: Failed header read:"
+		       " expected=%zd got=%zd\n",
+			sizeof(struct v2_disk_dqheader), size);
+		return 0;
+	}
+	return 1;
+}
+
 /* Check whether given file is really vfsv0 quotafile */
 static int v2_check_quota_file(struct super_block *sb, int type)
 {
 	struct v2_disk_dqheader dqhead;
-	ssize_t size;
 	static const uint quota_magics[] = V2_INITQMAGICS;
 	static const uint quota_versions[] = V2_INITQVERSIONS;
  
-	size = sb->s_op->quota_read(sb, type, (char *)&dqhead,
-				    sizeof(struct v2_disk_dqheader), 0);
-	if (size != sizeof(struct v2_disk_dqheader)) {
-		printk("quota_v2: failed read expected=%zd got=%zd\n",
-			sizeof(struct v2_disk_dqheader), size);
+	if (!v2_read_header(sb, type, &dqhead))
 		return 0;
-	}
 	if (le32_to_cpu(dqhead.dqh_magic) != quota_magics[type] ||
-	    le32_to_cpu(dqhead.dqh_version) != quota_versions[type])
+	    le32_to_cpu(dqhead.dqh_version) > quota_versions[type])
 		return 0;
 	return 1;
 }
@@ -71,14 +90,23 @@ static int v2_check_quota_file(struct super_block *sb, int type)
 static int v2_read_file_info(struct super_block *sb, int type)
 {
 	struct v2_disk_dqinfo dinfo;
+	struct v2_disk_dqheader dqhead;
 	struct mem_dqinfo *info = sb_dqinfo(sb, type);
 	struct qtree_mem_dqinfo *qinfo;
 	ssize_t size;
+	unsigned int version;
+
+	if (!v2_read_header(sb, type, &dqhead))
+		return -1;
+	version = le32_to_cpu(dqhead.dqh_version);
+	if ((info->dqi_fmt_id == QFMT_VFS_V0 && version != 0) ||
+	    (info->dqi_fmt_id == QFMT_VFS_V1 && version != 1))
+		return -1;
 
 	size = sb->s_op->quota_read(sb, type, (char *)&dinfo,
 	       sizeof(struct v2_disk_dqinfo), V2_DQINFOOFF);
 	if (size != sizeof(struct v2_disk_dqinfo)) {
-		printk(KERN_WARNING "Can't read info structure on device %s.\n",
+		printk(KERN_WARNING "quota_v2: Can't read info structure on device %s.\n",
 			sb->s_id);
 		return -1;
 	}
@@ -89,9 +117,15 @@ static int v2_read_file_info(struct super_block *sb, int type)
 		return -1;
 	}
 	qinfo = info->dqi_priv;
-	/* limits are stored as unsigned 32-bit data */
-	info->dqi_maxblimit = 0xffffffff;
-	info->dqi_maxilimit = 0xffffffff;
+	if (version == 0) {
+		/* limits are stored as unsigned 32-bit data */
+		info->dqi_maxblimit = 0xffffffff;
+		info->dqi_maxilimit = 0xffffffff;
+	} else {
+		/* used space is stored as unsigned 64-bit value */
+		info->dqi_maxblimit = 0xffffffffffffffffULL;	/* 2^64-1 */
+		info->dqi_maxilimit = 0xffffffffffffffffULL;
+	}
 	info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace);
 	info->dqi_igrace = le32_to_cpu(dinfo.dqi_igrace);
 	info->dqi_flags = le32_to_cpu(dinfo.dqi_flags);
@@ -103,8 +137,13 @@ static int v2_read_file_info(struct super_block *sb, int type)
 	qinfo->dqi_blocksize_bits = V2_DQBLKSIZE_BITS;
 	qinfo->dqi_usable_bs = 1 << V2_DQBLKSIZE_BITS;
 	qinfo->dqi_qtree_depth = qtree_depth(qinfo);
-	qinfo->dqi_entry_size = sizeof(struct v2_disk_dqblk);
-	qinfo->dqi_ops = &v2_qtree_ops;
+	if (version == 0) {
+		qinfo->dqi_entry_size = sizeof(struct v2r0_disk_dqblk);
+		qinfo->dqi_ops = &v2r0_qtree_ops;
+	} else {
+		qinfo->dqi_entry_size = sizeof(struct v2r1_disk_dqblk);
+		qinfo->dqi_ops = &v2r1_qtree_ops;
+	}
 	return 0;
 }
 
@@ -135,9 +174,9 @@ static int v2_write_file_info(struct super_block *sb, int type)
 	return 0;
 }
 
-static void v2_disk2memdqb(struct dquot *dquot, void *dp)
+static void v2r0_disk2memdqb(struct dquot *dquot, void *dp)
 {
-	struct v2_disk_dqblk *d = dp, empty;
+	struct v2r0_disk_dqblk *d = dp, empty;
 	struct mem_dqblk *m = &dquot->dq_dqb;
 
 	m->dqb_ihardlimit = le32_to_cpu(d->dqb_ihardlimit);
@@ -149,15 +188,15 @@ static void v2_disk2memdqb(struct dquot *dquot, void *dp)
 	m->dqb_curspace = le64_to_cpu(d->dqb_curspace);
 	m->dqb_btime = le64_to_cpu(d->dqb_btime);
 	/* We need to escape back all-zero structure */
-	memset(&empty, 0, sizeof(struct v2_disk_dqblk));
+	memset(&empty, 0, sizeof(struct v2r0_disk_dqblk));
 	empty.dqb_itime = cpu_to_le64(1);
-	if (!memcmp(&empty, dp, sizeof(struct v2_disk_dqblk)))
+	if (!memcmp(&empty, dp, sizeof(struct v2r0_disk_dqblk)))
 		m->dqb_itime = 0;
 }
 
-static void v2_mem2diskdqb(void *dp, struct dquot *dquot)
+static void v2r0_mem2diskdqb(void *dp, struct dquot *dquot)
 {
-	struct v2_disk_dqblk *d = dp;
+	struct v2r0_disk_dqblk *d = dp;
 	struct mem_dqblk *m = &dquot->dq_dqb;
 	struct qtree_mem_dqinfo *info =
 			sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
@@ -175,9 +214,60 @@ static void v2_mem2diskdqb(void *dp, struct dquot *dquot)
 		d->dqb_itime = cpu_to_le64(1);
 }
 
-static int v2_is_id(void *dp, struct dquot *dquot)
+static int v2r0_is_id(void *dp, struct dquot *dquot)
 {
-	struct v2_disk_dqblk *d = dp;
+	struct v2r0_disk_dqblk *d = dp;
+	struct qtree_mem_dqinfo *info =
+			sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
+
+	if (qtree_entry_unused(info, dp))
+		return 0;
+	return le32_to_cpu(d->dqb_id) == dquot->dq_id;
+}
+
+static void v2r1_disk2memdqb(struct dquot *dquot, void *dp)
+{
+	struct v2r1_disk_dqblk *d = dp, empty;
+	struct mem_dqblk *m = &dquot->dq_dqb;
+
+	m->dqb_ihardlimit = le64_to_cpu(d->dqb_ihardlimit);
+	m->dqb_isoftlimit = le64_to_cpu(d->dqb_isoftlimit);
+	m->dqb_curinodes = le64_to_cpu(d->dqb_curinodes);
+	m->dqb_itime = le64_to_cpu(d->dqb_itime);
+	m->dqb_bhardlimit = v2_qbtos(le64_to_cpu(d->dqb_bhardlimit));
+	m->dqb_bsoftlimit = v2_qbtos(le64_to_cpu(d->dqb_bsoftlimit));
+	m->dqb_curspace = le64_to_cpu(d->dqb_curspace);
+	m->dqb_btime = le64_to_cpu(d->dqb_btime);
+	/* We need to escape back all-zero structure */
+	memset(&empty, 0, sizeof(struct v2r1_disk_dqblk));
+	empty.dqb_itime = cpu_to_le64(1);
+	if (!memcmp(&empty, dp, sizeof(struct v2r1_disk_dqblk)))
+		m->dqb_itime = 0;
+}
+
+static void v2r1_mem2diskdqb(void *dp, struct dquot *dquot)
+{
+	struct v2r1_disk_dqblk *d = dp;
+	struct mem_dqblk *m = &dquot->dq_dqb;
+	struct qtree_mem_dqinfo *info =
+			sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
+
+	d->dqb_ihardlimit = cpu_to_le64(m->dqb_ihardlimit);
+	d->dqb_isoftlimit = cpu_to_le64(m->dqb_isoftlimit);
+	d->dqb_curinodes = cpu_to_le64(m->dqb_curinodes);
+	d->dqb_itime = cpu_to_le64(m->dqb_itime);
+	d->dqb_bhardlimit = cpu_to_le64(v2_stoqb(m->dqb_bhardlimit));
+	d->dqb_bsoftlimit = cpu_to_le64(v2_stoqb(m->dqb_bsoftlimit));
+	d->dqb_curspace = cpu_to_le64(m->dqb_curspace);
+	d->dqb_btime = cpu_to_le64(m->dqb_btime);
+	d->dqb_id = cpu_to_le32(dquot->dq_id);
+	if (qtree_entry_unused(info, dp))
+		d->dqb_itime = cpu_to_le64(1);
+}
+
+static int v2r1_is_id(void *dp, struct dquot *dquot)
+{
+	struct v2r1_disk_dqblk *d = dp;
 	struct qtree_mem_dqinfo *info =
 			sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv;
 
@@ -207,7 +297,7 @@ static int v2_free_file_info(struct super_block *sb, int type)
 	return 0;
 }
 
-static struct quota_format_ops v2_format_ops = {
+static const struct quota_format_ops v2_format_ops = {
 	.check_quota_file	= v2_check_quota_file,
 	.read_file_info		= v2_read_file_info,
 	.write_file_info	= v2_write_file_info,
@@ -217,20 +307,32 @@ static struct quota_format_ops v2_format_ops = {
 	.release_dqblk		= v2_release_dquot,
 };
 
-static struct quota_format_type v2_quota_format = {
+static struct quota_format_type v2r0_quota_format = {
 	.qf_fmt_id	= QFMT_VFS_V0,
 	.qf_ops		= &v2_format_ops,
 	.qf_owner	= THIS_MODULE
 };
 
+static struct quota_format_type v2r1_quota_format = {
+	.qf_fmt_id	= QFMT_VFS_V1,
+	.qf_ops		= &v2_format_ops,
+	.qf_owner	= THIS_MODULE
+};
+
 static int __init init_v2_quota_format(void)
 {
-	return register_quota_format(&v2_quota_format);
+	int ret;
+
+	ret = register_quota_format(&v2r0_quota_format);
+	if (ret)
+		return ret;
+	return register_quota_format(&v2r1_quota_format);
 }
 
 static void __exit exit_v2_quota_format(void)
 {
-	unregister_quota_format(&v2_quota_format);
+	unregister_quota_format(&v2r0_quota_format);
+	unregister_quota_format(&v2r1_quota_format);
 }
 
 module_init(init_v2_quota_format);
diff --git a/fs/quota/quotaio_v2.h b/fs/quota/quotaio_v2.h
index 530fe580685c..f1966b42c2fd 100644
--- a/fs/quota/quotaio_v2.h
+++ b/fs/quota/quotaio_v2.h
@@ -17,8 +17,8 @@
 }
 
 #define V2_INITQVERSIONS {\
-	0,		/* USRQUOTA */\
-	0		/* GRPQUOTA */\
+	1,		/* USRQUOTA */\
+	1		/* GRPQUOTA */\
 }
 
 /* First generic header */
@@ -32,7 +32,7 @@ struct v2_disk_dqheader {
  * (as it appears on disk) - the file is a radix tree whose leaves point
  * to blocks of these structures.
  */
-struct v2_disk_dqblk {
+struct v2r0_disk_dqblk {
 	__le32 dqb_id;		/* id this quota applies to */
 	__le32 dqb_ihardlimit;	/* absolute limit on allocated inodes */
 	__le32 dqb_isoftlimit;	/* preferred inode limit */
@@ -44,6 +44,19 @@ struct v2_disk_dqblk {
 	__le64 dqb_itime;	/* time limit for excessive inode use */
 };
 
+struct v2r1_disk_dqblk {
+	__le32 dqb_id;		/* id this quota applies to */
+	__le32 dqb_pad;
+	__le64 dqb_ihardlimit;	/* absolute limit on allocated inodes */
+	__le64 dqb_isoftlimit;	/* preferred inode limit */
+	__le64 dqb_curinodes;	/* current # allocated inodes */
+	__le64 dqb_bhardlimit;	/* absolute limit on disk space (in QUOTABLOCK_SIZE) */
+	__le64 dqb_bsoftlimit;	/* preferred limit on disk space (in QUOTABLOCK_SIZE) */
+	__le64 dqb_curspace;	/* current space occupied (in bytes) */
+	__le64 dqb_btime;	/* time limit for excessive disk use */
+	__le64 dqb_itime;	/* time limit for excessive inode use */
+};
+
 /* Header with type and version specific information */
 struct v2_disk_dqinfo {
 	__le32 dqi_bgrace;	/* Time before block soft limit becomes hard limit */
diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c
index 32fae4040ebf..1739a4aba25f 100644
--- a/fs/ramfs/file-nommu.c
+++ b/fs/ramfs/file-nommu.c
@@ -60,7 +60,7 @@ const struct inode_operations ramfs_file_inode_operations = {
  */
 int ramfs_nommu_expand_for_mapping(struct inode *inode, size_t newsize)
 {
-	unsigned long npages, xpages, loop, limit;
+	unsigned long npages, xpages, loop;
 	struct page *pages;
 	unsigned order;
 	void *data;
@@ -123,30 +123,6 @@ add_error:
 
 /*****************************************************************************/
 /*
- * check that file shrinkage doesn't leave any VMAs dangling in midair
- */
-static int ramfs_nommu_check_mappings(struct inode *inode,
-				      size_t newsize, size_t size)
-{
-	struct vm_area_struct *vma;
-	struct prio_tree_iter iter;
-
-	/* search for VMAs that fall within the dead zone */
-	vma_prio_tree_foreach(vma, &iter, &inode->i_mapping->i_mmap,
-			      newsize >> PAGE_SHIFT,
-			      (size + PAGE_SIZE - 1) >> PAGE_SHIFT
-			      ) {
-		/* found one - only interested if it's shared out of the page
-		 * cache */
-		if (vma->vm_flags & VM_SHARED)
-			return -ETXTBSY; /* not quite true, but near enough */
-	}
-
-	return 0;
-}
-
-/*****************************************************************************/
-/*
  *
  */
 static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size)
@@ -164,7 +140,7 @@ static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size)
 
 	/* check that a decrease in size doesn't cut off any shared mappings */
 	if (newsize < size) {
-		ret = ramfs_nommu_check_mappings(inode, newsize, size);
+		ret = nommu_shrink_inode_mappings(inode, size, newsize);
 		if (ret < 0)
 			return ret;
 	}
diff --git a/fs/reiserfs/Makefile b/fs/reiserfs/Makefile
index 6a9e30c041dd..792b3cb2cd18 100644
--- a/fs/reiserfs/Makefile
+++ b/fs/reiserfs/Makefile
@@ -7,7 +7,11 @@ obj-$(CONFIG_REISERFS_FS) += reiserfs.o
 reiserfs-objs := bitmap.o do_balan.o namei.o inode.o file.o dir.o fix_node.o \
 		 super.o prints.o objectid.o lbalance.o ibalance.o stree.o \
 		 hashes.o tail_conversion.o journal.o resize.o \
-		 item_ops.o ioctl.o procfs.o xattr.o lock.o
+		 item_ops.o ioctl.o xattr.o lock.o
+
+ifeq ($(CONFIG_REISERFS_PROC_INFO),y)
+reiserfs-objs += procfs.o
+endif
 
 ifeq ($(CONFIG_REISERFS_FS_XATTR),y)
 reiserfs-objs += xattr_user.o xattr_trusted.o
diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c
index 685495707181..65c872761177 100644
--- a/fs/reiserfs/bitmap.c
+++ b/fs/reiserfs/bitmap.c
@@ -1277,7 +1277,10 @@ int reiserfs_init_bitmap_cache(struct super_block *sb)
 	struct reiserfs_bitmap_info *bitmap;
 	unsigned int bmap_nr = reiserfs_bmap_count(sb);
 
+	/* Avoid lock recursion in fault case */
+	reiserfs_write_unlock(sb);
 	bitmap = vmalloc(sizeof(*bitmap) * bmap_nr);
+	reiserfs_write_lock(sb);
 	if (bitmap == NULL)
 		return -ENOMEM;
 
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 3a28e7751b3c..9087b10209e6 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -31,11 +31,12 @@ void reiserfs_delete_inode(struct inode *inode)
 	    JOURNAL_PER_BALANCE_CNT * 2 +
 	    2 * REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb);
 	struct reiserfs_transaction_handle th;
+	int depth;
 	int err;
 
 	truncate_inode_pages(&inode->i_data, 0);
 
-	reiserfs_write_lock(inode->i_sb);
+	depth = reiserfs_write_lock_once(inode->i_sb);
 
 	/* The = 0 happens when we abort creating a new inode for some reason like lack of space.. */
 	if (!(inode->i_state & I_NEW) && INODE_PKEY(inode)->k_objectid != 0) {	/* also handles bad_inode case */
@@ -74,7 +75,7 @@ void reiserfs_delete_inode(struct inode *inode)
       out:
 	clear_inode(inode);	/* note this must go after the journal_end to prevent deadlock */
 	inode->i_blocks = 0;
-	reiserfs_write_unlock(inode->i_sb);
+	reiserfs_write_unlock_once(inode->i_sb, depth);
 }
 
 static void _make_cpu_key(struct cpu_key *key, int version, __u32 dirid,
@@ -2538,6 +2539,12 @@ static int reiserfs_writepage(struct page *page, struct writeback_control *wbc)
 	return reiserfs_write_full_page(page, wbc);
 }
 
+static void reiserfs_truncate_failed_write(struct inode *inode)
+{
+	truncate_inode_pages(inode->i_mapping, inode->i_size);
+	reiserfs_truncate_file(inode, 0);
+}
+
 static int reiserfs_write_begin(struct file *file,
 				struct address_space *mapping,
 				loff_t pos, unsigned len, unsigned flags,
@@ -2604,6 +2611,8 @@ static int reiserfs_write_begin(struct file *file,
 	if (ret) {
 		unlock_page(page);
 		page_cache_release(page);
+		/* Truncate allocated blocks */
+		reiserfs_truncate_failed_write(inode);
 	}
 	return ret;
 }
@@ -2701,9 +2710,7 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 	 ** transaction tracking stuff when the size changes.  So, we have
 	 ** to do the i_size updates here.
 	 */
-	pos += copied;
-
-	if (pos > inode->i_size) {
+	if (pos + copied > inode->i_size) {
 		struct reiserfs_transaction_handle myth;
 		lock_depth = reiserfs_write_lock_once(inode->i_sb);
 		locked = true;
@@ -2721,7 +2728,7 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 			goto journal_error;
 
 		reiserfs_update_inode_transaction(inode);
-		inode->i_size = pos;
+		inode->i_size = pos + copied;
 		/*
 		 * this will just nest into our transaction.  It's important
 		 * to use mark_inode_dirty so the inode gets pushed around on the
@@ -2751,6 +2758,10 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
 	unlock_page(page);
 	page_cache_release(page);
+
+	if (pos + len > inode->i_size)
+		reiserfs_truncate_failed_write(inode);
+
 	return ret == 0 ? copied : ret;
 
       journal_error:
@@ -3051,13 +3062,14 @@ static ssize_t reiserfs_direct_IO(int rw, struct kiocb *iocb,
 int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = dentry->d_inode;
-	int error;
 	unsigned int ia_valid;
+	int depth;
+	int error;
 
 	/* must be turned off for recursive notify_change calls */
 	ia_valid = attr->ia_valid &= ~(ATTR_KILL_SUID|ATTR_KILL_SGID);
 
-	reiserfs_write_lock(inode->i_sb);
+	depth = reiserfs_write_lock_once(inode->i_sb);
 	if (attr->ia_valid & ATTR_SIZE) {
 		/* version 2 items will be caught by the s_maxbytes check
 		 ** done for us in vmtruncate
@@ -3138,8 +3150,17 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 				    journal_end(&th, inode->i_sb, jbegin_count);
 			}
 		}
-		if (!error)
+		if (!error) {
+			/*
+			 * Relax the lock here, as it might truncate the
+			 * inode pages and wait for inode pages locks.
+			 * To release such page lock, the owner needs the
+			 * reiserfs lock
+			 */
+			reiserfs_write_unlock_once(inode->i_sb, depth);
 			error = inode_setattr(inode, attr);
+			depth = reiserfs_write_lock_once(inode->i_sb);
+		}
 	}
 
 	if (!error && reiserfs_posixacl(inode->i_sb)) {
@@ -3148,7 +3169,8 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr)
 	}
 
       out:
-	reiserfs_write_unlock(inode->i_sb);
+	reiserfs_write_unlock_once(inode->i_sb, depth);
+
 	return error;
 }
 
diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c
index ace77451ceb1..f53505de0712 100644
--- a/fs/reiserfs/ioctl.c
+++ b/fs/reiserfs/ioctl.c
@@ -104,9 +104,10 @@ setflags_out:
 		err = put_user(inode->i_generation, (int __user *)arg);
 		break;
 	case REISERFS_IOC_SETVERSION:
-		if (!is_owner_or_cap(inode))
+		if (!is_owner_or_cap(inode)) {
 			err = -EPERM;
 			break;
+		}
 		err = mnt_want_write(filp->f_path.mnt);
 		if (err)
 			break;
diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index 2f8a7e7b8dab..ba98546fabbd 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -2009,10 +2009,11 @@ static int do_journal_release(struct reiserfs_transaction_handle *th,
 		destroy_workqueue(commit_wq);
 		commit_wq = NULL;
 	}
-	reiserfs_write_lock(sb);
 
 	free_journal_ram(sb);
 
+	reiserfs_write_lock(sb);
+
 	return 0;
 }
 
@@ -2758,11 +2759,18 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 	struct reiserfs_journal *journal;
 	struct reiserfs_journal_list *jl;
 	char b[BDEVNAME_SIZE];
+	int ret;
 
+	/*
+	 * Unlock here to avoid various RECLAIM-FS-ON <-> IN-RECLAIM-FS
+	 * dependency inversion warnings.
+	 */
+	reiserfs_write_unlock(sb);
 	journal = SB_JOURNAL(sb) = vmalloc(sizeof(struct reiserfs_journal));
 	if (!journal) {
 		reiserfs_warning(sb, "journal-1256",
 				 "unable to get memory for journal structure");
+		reiserfs_write_lock(sb);
 		return 1;
 	}
 	memset(journal, 0, sizeof(struct reiserfs_journal));
@@ -2771,10 +2779,12 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 	INIT_LIST_HEAD(&journal->j_working_list);
 	INIT_LIST_HEAD(&journal->j_journal_list);
 	journal->j_persistent_trans = 0;
-	if (reiserfs_allocate_list_bitmaps(sb,
-					   journal->j_list_bitmap,
-					   reiserfs_bmap_count(sb)))
+	ret = reiserfs_allocate_list_bitmaps(sb, journal->j_list_bitmap,
+					   reiserfs_bmap_count(sb));
+	reiserfs_write_lock(sb);
+	if (ret)
 		goto free_and_return;
+
 	allocate_bitmap_nodes(sb);
 
 	/* reserved for journal area support */
@@ -2903,7 +2913,9 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 	journal->j_mount_id = 10;
 	journal->j_state = 0;
 	atomic_set(&(journal->j_jlock), 0);
+	reiserfs_write_unlock(sb);
 	journal->j_cnode_free_list = allocate_cnodes(num_cnodes);
+	reiserfs_write_lock(sb);
 	journal->j_cnode_free_orig = journal->j_cnode_free_list;
 	journal->j_cnode_free = journal->j_cnode_free_list ? num_cnodes : 0;
 	journal->j_cnode_used = 0;
diff --git a/fs/reiserfs/lock.c b/fs/reiserfs/lock.c
index ee2cfc0fd8a7..b87aa2c1afc1 100644
--- a/fs/reiserfs/lock.c
+++ b/fs/reiserfs/lock.c
@@ -86,3 +86,12 @@ void reiserfs_check_lock_depth(struct super_block *sb, char *caller)
 		reiserfs_panic(sb, "%s called without kernel lock held %d",
 			       caller);
 }
+
+#ifdef CONFIG_REISERFS_CHECK
+void reiserfs_lock_check_recursive(struct super_block *sb)
+{
+	struct reiserfs_sb_info *sb_i = REISERFS_SB(sb);
+
+	WARN_ONCE((sb_i->lock_depth > 0), "Unwanted recursive reiserfs lock!\n");
+}
+#endif
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index e296ff72a6cc..9d4dcf0b07cb 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -921,6 +921,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry)
 	struct reiserfs_transaction_handle th;
 	int jbegin_count;
 	unsigned long savelink;
+	int depth;
 
 	inode = dentry->d_inode;
 
@@ -932,7 +933,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry)
 	    JOURNAL_PER_BALANCE_CNT * 2 + 2 +
 	    4 * REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb);
 
-	reiserfs_write_lock(dir->i_sb);
+	depth = reiserfs_write_lock_once(dir->i_sb);
 	retval = journal_begin(&th, dir->i_sb, jbegin_count);
 	if (retval)
 		goto out_unlink;
@@ -993,7 +994,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry)
 
 	retval = journal_end(&th, dir->i_sb, jbegin_count);
 	reiserfs_check_path(&path);
-	reiserfs_write_unlock(dir->i_sb);
+	reiserfs_write_unlock_once(dir->i_sb, depth);
 	return retval;
 
       end_unlink:
@@ -1003,7 +1004,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry)
 	if (err)
 		retval = err;
       out_unlink:
-	reiserfs_write_unlock(dir->i_sb);
+	reiserfs_write_unlock_once(dir->i_sb, depth);
 	return retval;
 }
 
diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c
index 9229e5514a4e..7a9981196c1c 100644
--- a/fs/reiserfs/procfs.c
+++ b/fs/reiserfs/procfs.c
@@ -17,8 +17,6 @@
 #include <linux/init.h>
 #include <linux/proc_fs.h>
 
-#ifdef CONFIG_REISERFS_PROC_INFO
-
 /*
  * LOCKING:
  *
@@ -48,14 +46,6 @@ static int show_version(struct seq_file *m, struct super_block *sb)
 	return 0;
 }
 
-int reiserfs_global_version_in_proc(char *buffer, char **start, off_t offset,
-				    int count, int *eof, void *data)
-{
-	*start = buffer;
-	*eof = 1;
-	return 0;
-}
-
 #define SF( x ) ( r -> x )
 #define SFP( x ) SF( s_proc_info_data.x )
 #define SFPL( x ) SFP( x[ level ] )
@@ -538,19 +528,6 @@ int reiserfs_proc_info_done(struct super_block *sb)
 	return 0;
 }
 
-struct proc_dir_entry *reiserfs_proc_register_global(char *name,
-						     read_proc_t * func)
-{
-	return (proc_info_root) ? create_proc_read_entry(name, 0,
-							 proc_info_root,
-							 func, NULL) : NULL;
-}
-
-void reiserfs_proc_unregister_global(const char *name)
-{
-	remove_proc_entry(name, proc_info_root);
-}
-
 int reiserfs_proc_info_global_init(void)
 {
 	if (proc_info_root == NULL) {
@@ -572,48 +549,6 @@ int reiserfs_proc_info_global_done(void)
 	}
 	return 0;
 }
-
-/* REISERFS_PROC_INFO */
-#else
-
-int reiserfs_proc_info_init(struct super_block *sb)
-{
-	return 0;
-}
-int reiserfs_proc_info_done(struct super_block *sb)
-{
-	return 0;
-}
-
-struct proc_dir_entry *reiserfs_proc_register_global(char *name,
-						     read_proc_t * func)
-{
-	return NULL;
-}
-
-void reiserfs_proc_unregister_global(const char *name)
-{;
-}
-
-int reiserfs_proc_info_global_init(void)
-{
-	return 0;
-}
-int reiserfs_proc_info_global_done(void)
-{
-	return 0;
-}
-
-int reiserfs_global_version_in_proc(char *buffer, char **start,
-				    off_t offset,
-				    int count, int *eof, void *data)
-{
-	return 0;
-}
-
-/* REISERFS_PROC_INFO */
-#endif
-
 /*
  * Revision 1.1.8.2  2001/07/15 17:08:42  god
  *  . use get_super() in procfs.c
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 339b0baf2af6..b4a7dd03bdb9 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -2222,8 +2222,6 @@ static int __init init_reiserfs_fs(void)
 	}
 
 	reiserfs_proc_info_global_init();
-	reiserfs_proc_register_global("version",
-				      reiserfs_global_version_in_proc);
 
 	ret = register_filesystem(&reiserfs_fs_type);
 
@@ -2231,7 +2229,6 @@ static int __init init_reiserfs_fs(void)
 		return 0;
 	}
 
-	reiserfs_proc_unregister_global("version");
 	reiserfs_proc_info_global_done();
 	destroy_inodecache();
 
@@ -2240,7 +2237,6 @@ static int __init init_reiserfs_fs(void)
 
 static void __exit exit_reiserfs_fs(void)
 {
-	reiserfs_proc_unregister_global("version");
 	reiserfs_proc_info_global_done();
 	unregister_filesystem(&reiserfs_fs_type);
 	destroy_inodecache();
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 58aa8e75f7f5..81f09fab8ae4 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -48,6 +48,7 @@
 #include <net/checksum.h>
 #include <linux/stat.h>
 #include <linux/quotaops.h>
+#include <linux/security.h>
 
 #define PRIVROOT_NAME ".reiserfs_priv"
 #define XAROOT_NAME   "xattrs"
@@ -82,7 +83,8 @@ static int xattr_unlink(struct inode *dir, struct dentry *dentry)
 	BUG_ON(!mutex_is_locked(&dir->i_mutex));
 	vfs_dq_init(dir);
 
-	mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD);
+	reiserfs_mutex_lock_nested_safe(&dentry->d_inode->i_mutex,
+					I_MUTEX_CHILD, dir->i_sb);
 	error = dir->i_op->unlink(dir, dentry);
 	mutex_unlock(&dentry->d_inode->i_mutex);
 
@@ -97,7 +99,8 @@ static int xattr_rmdir(struct inode *dir, struct dentry *dentry)
 	BUG_ON(!mutex_is_locked(&dir->i_mutex));
 	vfs_dq_init(dir);
 
-	mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD);
+	reiserfs_mutex_lock_nested_safe(&dentry->d_inode->i_mutex,
+					I_MUTEX_CHILD, dir->i_sb);
 	dentry_unhash(dentry);
 	error = dir->i_op->rmdir(dir, dentry);
 	if (!error)
@@ -234,16 +237,22 @@ static int reiserfs_for_each_xattr(struct inode *inode,
 	if (IS_PRIVATE(inode) || get_inode_sd_version(inode) == STAT_DATA_V1)
 		return 0;
 
+	reiserfs_write_unlock(inode->i_sb);
 	dir = open_xa_dir(inode, XATTR_REPLACE);
 	if (IS_ERR(dir)) {
 		err = PTR_ERR(dir);
+		reiserfs_write_lock(inode->i_sb);
 		goto out;
 	} else if (!dir->d_inode) {
 		err = 0;
+		reiserfs_write_lock(inode->i_sb);
 		goto out_dir;
 	}
 
 	mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_XATTR);
+
+	reiserfs_write_lock(inode->i_sb);
+
 	buf.xadir = dir;
 	err = reiserfs_readdir_dentry(dir, &buf, fill_with_dentries, &pos);
 	while ((err == 0 || err == -ENOSPC) && buf.count) {
@@ -282,8 +291,9 @@ static int reiserfs_for_each_xattr(struct inode *inode,
 		err = journal_begin(&th, inode->i_sb, blocks);
 		if (!err) {
 			int jerror;
-			mutex_lock_nested(&dir->d_parent->d_inode->i_mutex,
-					  I_MUTEX_XATTR);
+			reiserfs_mutex_lock_nested_safe(
+					  &dir->d_parent->d_inode->i_mutex,
+					  I_MUTEX_XATTR, inode->i_sb);
 			err = action(dir, data);
 			jerror = journal_end(&th, inode->i_sb, blocks);
 			mutex_unlock(&dir->d_parent->d_inode->i_mutex);
@@ -442,7 +452,9 @@ static int lookup_and_delete_xattr(struct inode *inode, const char *name)
 	}
 
 	if (dentry->d_inode) {
+		reiserfs_write_lock(inode->i_sb);
 		err = xattr_unlink(xadir->d_inode, dentry);
+		reiserfs_write_unlock(inode->i_sb);
 		update_ctime(inode);
 	}
 
@@ -476,15 +488,24 @@ reiserfs_xattr_set_handle(struct reiserfs_transaction_handle *th,
 	if (get_inode_sd_version(inode) == STAT_DATA_V1)
 		return -EOPNOTSUPP;
 
-	if (!buffer)
-		return lookup_and_delete_xattr(inode, name);
+	reiserfs_write_unlock(inode->i_sb);
+
+	if (!buffer) {
+		err = lookup_and_delete_xattr(inode, name);
+		reiserfs_write_lock(inode->i_sb);
+		return err;
+	}
 
 	dentry = xattr_lookup(inode, name, flags);
-	if (IS_ERR(dentry))
+	if (IS_ERR(dentry)) {
+		reiserfs_write_lock(inode->i_sb);
 		return PTR_ERR(dentry);
+	}
 
 	down_write(&REISERFS_I(inode)->i_xattr_sem);
 
+	reiserfs_write_lock(inode->i_sb);
+
 	xahash = xattr_hash(buffer, buffer_size);
 	while (buffer_pos < buffer_size || buffer_pos == 0) {
 		size_t chunk;
@@ -539,8 +560,12 @@ reiserfs_xattr_set_handle(struct reiserfs_transaction_handle *th,
 			.ia_size = buffer_size,
 			.ia_valid = ATTR_SIZE | ATTR_CTIME,
 		};
+
+		reiserfs_write_unlock(inode->i_sb);
 		mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_XATTR);
 		down_write(&dentry->d_inode->i_alloc_sem);
+		reiserfs_write_lock(inode->i_sb);
+
 		err = reiserfs_setattr(dentry, &newattrs);
 		up_write(&dentry->d_inode->i_alloc_sem);
 		mutex_unlock(&dentry->d_inode->i_mutex);
@@ -726,15 +751,14 @@ ssize_t
 reiserfs_getxattr(struct dentry * dentry, const char *name, void *buffer,
 		  size_t size)
 {
-	struct inode *inode = dentry->d_inode;
 	struct xattr_handler *handler;
 
-	handler = find_xattr_handler_prefix(inode->i_sb->s_xattr, name);
+	handler = find_xattr_handler_prefix(dentry->d_sb->s_xattr, name);
 
-	if (!handler || get_inode_sd_version(inode) == STAT_DATA_V1)
+	if (!handler || get_inode_sd_version(dentry->d_inode) == STAT_DATA_V1)
 		return -EOPNOTSUPP;
 
-	return handler->get(inode, name, buffer, size);
+	return handler->get(dentry, name, buffer, size, handler->flags);
 }
 
 /*
@@ -746,15 +770,14 @@ int
 reiserfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 		  size_t size, int flags)
 {
-	struct inode *inode = dentry->d_inode;
 	struct xattr_handler *handler;
 
-	handler = find_xattr_handler_prefix(inode->i_sb->s_xattr, name);
+	handler = find_xattr_handler_prefix(dentry->d_sb->s_xattr, name);
 
-	if (!handler || get_inode_sd_version(inode) == STAT_DATA_V1)
+	if (!handler || get_inode_sd_version(dentry->d_inode) == STAT_DATA_V1)
 		return -EOPNOTSUPP;
 
-	return handler->set(inode, name, value, size, flags);
+	return handler->set(dentry, name, value, size, flags, handler->flags);
 }
 
 /*
@@ -764,21 +787,20 @@ reiserfs_setxattr(struct dentry *dentry, const char *name, const void *value,
  */
 int reiserfs_removexattr(struct dentry *dentry, const char *name)
 {
-	struct inode *inode = dentry->d_inode;
 	struct xattr_handler *handler;
-	handler = find_xattr_handler_prefix(inode->i_sb->s_xattr, name);
+	handler = find_xattr_handler_prefix(dentry->d_sb->s_xattr, name);
 
-	if (!handler || get_inode_sd_version(inode) == STAT_DATA_V1)
+	if (!handler || get_inode_sd_version(dentry->d_inode) == STAT_DATA_V1)
 		return -EOPNOTSUPP;
 
-	return handler->set(inode, name, NULL, 0, XATTR_REPLACE);
+	return handler->set(dentry, name, NULL, 0, XATTR_REPLACE, handler->flags);
 }
 
 struct listxattr_buf {
 	size_t size;
 	size_t pos;
 	char *buf;
-	struct inode *inode;
+	struct dentry *dentry;
 };
 
 static int listxattr_filler(void *buf, const char *name, int namelen,
@@ -789,17 +811,19 @@ static int listxattr_filler(void *buf, const char *name, int namelen,
 	if (name[0] != '.' ||
 	    (namelen != 1 && (name[1] != '.' || namelen != 2))) {
 		struct xattr_handler *handler;
-		handler = find_xattr_handler_prefix(b->inode->i_sb->s_xattr,
+		handler = find_xattr_handler_prefix(b->dentry->d_sb->s_xattr,
 						    name);
 		if (!handler)	/* Unsupported xattr name */
 			return 0;
 		if (b->buf) {
-			size = handler->list(b->inode, b->buf + b->pos,
-					 b->size, name, namelen);
+			size = handler->list(b->dentry, b->buf + b->pos,
+					 b->size, name, namelen,
+					 handler->flags);
 			if (size > b->size)
 				return -ERANGE;
 		} else {
-			size = handler->list(b->inode, NULL, 0, name, namelen);
+			size = handler->list(b->dentry, NULL, 0, name,
+					     namelen, handler->flags);
 		}
 
 		b->pos += size;
@@ -820,7 +844,7 @@ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size)
 	int err = 0;
 	loff_t pos = 0;
 	struct listxattr_buf buf = {
-		.inode = dentry->d_inode,
+		.dentry = dentry,
 		.buf = buffer,
 		.size = buffer ? size : 0,
 	};
diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index 35d6e672a279..dd20a7883f0f 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -15,8 +15,10 @@ static int reiserfs_set_acl(struct reiserfs_transaction_handle *th,
 			    struct posix_acl *acl);
 
 static int
-xattr_set_acl(struct inode *inode, int type, const void *value, size_t size)
+posix_acl_set(struct dentry *dentry, const char *name, const void *value,
+		size_t size, int flags, int type)
 {
+	struct inode *inode = dentry->d_inode;
 	struct posix_acl *acl;
 	int error, error2;
 	struct reiserfs_transaction_handle th;
@@ -60,15 +62,16 @@ xattr_set_acl(struct inode *inode, int type, const void *value, size_t size)
 }
 
 static int
-xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size)
+posix_acl_get(struct dentry *dentry, const char *name, void *buffer,
+		size_t size, int type)
 {
 	struct posix_acl *acl;
 	int error;
 
-	if (!reiserfs_posixacl(inode->i_sb))
+	if (!reiserfs_posixacl(dentry->d_sb))
 		return -EOPNOTSUPP;
 
-	acl = reiserfs_get_acl(inode, type);
+	acl = reiserfs_get_acl(dentry->d_inode, type);
 	if (IS_ERR(acl))
 		return PTR_ERR(acl);
 	if (acl == NULL)
@@ -452,7 +455,9 @@ int reiserfs_acl_chmod(struct inode *inode)
 		return 0;
 	}
 
+	reiserfs_write_unlock(inode->i_sb);
 	acl = reiserfs_get_acl(inode, ACL_TYPE_ACCESS);
+	reiserfs_write_lock(inode->i_sb);
 	if (!acl)
 		return 0;
 	if (IS_ERR(acl))
@@ -482,30 +487,12 @@ int reiserfs_acl_chmod(struct inode *inode)
 	return error;
 }
 
-static int
-posix_acl_access_get(struct inode *inode, const char *name,
-		     void *buffer, size_t size)
-{
-	if (strlen(name) != sizeof(POSIX_ACL_XATTR_ACCESS) - 1)
-		return -EINVAL;
-	return xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size);
-}
-
-static int
-posix_acl_access_set(struct inode *inode, const char *name,
-		     const void *value, size_t size, int flags)
-{
-	if (strlen(name) != sizeof(POSIX_ACL_XATTR_ACCESS) - 1)
-		return -EINVAL;
-	return xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size);
-}
-
-static size_t posix_acl_access_list(struct inode *inode, char *list,
+static size_t posix_acl_access_list(struct dentry *dentry, char *list,
 				    size_t list_size, const char *name,
-				    size_t name_len)
+				    size_t name_len, int type)
 {
 	const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS);
-	if (!reiserfs_posixacl(inode->i_sb))
+	if (!reiserfs_posixacl(dentry->d_sb))
 		return 0;
 	if (list && size <= list_size)
 		memcpy(list, POSIX_ACL_XATTR_ACCESS, size);
@@ -514,35 +501,18 @@ static size_t posix_acl_access_list(struct inode *inode, char *list,
 
 struct xattr_handler reiserfs_posix_acl_access_handler = {
 	.prefix = POSIX_ACL_XATTR_ACCESS,
-	.get = posix_acl_access_get,
-	.set = posix_acl_access_set,
+	.flags = ACL_TYPE_ACCESS,
+	.get = posix_acl_get,
+	.set = posix_acl_set,
 	.list = posix_acl_access_list,
 };
 
-static int
-posix_acl_default_get(struct inode *inode, const char *name,
-		      void *buffer, size_t size)
-{
-	if (strlen(name) != sizeof(POSIX_ACL_XATTR_DEFAULT) - 1)
-		return -EINVAL;
-	return xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size);
-}
-
-static int
-posix_acl_default_set(struct inode *inode, const char *name,
-		      const void *value, size_t size, int flags)
-{
-	if (strlen(name) != sizeof(POSIX_ACL_XATTR_DEFAULT) - 1)
-		return -EINVAL;
-	return xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size);
-}
-
-static size_t posix_acl_default_list(struct inode *inode, char *list,
+static size_t posix_acl_default_list(struct dentry *dentry, char *list,
 				     size_t list_size, const char *name,
-				     size_t name_len)
+				     size_t name_len, int type)
 {
 	const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT);
-	if (!reiserfs_posixacl(inode->i_sb))
+	if (!reiserfs_posixacl(dentry->d_sb))
 		return 0;
 	if (list && size <= list_size)
 		memcpy(list, POSIX_ACL_XATTR_DEFAULT, size);
@@ -551,7 +521,8 @@ static size_t posix_acl_default_list(struct inode *inode, char *list,
 
 struct xattr_handler reiserfs_posix_acl_default_handler = {
 	.prefix = POSIX_ACL_XATTR_DEFAULT,
-	.get = posix_acl_default_get,
-	.set = posix_acl_default_set,
+	.flags = ACL_TYPE_DEFAULT,
+	.get = posix_acl_get,
+	.set = posix_acl_set,
 	.list = posix_acl_default_list,
 };
diff --git a/fs/reiserfs/xattr_security.c b/fs/reiserfs/xattr_security.c
index a92c8792c0f6..d8b5bfcbdd30 100644
--- a/fs/reiserfs/xattr_security.c
+++ b/fs/reiserfs/xattr_security.c
@@ -8,36 +8,37 @@
 #include <asm/uaccess.h>
 
 static int
-security_get(struct inode *inode, const char *name, void *buffer, size_t size)
+security_get(struct dentry *dentry, const char *name, void *buffer, size_t size,
+		int handler_flags)
 {
 	if (strlen(name) < sizeof(XATTR_SECURITY_PREFIX))
 		return -EINVAL;
 
-	if (IS_PRIVATE(inode))
+	if (IS_PRIVATE(dentry->d_inode))
 		return -EPERM;
 
-	return reiserfs_xattr_get(inode, name, buffer, size);
+	return reiserfs_xattr_get(dentry->d_inode, name, buffer, size);
 }
 
 static int
-security_set(struct inode *inode, const char *name, const void *buffer,
-	     size_t size, int flags)
+security_set(struct dentry *dentry, const char *name, const void *buffer,
+	     size_t size, int flags, int handler_flags)
 {
 	if (strlen(name) < sizeof(XATTR_SECURITY_PREFIX))
 		return -EINVAL;
 
-	if (IS_PRIVATE(inode))
+	if (IS_PRIVATE(dentry->d_inode))
 		return -EPERM;
 
-	return reiserfs_xattr_set(inode, name, buffer, size, flags);
+	return reiserfs_xattr_set(dentry->d_inode, name, buffer, size, flags);
 }
 
-static size_t security_list(struct inode *inode, char *list, size_t list_len,
-			    const char *name, size_t namelen)
+static size_t security_list(struct dentry *dentry, char *list, size_t list_len,
+			    const char *name, size_t namelen, int handler_flags)
 {
 	const size_t len = namelen + 1;
 
-	if (IS_PRIVATE(inode))
+	if (IS_PRIVATE(dentry->d_inode))
 		return 0;
 
 	if (list && len <= list_len) {
diff --git a/fs/reiserfs/xattr_trusted.c b/fs/reiserfs/xattr_trusted.c
index a865042f75e2..5b08aaca3daf 100644
--- a/fs/reiserfs/xattr_trusted.c
+++ b/fs/reiserfs/xattr_trusted.c
@@ -8,36 +8,37 @@
 #include <asm/uaccess.h>
 
 static int
-trusted_get(struct inode *inode, const char *name, void *buffer, size_t size)
+trusted_get(struct dentry *dentry, const char *name, void *buffer, size_t size,
+	    int handler_flags)
 {
 	if (strlen(name) < sizeof(XATTR_TRUSTED_PREFIX))
 		return -EINVAL;
 
-	if (!capable(CAP_SYS_ADMIN) || IS_PRIVATE(inode))
+	if (!capable(CAP_SYS_ADMIN) || IS_PRIVATE(dentry->d_inode))
 		return -EPERM;
 
-	return reiserfs_xattr_get(inode, name, buffer, size);
+	return reiserfs_xattr_get(dentry->d_inode, name, buffer, size);
 }
 
 static int
-trusted_set(struct inode *inode, const char *name, const void *buffer,
-	    size_t size, int flags)
+trusted_set(struct dentry *dentry, const char *name, const void *buffer,
+	    size_t size, int flags, int handler_flags)
 {
 	if (strlen(name) < sizeof(XATTR_TRUSTED_PREFIX))
 		return -EINVAL;
 
-	if (!capable(CAP_SYS_ADMIN) || IS_PRIVATE(inode))
+	if (!capable(CAP_SYS_ADMIN) || IS_PRIVATE(dentry->d_inode))
 		return -EPERM;
 
-	return reiserfs_xattr_set(inode, name, buffer, size, flags);
+	return reiserfs_xattr_set(dentry->d_inode, name, buffer, size, flags);
 }
 
-static size_t trusted_list(struct inode *inode, char *list, size_t list_size,
-			   const char *name, size_t name_len)
+static size_t trusted_list(struct dentry *dentry, char *list, size_t list_size,
+			   const char *name, size_t name_len, int handler_flags)
 {
 	const size_t len = name_len + 1;
 
-	if (!capable(CAP_SYS_ADMIN) || IS_PRIVATE(inode))
+	if (!capable(CAP_SYS_ADMIN) || IS_PRIVATE(dentry->d_inode))
 		return 0;
 
 	if (list && len <= list_size) {
diff --git a/fs/reiserfs/xattr_user.c b/fs/reiserfs/xattr_user.c
index e3238dc4f3db..75d59c49b911 100644
--- a/fs/reiserfs/xattr_user.c
+++ b/fs/reiserfs/xattr_user.c
@@ -7,34 +7,35 @@
 #include <asm/uaccess.h>
 
 static int
-user_get(struct inode *inode, const char *name, void *buffer, size_t size)
+user_get(struct dentry *dentry, const char *name, void *buffer, size_t size,
+	 int handler_flags)
 {
 
 	if (strlen(name) < sizeof(XATTR_USER_PREFIX))
 		return -EINVAL;
-	if (!reiserfs_xattrs_user(inode->i_sb))
+	if (!reiserfs_xattrs_user(dentry->d_sb))
 		return -EOPNOTSUPP;
-	return reiserfs_xattr_get(inode, name, buffer, size);
+	return reiserfs_xattr_get(dentry->d_inode, name, buffer, size);
 }
 
 static int
-user_set(struct inode *inode, const char *name, const void *buffer,
-	 size_t size, int flags)
+user_set(struct dentry *dentry, const char *name, const void *buffer,
+	 size_t size, int flags, int handler_flags)
 {
 	if (strlen(name) < sizeof(XATTR_USER_PREFIX))
 		return -EINVAL;
 
-	if (!reiserfs_xattrs_user(inode->i_sb))
+	if (!reiserfs_xattrs_user(dentry->d_sb))
 		return -EOPNOTSUPP;
-	return reiserfs_xattr_set(inode, name, buffer, size, flags);
+	return reiserfs_xattr_set(dentry->d_inode, name, buffer, size, flags);
 }
 
-static size_t user_list(struct inode *inode, char *list, size_t list_size,
-			const char *name, size_t name_len)
+static size_t user_list(struct dentry *dentry, char *list, size_t list_size,
+			const char *name, size_t name_len, int handler_flags)
 {
 	const size_t len = name_len + 1;
 
-	if (!reiserfs_xattrs_user(inode->i_sb))
+	if (!reiserfs_xattrs_user(dentry->d_sb))
 		return 0;
 	if (list && len <= list_size) {
 		memcpy(list, name, name_len);
diff --git a/fs/romfs/super.c b/fs/romfs/super.c
index c117fa80d1e9..42d213546894 100644
--- a/fs/romfs/super.c
+++ b/fs/romfs/super.c
@@ -544,6 +544,7 @@ error:
 error_rsb_inval:
 	ret = -EINVAL;
 error_rsb:
+	kfree(rsb);
 	return ret;
 }
 
diff --git a/fs/signalfd.c b/fs/signalfd.c
index b07565c94386..1dabe4ee02fe 100644
--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -236,7 +236,7 @@ SYSCALL_DEFINE4(signalfd4, int, ufd, sigset_t __user *, user_mask,
 		 * anon_inode_getfd() will install the fd.
 		 */
 		ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx,
-				       flags & (O_CLOEXEC | O_NONBLOCK));
+				       O_RDWR | (flags & (O_CLOEXEC | O_NONBLOCK)));
 		if (ufd < 0)
 			kfree(ctx);
 	} else {
diff --git a/fs/stack.c b/fs/stack.c
index 67716f6a1a4a..4a6f7f440658 100644
--- a/fs/stack.c
+++ b/fs/stack.c
@@ -7,18 +7,63 @@
  * This function cannot be inlined since i_size_{read,write} is rather
  * heavy-weight on 32-bit systems
  */
-void fsstack_copy_inode_size(struct inode *dst, const struct inode *src)
+void fsstack_copy_inode_size(struct inode *dst, struct inode *src)
 {
-	i_size_write(dst, i_size_read((struct inode *)src));
-	dst->i_blocks = src->i_blocks;
+	loff_t i_size;
+	blkcnt_t i_blocks;
+
+	/*
+	 * i_size_read() includes its own seqlocking and protection from
+	 * preemption (see include/linux/fs.h): we need nothing extra for
+	 * that here, and prefer to avoid nesting locks than attempt to keep
+	 * i_size and i_blocks in sync together.
+	 */
+	i_size = i_size_read(src);
+
+	/*
+	 * But if CONFIG_LBDAF (on 32-bit), we ought to make an effort to
+	 * keep the two halves of i_blocks in sync despite SMP or PREEMPT -
+	 * though stat's generic_fillattr() doesn't bother, and we won't be
+	 * applying quotas (where i_blocks does become important) at the
+	 * upper level.
+	 *
+	 * We don't actually know what locking is used at the lower level;
+	 * but if it's a filesystem that supports quotas, it will be using
+	 * i_lock as in inode_add_bytes().  tmpfs uses other locking, and
+	 * its 32-bit is (just) able to exceed 2TB i_size with the aid of
+	 * holes; but its i_blocks cannot carry into the upper long without
+	 * almost 2TB swap - let's ignore that case.
+	 */
+	if (sizeof(i_blocks) > sizeof(long))
+		spin_lock(&src->i_lock);
+	i_blocks = src->i_blocks;
+	if (sizeof(i_blocks) > sizeof(long))
+		spin_unlock(&src->i_lock);
+
+	/*
+	 * If CONFIG_SMP or CONFIG_PREEMPT on 32-bit, it's vital for
+	 * fsstack_copy_inode_size() to hold some lock around
+	 * i_size_write(), otherwise i_size_read() may spin forever (see
+	 * include/linux/fs.h).  We don't necessarily hold i_mutex when this
+	 * is called, so take i_lock for that case.
+	 *
+	 * And if CONFIG_LBADF (on 32-bit), continue our effort to keep the
+	 * two halves of i_blocks in sync despite SMP or PREEMPT: use i_lock
+	 * for that case too, and do both at once by combining the tests.
+	 *
+	 * There is none of this locking overhead in the 64-bit case.
+	 */
+	if (sizeof(i_size) > sizeof(long) || sizeof(i_blocks) > sizeof(long))
+		spin_lock(&dst->i_lock);
+	i_size_write(dst, i_size);
+	dst->i_blocks = i_blocks;
+	if (sizeof(i_size) > sizeof(long) || sizeof(i_blocks) > sizeof(long))
+		spin_unlock(&dst->i_lock);
 }
 EXPORT_SYMBOL_GPL(fsstack_copy_inode_size);
 
-/* copy all attributes; get_nlinks is optional way to override the i_nlink
- * copying
- */
-void fsstack_copy_attr_all(struct inode *dest, const struct inode *src,
-				int (*get_nlinks)(struct inode *))
+/* copy all attributes */
+void fsstack_copy_attr_all(struct inode *dest, const struct inode *src)
 {
 	dest->i_mode = src->i_mode;
 	dest->i_uid = src->i_uid;
@@ -29,14 +74,6 @@ void fsstack_copy_attr_all(struct inode *dest, const struct inode *src,
 	dest->i_ctime = src->i_ctime;
 	dest->i_blkbits = src->i_blkbits;
 	dest->i_flags = src->i_flags;
-
-	/*
-	 * Update the nlinks AFTER updating the above fields, because the
-	 * get_links callback may depend on them.
-	 */
-	if (!get_nlinks)
-		dest->i_nlink = src->i_nlink;
-	else
-		dest->i_nlink = (*get_nlinks)(dest);
+	dest->i_nlink = src->i_nlink;
 }
 EXPORT_SYMBOL_GPL(fsstack_copy_attr_all);
diff --git a/fs/stat.c b/fs/stat.c
index 075694e31d8b..c4ecd52c5737 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -401,9 +401,9 @@ SYSCALL_DEFINE4(fstatat64, int, dfd, char __user *, filename,
 }
 #endif /* __ARCH_WANT_STAT64 */
 
-void inode_add_bytes(struct inode *inode, loff_t bytes)
+/* Caller is here responsible for sufficient locking (ie. inode->i_lock) */
+void __inode_add_bytes(struct inode *inode, loff_t bytes)
 {
-	spin_lock(&inode->i_lock);
 	inode->i_blocks += bytes >> 9;
 	bytes &= 511;
 	inode->i_bytes += bytes;
@@ -411,6 +411,12 @@ void inode_add_bytes(struct inode *inode, loff_t bytes)
 		inode->i_blocks++;
 		inode->i_bytes -= 512;
 	}
+}
+
+void inode_add_bytes(struct inode *inode, loff_t bytes)
+{
+	spin_lock(&inode->i_lock);
+	__inode_add_bytes(inode, bytes);
 	spin_unlock(&inode->i_lock);
 }
 
diff --git a/fs/super.c b/fs/super.c
index 19eb70b374bc..aff046b0fe78 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -901,8 +901,9 @@ int get_sb_single(struct file_system_type *fs_type,
 			return error;
 		}
 		s->s_flags |= MS_ACTIVE;
+	} else {
+		do_remount_sb(s, flags, data, 0);
 	}
-	do_remount_sb(s, flags, data, 0);
 	simple_set_mnt(mnt, s);
 	return 0;
 }
diff --git a/fs/sync.c b/fs/sync.c
index d104591b066b..418727a2a239 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -295,10 +295,11 @@ SYSCALL_DEFINE1(fdatasync, unsigned int, fd)
  */
 int generic_write_sync(struct file *file, loff_t pos, loff_t count)
 {
-	if (!(file->f_flags & O_SYNC) && !IS_SYNC(file->f_mapping->host))
+	if (!(file->f_flags & O_DSYNC) && !IS_SYNC(file->f_mapping->host))
 		return 0;
 	return vfs_fsync_range(file, file->f_path.dentry, pos,
-			       pos + count - 1, 1);
+			       pos + count - 1,
+			       (file->f_flags & __O_SYNC) ? 0 : 1);
 }
 EXPORT_SYMBOL(generic_write_sync);
 
@@ -354,6 +355,7 @@ SYSCALL_DEFINE(sync_file_range)(int fd, loff_t offset, loff_t nbytes,
 {
 	int ret;
 	struct file *file;
+	struct address_space *mapping;
 	loff_t endbyte;			/* inclusive */
 	int fput_needed;
 	umode_t i_mode;
@@ -404,7 +406,28 @@ SYSCALL_DEFINE(sync_file_range)(int fd, loff_t offset, loff_t nbytes,
 			!S_ISLNK(i_mode))
 		goto out_put;
 
-	ret = do_sync_mapping_range(file->f_mapping, offset, endbyte, flags);
+	mapping = file->f_mapping;
+	if (!mapping) {
+		ret = -EINVAL;
+		goto out_put;
+	}
+
+	ret = 0;
+	if (flags & SYNC_FILE_RANGE_WAIT_BEFORE) {
+		ret = filemap_fdatawait_range(mapping, offset, endbyte);
+		if (ret < 0)
+			goto out_put;
+	}
+
+	if (flags & SYNC_FILE_RANGE_WRITE) {
+		ret = filemap_fdatawrite_range(mapping, offset, endbyte);
+		if (ret < 0)
+			goto out_put;
+	}
+
+	if (flags & SYNC_FILE_RANGE_WAIT_AFTER)
+		ret = filemap_fdatawait_range(mapping, offset, endbyte);
+
 out_put:
 	fput_light(file, fput_needed);
 out:
@@ -436,42 +459,3 @@ asmlinkage long SyS_sync_file_range2(long fd, long flags,
 }
 SYSCALL_ALIAS(sys_sync_file_range2, SyS_sync_file_range2);
 #endif
-
-/*
- * `endbyte' is inclusive
- */
-int do_sync_mapping_range(struct address_space *mapping, loff_t offset,
-			  loff_t endbyte, unsigned int flags)
-{
-	int ret;
-
-	if (!mapping) {
-		ret = -EINVAL;
-		goto out;
-	}
-
-	ret = 0;
-	if (flags & SYNC_FILE_RANGE_WAIT_BEFORE) {
-		ret = wait_on_page_writeback_range(mapping,
-					offset >> PAGE_CACHE_SHIFT,
-					endbyte >> PAGE_CACHE_SHIFT);
-		if (ret < 0)
-			goto out;
-	}
-
-	if (flags & SYNC_FILE_RANGE_WRITE) {
-		ret = __filemap_fdatawrite_range(mapping, offset, endbyte,
-						WB_SYNC_ALL);
-		if (ret < 0)
-			goto out;
-	}
-
-	if (flags & SYNC_FILE_RANGE_WAIT_AFTER) {
-		ret = wait_on_page_writeback_range(mapping,
-					offset >> PAGE_CACHE_SHIFT,
-					endbyte >> PAGE_CACHE_SHIFT);
-	}
-out:
-	return ret;
-}
-EXPORT_SYMBOL_GPL(do_sync_mapping_range);
diff --git a/fs/sysfs/bin.c b/fs/sysfs/bin.c
index 60c702bc10ae..a0a500af24a1 100644
--- a/fs/sysfs/bin.c
+++ b/fs/sysfs/bin.c
@@ -483,7 +483,8 @@ void unmap_bin_file(struct sysfs_dirent *attr_sd)
  *	@attr:	attribute descriptor.
  */
 
-int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr)
+int sysfs_create_bin_file(struct kobject *kobj,
+			  const struct bin_attribute *attr)
 {
 	BUG_ON(!kobj || !kobj->sd || !attr);
 
@@ -497,7 +498,8 @@ int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr)
  *	@attr:	attribute descriptor.
  */
 
-void sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr)
+void sysfs_remove_bin_file(struct kobject *kobj,
+			   const struct bin_attribute *attr)
 {
 	sysfs_hash_and_remove(kobj->sd, attr->attr.name);
 }
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index e0201837d244..699f371b9f12 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -25,7 +25,6 @@
 #include "sysfs.h"
 
 DEFINE_MUTEX(sysfs_mutex);
-DEFINE_MUTEX(sysfs_rename_mutex);
 DEFINE_SPINLOCK(sysfs_assoc_lock);
 
 static DEFINE_SPINLOCK(sysfs_ino_lock);
@@ -85,46 +84,6 @@ static void sysfs_unlink_sibling(struct sysfs_dirent *sd)
 }
 
 /**
- *	sysfs_get_dentry - get dentry for the given sysfs_dirent
- *	@sd: sysfs_dirent of interest
- *
- *	Get dentry for @sd.  Dentry is looked up if currently not
- *	present.  This function descends from the root looking up
- *	dentry for each step.
- *
- *	LOCKING:
- *	mutex_lock(sysfs_rename_mutex)
- *
- *	RETURNS:
- *	Pointer to found dentry on success, ERR_PTR() value on error.
- */
-struct dentry *sysfs_get_dentry(struct sysfs_dirent *sd)
-{
-	struct dentry *dentry = dget(sysfs_sb->s_root);
-
-	while (dentry->d_fsdata != sd) {
-		struct sysfs_dirent *cur;
-		struct dentry *parent;
-
-		/* find the first ancestor which hasn't been looked up */
-		cur = sd;
-		while (cur->s_parent != dentry->d_fsdata)
-			cur = cur->s_parent;
-
-		/* look it up */
-		parent = dentry;
-		mutex_lock(&parent->d_inode->i_mutex);
-		dentry = lookup_one_noperm(cur->s_name, parent);
-		mutex_unlock(&parent->d_inode->i_mutex);
-		dput(parent);
-
-		if (IS_ERR(dentry))
-			break;
-	}
-	return dentry;
-}
-
-/**
  *	sysfs_get_active - get an active reference to sysfs_dirent
  *	@sd: sysfs_dirent to get an active reference to
  *
@@ -147,8 +106,10 @@ static struct sysfs_dirent *sysfs_get_active(struct sysfs_dirent *sd)
 			return NULL;
 
 		t = atomic_cmpxchg(&sd->s_active, v, v + 1);
-		if (likely(t == v))
+		if (likely(t == v)) {
+			rwsem_acquire_read(&sd->dep_map, 0, 1, _RET_IP_);
 			return sd;
+		}
 		if (t < 0)
 			return NULL;
 
@@ -171,6 +132,7 @@ static void sysfs_put_active(struct sysfs_dirent *sd)
 	if (unlikely(!sd))
 		return;
 
+	rwsem_release(&sd->dep_map, 1, _RET_IP_);
 	v = atomic_dec_return(&sd->s_active);
 	if (likely(v != SD_DEACTIVATED_BIAS))
 		return;
@@ -235,15 +197,21 @@ static void sysfs_deactivate(struct sysfs_dirent *sd)
 	BUG_ON(sd->s_sibling || !(sd->s_flags & SYSFS_FLAG_REMOVED));
 	sd->s_sibling = (void *)&wait;
 
+	rwsem_acquire(&sd->dep_map, 0, 0, _RET_IP_);
 	/* atomic_add_return() is a mb(), put_active() will always see
 	 * the updated sd->s_sibling.
 	 */
 	v = atomic_add_return(SD_DEACTIVATED_BIAS, &sd->s_active);
 
-	if (v != SD_DEACTIVATED_BIAS)
+	if (v != SD_DEACTIVATED_BIAS) {
+		lock_contended(&sd->dep_map, _RET_IP_);
 		wait_for_completion(&wait);
+	}
 
 	sd->s_sibling = NULL;
+
+	lock_acquired(&sd->dep_map, _RET_IP_);
+	rwsem_release(&sd->dep_map, 1, _RET_IP_);
 }
 
 static int sysfs_alloc_ino(ino_t *pino)
@@ -298,7 +266,61 @@ void release_sysfs_dirent(struct sysfs_dirent * sd)
 		goto repeat;
 }
 
-static void sysfs_d_iput(struct dentry * dentry, struct inode * inode)
+static int sysfs_dentry_delete(struct dentry *dentry)
+{
+	struct sysfs_dirent *sd = dentry->d_fsdata;
+	return !!(sd->s_flags & SYSFS_FLAG_REMOVED);
+}
+
+static int sysfs_dentry_revalidate(struct dentry *dentry, struct nameidata *nd)
+{
+	struct sysfs_dirent *sd = dentry->d_fsdata;
+	int is_dir;
+
+	mutex_lock(&sysfs_mutex);
+
+	/* The sysfs dirent has been deleted */
+	if (sd->s_flags & SYSFS_FLAG_REMOVED)
+		goto out_bad;
+
+	/* The sysfs dirent has been moved? */
+	if (dentry->d_parent->d_fsdata != sd->s_parent)
+		goto out_bad;
+
+	/* The sysfs dirent has been renamed */
+	if (strcmp(dentry->d_name.name, sd->s_name) != 0)
+		goto out_bad;
+
+	mutex_unlock(&sysfs_mutex);
+out_valid:
+	return 1;
+out_bad:
+	/* Remove the dentry from the dcache hashes.
+	 * If this is a deleted dentry we use d_drop instead of d_delete
+	 * so sysfs doesn't need to cope with negative dentries.
+	 *
+	 * If this is a dentry that has simply been renamed we
+	 * use d_drop to remove it from the dcache lookup on its
+	 * old parent.  If this dentry persists later when a lookup
+	 * is performed at its new name the dentry will be readded
+	 * to the dcache hashes.
+	 */
+	is_dir = (sysfs_type(sd) == SYSFS_DIR);
+	mutex_unlock(&sysfs_mutex);
+	if (is_dir) {
+		/* If we have submounts we must allow the vfs caches
+		 * to lie about the state of the filesystem to prevent
+		 * leaks and other nasty things.
+		 */
+		if (have_submounts(dentry))
+			goto out_valid;
+		shrink_dcache_parent(dentry);
+	}
+	d_drop(dentry);
+	return 0;
+}
+
+static void sysfs_dentry_iput(struct dentry *dentry, struct inode *inode)
 {
 	struct sysfs_dirent * sd = dentry->d_fsdata;
 
@@ -307,7 +329,9 @@ static void sysfs_d_iput(struct dentry * dentry, struct inode * inode)
 }
 
 static const struct dentry_operations sysfs_dentry_ops = {
-	.d_iput		= sysfs_d_iput,
+	.d_revalidate	= sysfs_dentry_revalidate,
+	.d_delete	= sysfs_dentry_delete,
+	.d_iput		= sysfs_dentry_iput,
 };
 
 struct sysfs_dirent *sysfs_new_dirent(const char *name, umode_t mode, int type)
@@ -330,6 +354,7 @@ struct sysfs_dirent *sysfs_new_dirent(const char *name, umode_t mode, int type)
 
 	atomic_set(&sd->s_count, 1);
 	atomic_set(&sd->s_active, 0);
+	sysfs_dirent_init_lockdep(sd);
 
 	sd->s_name = name;
 	sd->s_mode = mode;
@@ -344,12 +369,6 @@ struct sysfs_dirent *sysfs_new_dirent(const char *name, umode_t mode, int type)
 	return NULL;
 }
 
-static int sysfs_ilookup_test(struct inode *inode, void *arg)
-{
-	struct sysfs_dirent *sd = arg;
-	return inode->i_ino == sd->s_ino;
-}
-
 /**
  *	sysfs_addrm_start - prepare for sysfs_dirent add/remove
  *	@acxt: pointer to sysfs_addrm_cxt to be used
@@ -357,47 +376,20 @@ static int sysfs_ilookup_test(struct inode *inode, void *arg)
  *
  *	This function is called when the caller is about to add or
  *	remove sysfs_dirent under @parent_sd.  This function acquires
- *	sysfs_mutex, grabs inode for @parent_sd if available and lock
- *	i_mutex of it.  @acxt is used to keep and pass context to
+ *	sysfs_mutex.  @acxt is used to keep and pass context to
  *	other addrm functions.
  *
  *	LOCKING:
  *	Kernel thread context (may sleep).  sysfs_mutex is locked on
- *	return.  i_mutex of parent inode is locked on return if
- *	available.
+ *	return.
  */
 void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt,
 		       struct sysfs_dirent *parent_sd)
 {
-	struct inode *inode;
-
 	memset(acxt, 0, sizeof(*acxt));
 	acxt->parent_sd = parent_sd;
 
-	/* Lookup parent inode.  inode initialization is protected by
-	 * sysfs_mutex, so inode existence can be determined by
-	 * looking up inode while holding sysfs_mutex.
-	 */
 	mutex_lock(&sysfs_mutex);
-
-	inode = ilookup5(sysfs_sb, parent_sd->s_ino, sysfs_ilookup_test,
-			 parent_sd);
-	if (inode) {
-		WARN_ON(inode->i_state & I_NEW);
-
-		/* parent inode available */
-		acxt->parent_inode = inode;
-
-		/* sysfs_mutex is below i_mutex in lock hierarchy.
-		 * First, trylock i_mutex.  If fails, unlock
-		 * sysfs_mutex and lock them in order.
-		 */
-		if (!mutex_trylock(&inode->i_mutex)) {
-			mutex_unlock(&sysfs_mutex);
-			mutex_lock(&inode->i_mutex);
-			mutex_lock(&sysfs_mutex);
-		}
-	}
 }
 
 /**
@@ -422,18 +414,22 @@ void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt,
  */
 int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
 {
+	struct sysfs_inode_attrs *ps_iattr;
+
 	if (sysfs_find_dirent(acxt->parent_sd, sd->s_name))
 		return -EEXIST;
 
 	sd->s_parent = sysfs_get(acxt->parent_sd);
 
-	if (sysfs_type(sd) == SYSFS_DIR && acxt->parent_inode)
-		inc_nlink(acxt->parent_inode);
-
-	acxt->cnt++;
-
 	sysfs_link_sibling(sd);
 
+	/* Update timestamps on the parent */
+	ps_iattr = acxt->parent_sd->s_iattr;
+	if (ps_iattr) {
+		struct iattr *ps_iattrs = &ps_iattr->ia_iattr;
+		ps_iattrs->ia_ctime = ps_iattrs->ia_mtime = CURRENT_TIME;
+	}
+
 	return 0;
 }
 
@@ -512,70 +508,22 @@ int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
  */
 void sysfs_remove_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
 {
+	struct sysfs_inode_attrs *ps_iattr;
+
 	BUG_ON(sd->s_flags & SYSFS_FLAG_REMOVED);
 
 	sysfs_unlink_sibling(sd);
 
+	/* Update timestamps on the parent */
+	ps_iattr = acxt->parent_sd->s_iattr;
+	if (ps_iattr) {
+		struct iattr *ps_iattrs = &ps_iattr->ia_iattr;
+		ps_iattrs->ia_ctime = ps_iattrs->ia_mtime = CURRENT_TIME;
+	}
+
 	sd->s_flags |= SYSFS_FLAG_REMOVED;
 	sd->s_sibling = acxt->removed;
 	acxt->removed = sd;
-
-	if (sysfs_type(sd) == SYSFS_DIR && acxt->parent_inode)
-		drop_nlink(acxt->parent_inode);
-
-	acxt->cnt++;
-}
-
-/**
- *	sysfs_drop_dentry - drop dentry for the specified sysfs_dirent
- *	@sd: target sysfs_dirent
- *
- *	Drop dentry for @sd.  @sd must have been unlinked from its
- *	parent on entry to this function such that it can't be looked
- *	up anymore.
- */
-static void sysfs_drop_dentry(struct sysfs_dirent *sd)
-{
-	struct inode *inode;
-	struct dentry *dentry;
-
-	inode = ilookup(sysfs_sb, sd->s_ino);
-	if (!inode)
-		return;
-
-	/* Drop any existing dentries associated with sd.
-	 *
-	 * For the dentry to be properly freed we need to grab a
-	 * reference to the dentry under the dcache lock,  unhash it,
-	 * and then put it.  The playing with the dentry count allows
-	 * dput to immediately free the dentry  if it is not in use.
-	 */
-repeat:
-	spin_lock(&dcache_lock);
-	list_for_each_entry(dentry, &inode->i_dentry, d_alias) {
-		if (d_unhashed(dentry))
-			continue;
-		dget_locked(dentry);
-		spin_lock(&dentry->d_lock);
-		__d_drop(dentry);
-		spin_unlock(&dentry->d_lock);
-		spin_unlock(&dcache_lock);
-		dput(dentry);
-		goto repeat;
-	}
-	spin_unlock(&dcache_lock);
-
-	/* adjust nlink and update timestamp */
-	mutex_lock(&inode->i_mutex);
-
-	inode->i_ctime = CURRENT_TIME;
-	drop_nlink(inode);
-	if (sysfs_type(sd) == SYSFS_DIR)
-		drop_nlink(inode);
-
-	mutex_unlock(&inode->i_mutex);
-
-	iput(inode);
 }
 
 /**
@@ -584,25 +532,15 @@ repeat:
  *
  *	Finish up sysfs_dirent add/remove.  Resources acquired by
  *	sysfs_addrm_start() are released and removed sysfs_dirents are
- *	cleaned up.  Timestamps on the parent inode are updated.
+ *	cleaned up.
  *
  *	LOCKING:
- *	All mutexes acquired by sysfs_addrm_start() are released.
+ *	sysfs_mutex is released.
  */
 void sysfs_addrm_finish(struct sysfs_addrm_cxt *acxt)
 {
 	/* release resources acquired by sysfs_addrm_start() */
 	mutex_unlock(&sysfs_mutex);
-	if (acxt->parent_inode) {
-		struct inode *inode = acxt->parent_inode;
-
-		/* if added/removed, update timestamps on the parent */
-		if (acxt->cnt)
-			inode->i_ctime = inode->i_mtime = CURRENT_TIME;
-
-		mutex_unlock(&inode->i_mutex);
-		iput(inode);
-	}
 
 	/* kill removed sysfs_dirents */
 	while (acxt->removed) {
@@ -611,7 +549,6 @@ void sysfs_addrm_finish(struct sysfs_addrm_cxt *acxt)
 		acxt->removed = sd->s_sibling;
 		sd->s_sibling = NULL;
 
-		sysfs_drop_dentry(sd);
 		sysfs_deactivate(sd);
 		unmap_bin_file(sd);
 		sysfs_put(sd);
@@ -751,10 +688,15 @@ static struct dentry * sysfs_lookup(struct inode *dir, struct dentry *dentry,
 	}
 
 	/* instantiate and hash dentry */
-	dentry->d_op = &sysfs_dentry_ops;
-	dentry->d_fsdata = sysfs_get(sd);
-	d_instantiate(dentry, inode);
-	d_rehash(dentry);
+	ret = d_find_alias(inode);
+	if (!ret) {
+		dentry->d_op = &sysfs_dentry_ops;
+		dentry->d_fsdata = sysfs_get(sd);
+		d_add(dentry, inode);
+	} else {
+		d_move(ret, dentry);
+		iput(inode);
+	}
 
  out_unlock:
 	mutex_unlock(&sysfs_mutex);
@@ -763,7 +705,9 @@ static struct dentry * sysfs_lookup(struct inode *dir, struct dentry *dentry,
 
 const struct inode_operations sysfs_dir_inode_operations = {
 	.lookup		= sysfs_lookup,
+	.permission	= sysfs_permission,
 	.setattr	= sysfs_setattr,
+	.getattr	= sysfs_getattr,
 	.setxattr	= sysfs_setxattr,
 };
 
@@ -826,141 +770,65 @@ void sysfs_remove_dir(struct kobject * kobj)
 	__sysfs_remove_dir(sd);
 }
 
-int sysfs_rename_dir(struct kobject * kobj, const char *new_name)
+int sysfs_rename(struct sysfs_dirent *sd,
+	struct sysfs_dirent *new_parent_sd, const char *new_name)
 {
-	struct sysfs_dirent *sd = kobj->sd;
-	struct dentry *parent = NULL;
-	struct dentry *old_dentry = NULL, *new_dentry = NULL;
 	const char *dup_name = NULL;
 	int error;
 
-	mutex_lock(&sysfs_rename_mutex);
+	mutex_lock(&sysfs_mutex);
 
 	error = 0;
-	if (strcmp(sd->s_name, new_name) == 0)
+	if ((sd->s_parent == new_parent_sd) &&
+	    (strcmp(sd->s_name, new_name) == 0))
 		goto out;	/* nothing to rename */
 
-	/* get the original dentry */
-	old_dentry = sysfs_get_dentry(sd);
-	if (IS_ERR(old_dentry)) {
-		error = PTR_ERR(old_dentry);
-		old_dentry = NULL;
-		goto out;
-	}
-
-	parent = old_dentry->d_parent;
-
-	/* lock parent and get dentry for new name */
-	mutex_lock(&parent->d_inode->i_mutex);
-	mutex_lock(&sysfs_mutex);
-
 	error = -EEXIST;
-	if (sysfs_find_dirent(sd->s_parent, new_name))
-		goto out_unlock;
-
-	error = -ENOMEM;
-	new_dentry = d_alloc_name(parent, new_name);
-	if (!new_dentry)
-		goto out_unlock;
+	if (sysfs_find_dirent(new_parent_sd, new_name))
+		goto out;
 
 	/* rename sysfs_dirent */
-	error = -ENOMEM;
-	new_name = dup_name = kstrdup(new_name, GFP_KERNEL);
-	if (!new_name)
-		goto out_unlock;
-
-	dup_name = sd->s_name;
-	sd->s_name = new_name;
+	if (strcmp(sd->s_name, new_name) != 0) {
+		error = -ENOMEM;
+		new_name = dup_name = kstrdup(new_name, GFP_KERNEL);
+		if (!new_name)
+			goto out;
+
+		dup_name = sd->s_name;
+		sd->s_name = new_name;
+	}
 
-	/* rename */
-	d_add(new_dentry, NULL);
-	d_move(old_dentry, new_dentry);
+	/* Remove from old parent's list and insert into new parent's list. */
+	if (sd->s_parent != new_parent_sd) {
+		sysfs_unlink_sibling(sd);
+		sysfs_get(new_parent_sd);
+		sysfs_put(sd->s_parent);
+		sd->s_parent = new_parent_sd;
+		sysfs_link_sibling(sd);
+	}
 
 	error = 0;
- out_unlock:
+ out:
 	mutex_unlock(&sysfs_mutex);
-	mutex_unlock(&parent->d_inode->i_mutex);
 	kfree(dup_name);
-	dput(old_dentry);
-	dput(new_dentry);
- out:
-	mutex_unlock(&sysfs_rename_mutex);
 	return error;
 }
 
+int sysfs_rename_dir(struct kobject *kobj, const char *new_name)
+{
+	return sysfs_rename(kobj->sd, kobj->sd->s_parent, new_name);
+}
+
 int sysfs_move_dir(struct kobject *kobj, struct kobject *new_parent_kobj)
 {
 	struct sysfs_dirent *sd = kobj->sd;
 	struct sysfs_dirent *new_parent_sd;
-	struct dentry *old_parent, *new_parent = NULL;
-	struct dentry *old_dentry = NULL, *new_dentry = NULL;
-	int error;
 
-	mutex_lock(&sysfs_rename_mutex);
 	BUG_ON(!sd->s_parent);
-	new_parent_sd = (new_parent_kobj && new_parent_kobj->sd) ?
+	new_parent_sd = new_parent_kobj && new_parent_kobj->sd ?
 		new_parent_kobj->sd : &sysfs_root;
 
-	error = 0;
-	if (sd->s_parent == new_parent_sd)
-		goto out;	/* nothing to move */
-
-	/* get dentries */
-	old_dentry = sysfs_get_dentry(sd);
-	if (IS_ERR(old_dentry)) {
-		error = PTR_ERR(old_dentry);
-		old_dentry = NULL;
-		goto out;
-	}
-	old_parent = old_dentry->d_parent;
-
-	new_parent = sysfs_get_dentry(new_parent_sd);
-	if (IS_ERR(new_parent)) {
-		error = PTR_ERR(new_parent);
-		new_parent = NULL;
-		goto out;
-	}
-
-again:
-	mutex_lock(&old_parent->d_inode->i_mutex);
-	if (!mutex_trylock(&new_parent->d_inode->i_mutex)) {
-		mutex_unlock(&old_parent->d_inode->i_mutex);
-		goto again;
-	}
-	mutex_lock(&sysfs_mutex);
-
-	error = -EEXIST;
-	if (sysfs_find_dirent(new_parent_sd, sd->s_name))
-		goto out_unlock;
-
-	error = -ENOMEM;
-	new_dentry = d_alloc_name(new_parent, sd->s_name);
-	if (!new_dentry)
-		goto out_unlock;
-
-	error = 0;
-	d_add(new_dentry, NULL);
-	d_move(old_dentry, new_dentry);
-
-	/* Remove from old parent's list and insert into new parent's list. */
-	sysfs_unlink_sibling(sd);
-	sysfs_get(new_parent_sd);
-	drop_nlink(old_parent->d_inode);
-	sysfs_put(sd->s_parent);
-	sd->s_parent = new_parent_sd;
-	inc_nlink(new_parent->d_inode);
-	sysfs_link_sibling(sd);
-
- out_unlock:
-	mutex_unlock(&sysfs_mutex);
-	mutex_unlock(&new_parent->d_inode->i_mutex);
-	mutex_unlock(&old_parent->d_inode->i_mutex);
- out:
-	dput(new_parent);
-	dput(old_dentry);
-	dput(new_dentry);
-	mutex_unlock(&sysfs_rename_mutex);
-	return error;
+	return sysfs_rename(sd, new_parent_sd, sd->s_name);
 }
 
 /* Relationship between s_mode and the DT_xxx types */
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index f5ea4680f15f..dc30d9e31683 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -579,46 +579,23 @@ EXPORT_SYMBOL_GPL(sysfs_add_file_to_group);
  */
 int sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode)
 {
-	struct sysfs_dirent *victim_sd = NULL;
-	struct dentry *victim = NULL;
-	struct inode * inode;
+	struct sysfs_dirent *sd;
 	struct iattr newattrs;
 	int rc;
 
-	rc = -ENOENT;
-	victim_sd = sysfs_get_dirent(kobj->sd, attr->name);
-	if (!victim_sd)
-		goto out;
+	mutex_lock(&sysfs_mutex);
 
-	mutex_lock(&sysfs_rename_mutex);
-	victim = sysfs_get_dentry(victim_sd);
-	mutex_unlock(&sysfs_rename_mutex);
-	if (IS_ERR(victim)) {
-		rc = PTR_ERR(victim);
-		victim = NULL;
+	rc = -ENOENT;
+	sd = sysfs_find_dirent(kobj->sd, attr->name);
+	if (!sd)
 		goto out;
-	}
-
-	inode = victim->d_inode;
-
-	mutex_lock(&inode->i_mutex);
 
-	newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO);
-	newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
-	newattrs.ia_ctime = current_fs_time(inode->i_sb);
-	rc = sysfs_setattr(victim, &newattrs);
+	newattrs.ia_mode = (mode & S_IALLUGO) | (sd->s_mode & ~S_IALLUGO);
+	newattrs.ia_valid = ATTR_MODE;
+	rc = sysfs_sd_setattr(sd, &newattrs);
 
-	if (rc == 0) {
-		fsnotify_change(victim, newattrs.ia_valid);
-		mutex_lock(&sysfs_mutex);
-		victim_sd->s_mode = newattrs.ia_mode;
-		mutex_unlock(&sysfs_mutex);
-	}
-
-	mutex_unlock(&inode->i_mutex);
  out:
-	dput(victim);
-	sysfs_put(victim_sd);
+	mutex_unlock(&sysfs_mutex);
 	return rc;
 }
 EXPORT_SYMBOL_GPL(sysfs_chmod_file);
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index e28cecf179f5..220b758523ae 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -37,7 +37,9 @@ static struct backing_dev_info sysfs_backing_dev_info = {
 };
 
 static const struct inode_operations sysfs_inode_operations ={
+	.permission	= sysfs_permission,
 	.setattr	= sysfs_setattr,
+	.getattr	= sysfs_getattr,
 	.setxattr	= sysfs_setxattr,
 };
 
@@ -46,7 +48,7 @@ int __init sysfs_inode_init(void)
 	return bdi_init(&sysfs_backing_dev_info);
 }
 
-struct sysfs_inode_attrs *sysfs_init_inode_attrs(struct sysfs_dirent *sd)
+static struct sysfs_inode_attrs *sysfs_init_inode_attrs(struct sysfs_dirent *sd)
 {
 	struct sysfs_inode_attrs *attrs;
 	struct iattr *iattrs;
@@ -64,30 +66,15 @@ struct sysfs_inode_attrs *sysfs_init_inode_attrs(struct sysfs_dirent *sd)
 
 	return attrs;
 }
-int sysfs_setattr(struct dentry * dentry, struct iattr * iattr)
+
+int sysfs_sd_setattr(struct sysfs_dirent *sd, struct iattr * iattr)
 {
-	struct inode * inode = dentry->d_inode;
-	struct sysfs_dirent * sd = dentry->d_fsdata;
 	struct sysfs_inode_attrs *sd_attrs;
 	struct iattr *iattrs;
 	unsigned int ia_valid = iattr->ia_valid;
-	int error;
-
-	if (!sd)
-		return -EINVAL;
 
 	sd_attrs = sd->s_iattr;
 
-	error = inode_change_ok(inode, iattr);
-	if (error)
-		return error;
-
-	iattr->ia_valid &= ~ATTR_SIZE; /* ignore size changes */
-
-	error = inode_setattr(inode, iattr);
-	if (error)
-		return error;
-
 	if (!sd_attrs) {
 		/* setting attributes for the first time, allocate now */
 		sd_attrs = sysfs_init_inode_attrs(sd);
@@ -103,42 +90,78 @@ int sysfs_setattr(struct dentry * dentry, struct iattr * iattr)
 		if (ia_valid & ATTR_GID)
 			iattrs->ia_gid = iattr->ia_gid;
 		if (ia_valid & ATTR_ATIME)
-			iattrs->ia_atime = timespec_trunc(iattr->ia_atime,
-					inode->i_sb->s_time_gran);
+			iattrs->ia_atime = iattr->ia_atime;
 		if (ia_valid & ATTR_MTIME)
-			iattrs->ia_mtime = timespec_trunc(iattr->ia_mtime,
-					inode->i_sb->s_time_gran);
+			iattrs->ia_mtime = iattr->ia_mtime;
 		if (ia_valid & ATTR_CTIME)
-			iattrs->ia_ctime = timespec_trunc(iattr->ia_ctime,
-					inode->i_sb->s_time_gran);
+			iattrs->ia_ctime = iattr->ia_ctime;
 		if (ia_valid & ATTR_MODE) {
 			umode_t mode = iattr->ia_mode;
-
-			if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
-				mode &= ~S_ISGID;
 			iattrs->ia_mode = sd->s_mode = mode;
 		}
 	}
+	return 0;
+}
+
+int sysfs_setattr(struct dentry *dentry, struct iattr *iattr)
+{
+	struct inode *inode = dentry->d_inode;
+	struct sysfs_dirent *sd = dentry->d_fsdata;
+	int error;
+
+	if (!sd)
+		return -EINVAL;
+
+	error = inode_change_ok(inode, iattr);
+	if (error)
+		return error;
+
+	iattr->ia_valid &= ~ATTR_SIZE; /* ignore size changes */
+
+	error = inode_setattr(inode, iattr);
+	if (error)
+		return error;
+
+	mutex_lock(&sysfs_mutex);
+	error = sysfs_sd_setattr(sd, iattr);
+	mutex_unlock(&sysfs_mutex);
+
 	return error;
 }
 
+static int sysfs_sd_setsecdata(struct sysfs_dirent *sd, void **secdata, u32 *secdata_len)
+{
+	struct sysfs_inode_attrs *iattrs;
+	void *old_secdata;
+	size_t old_secdata_len;
+
+	iattrs = sd->s_iattr;
+	if (!iattrs)
+		iattrs = sysfs_init_inode_attrs(sd);
+	if (!iattrs)
+		return -ENOMEM;
+
+	old_secdata = iattrs->ia_secdata;
+	old_secdata_len = iattrs->ia_secdata_len;
+
+	iattrs->ia_secdata = *secdata;
+	iattrs->ia_secdata_len = *secdata_len;
+
+	*secdata = old_secdata;
+	*secdata_len = old_secdata_len;
+	return 0;
+}
+
 int sysfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 		size_t size, int flags)
 {
 	struct sysfs_dirent *sd = dentry->d_fsdata;
-	struct sysfs_inode_attrs *iattrs;
 	void *secdata;
 	int error;
 	u32 secdata_len = 0;
 
 	if (!sd)
 		return -EINVAL;
-	if (!sd->s_iattr)
-		sd->s_iattr = sysfs_init_inode_attrs(sd);
-	if (!sd->s_iattr)
-		return -ENOMEM;
-
-	iattrs = sd->s_iattr;
 
 	if (!strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN)) {
 		const char *suffix = name + XATTR_SECURITY_PREFIX_LEN;
@@ -150,12 +173,13 @@ int sysfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 						&secdata, &secdata_len);
 		if (error)
 			goto out;
-		if (iattrs->ia_secdata)
-			security_release_secctx(iattrs->ia_secdata,
-						iattrs->ia_secdata_len);
-		iattrs->ia_secdata = secdata;
-		iattrs->ia_secdata_len = secdata_len;
 
+		mutex_lock(&sysfs_mutex);
+		error = sysfs_sd_setsecdata(sd, &secdata, &secdata_len);
+		mutex_unlock(&sysfs_mutex);
+
+		if (secdata)
+			security_release_secctx(secdata, secdata_len);
 	} else
 		return -EINVAL;
 out:
@@ -170,7 +194,6 @@ static inline void set_default_inode_attr(struct inode * inode, mode_t mode)
 
 static inline void set_inode_attr(struct inode * inode, struct iattr * iattr)
 {
-	inode->i_mode = iattr->ia_mode;
 	inode->i_uid = iattr->ia_uid;
 	inode->i_gid = iattr->ia_gid;
 	inode->i_atime = iattr->ia_atime;
@@ -178,17 +201,6 @@ static inline void set_inode_attr(struct inode * inode, struct iattr * iattr)
 	inode->i_ctime = iattr->ia_ctime;
 }
 
-
-/*
- * sysfs has a different i_mutex lock order behavior for i_mutex than other
- * filesystems; sysfs i_mutex is called in many places with subsystem locks
- * held. At the same time, many of the VFS locking rules do not apply to
- * sysfs at all (cross directory rename for example). To untangle this mess
- * (which gives false positives in lockdep), we're giving sysfs inodes their
- * own class for i_mutex.
- */
-static struct lock_class_key sysfs_inode_imutex_key;
-
 static int sysfs_count_nlink(struct sysfs_dirent *sd)
 {
 	struct sysfs_dirent *child;
@@ -201,38 +213,55 @@ static int sysfs_count_nlink(struct sysfs_dirent *sd)
 	return nr + 2;
 }
 
+static void sysfs_refresh_inode(struct sysfs_dirent *sd, struct inode *inode)
+{
+	struct sysfs_inode_attrs *iattrs = sd->s_iattr;
+
+	inode->i_mode = sd->s_mode;
+	if (iattrs) {
+		/* sysfs_dirent has non-default attributes
+		 * get them from persistent copy in sysfs_dirent
+		 */
+		set_inode_attr(inode, &iattrs->ia_iattr);
+		security_inode_notifysecctx(inode,
+					    iattrs->ia_secdata,
+					    iattrs->ia_secdata_len);
+	}
+
+	if (sysfs_type(sd) == SYSFS_DIR)
+		inode->i_nlink = sysfs_count_nlink(sd);
+}
+
+int sysfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
+{
+	struct sysfs_dirent *sd = dentry->d_fsdata;
+	struct inode *inode = dentry->d_inode;
+
+	mutex_lock(&sysfs_mutex);
+	sysfs_refresh_inode(sd, inode);
+	mutex_unlock(&sysfs_mutex);
+
+	generic_fillattr(inode, stat);
+	return 0;
+}
+
 static void sysfs_init_inode(struct sysfs_dirent *sd, struct inode *inode)
 {
 	struct bin_attribute *bin_attr;
-	struct sysfs_inode_attrs *iattrs;
 
 	inode->i_private = sysfs_get(sd);
 	inode->i_mapping->a_ops = &sysfs_aops;
 	inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info;
 	inode->i_op = &sysfs_inode_operations;
-	inode->i_ino = sd->s_ino;
-	lockdep_set_class(&inode->i_mutex, &sysfs_inode_imutex_key);
 
-	iattrs = sd->s_iattr;
-	if (iattrs) {
-		/* sysfs_dirent has non-default attributes
-		 * get them for the new inode from persistent copy
-		 * in sysfs_dirent
-		 */
-		set_inode_attr(inode, &iattrs->ia_iattr);
-		if (iattrs->ia_secdata)
-			security_inode_notifysecctx(inode,
-						iattrs->ia_secdata,
-						iattrs->ia_secdata_len);
-	} else
-		set_default_inode_attr(inode, sd->s_mode);
+	set_default_inode_attr(inode, sd->s_mode);
+	sysfs_refresh_inode(sd, inode);
 
 	/* initialize inode according to type */
 	switch (sysfs_type(sd)) {
 	case SYSFS_DIR:
 		inode->i_op = &sysfs_dir_inode_operations;
 		inode->i_fop = &sysfs_dir_operations;
-		inode->i_nlink = sysfs_count_nlink(sd);
 		break;
 	case SYSFS_KOBJ_ATTR:
 		inode->i_size = PAGE_SIZE;
@@ -315,3 +344,14 @@ int sysfs_hash_and_remove(struct sysfs_dirent *dir_sd, const char *name)
 	else
 		return -ENOENT;
 }
+
+int sysfs_permission(struct inode *inode, int mask)
+{
+	struct sysfs_dirent *sd = inode->i_private;
+
+	mutex_lock(&sysfs_mutex);
+	sysfs_refresh_inode(sd, inode);
+	mutex_unlock(&sysfs_mutex);
+
+	return generic_permission(inode, mask, NULL);
+}
diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c
index c5081ad77026..c5eff49fa41b 100644
--- a/fs/sysfs/symlink.c
+++ b/fs/sysfs/symlink.c
@@ -210,10 +210,13 @@ static void sysfs_put_link(struct dentry *dentry, struct nameidata *nd, void *co
 }
 
 const struct inode_operations sysfs_symlink_inode_operations = {
-	.setxattr = sysfs_setxattr,
-	.readlink = generic_readlink,
-	.follow_link = sysfs_follow_link,
-	.put_link = sysfs_put_link,
+	.setxattr	= sysfs_setxattr,
+	.readlink	= generic_readlink,
+	.follow_link	= sysfs_follow_link,
+	.put_link	= sysfs_put_link,
+	.setattr	= sysfs_setattr,
+	.getattr	= sysfs_getattr,
+	.permission	= sysfs_permission,
 };
 
 
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index af4c4e7482ac..cdd9377a6e06 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -8,6 +8,7 @@
  * This file is released under the GPLv2.
  */
 
+#include <linux/lockdep.h>
 #include <linux/fs.h>
 
 struct sysfs_open_dirent;
@@ -50,6 +51,9 @@ struct sysfs_inode_attrs {
 struct sysfs_dirent {
 	atomic_t		s_count;
 	atomic_t		s_active;
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+	struct lockdep_map	dep_map;
+#endif
 	struct sysfs_dirent	*s_parent;
 	struct sysfs_dirent	*s_sibling;
 	const char		*s_name;
@@ -84,14 +88,23 @@ static inline unsigned int sysfs_type(struct sysfs_dirent *sd)
 	return sd->s_flags & SYSFS_TYPE_MASK;
 }
 
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+#define sysfs_dirent_init_lockdep(sd)				\
+do {								\
+	static struct lock_class_key __key;			\
+								\
+	lockdep_init_map(&sd->dep_map, "s_active", &__key, 0);	\
+} while(0)
+#else
+#define sysfs_dirent_init_lockdep(sd) do {} while(0)
+#endif
+
 /*
  * Context structure to be used while adding/removing nodes.
  */
 struct sysfs_addrm_cxt {
 	struct sysfs_dirent	*parent_sd;
-	struct inode		*parent_inode;
 	struct sysfs_dirent	*removed;
-	int			cnt;
 };
 
 /*
@@ -105,7 +118,6 @@ extern struct kmem_cache *sysfs_dir_cachep;
  * dir.c
  */
 extern struct mutex sysfs_mutex;
-extern struct mutex sysfs_rename_mutex;
 extern spinlock_t sysfs_assoc_lock;
 
 extern const struct file_operations sysfs_dir_operations;
@@ -133,6 +145,9 @@ int sysfs_create_subdir(struct kobject *kobj, const char *name,
 			struct sysfs_dirent **p_sd);
 void sysfs_remove_subdir(struct sysfs_dirent *sd);
 
+int sysfs_rename(struct sysfs_dirent *sd,
+	struct sysfs_dirent *new_parent_sd, const char *new_name);
+
 static inline struct sysfs_dirent *__sysfs_get(struct sysfs_dirent *sd)
 {
 	if (sd) {
@@ -155,7 +170,10 @@ static inline void __sysfs_put(struct sysfs_dirent *sd)
  */
 struct inode *sysfs_get_inode(struct sysfs_dirent *sd);
 void sysfs_delete_inode(struct inode *inode);
+int sysfs_sd_setattr(struct sysfs_dirent *sd, struct iattr *iattr);
+int sysfs_permission(struct inode *inode, int mask);
 int sysfs_setattr(struct dentry *dentry, struct iattr *iattr);
+int sysfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat);
 int sysfs_setxattr(struct dentry *dentry, const char *name, const void *value,
 		size_t size, int flags);
 int sysfs_hash_and_remove(struct sysfs_dirent *dir_sd, const char *name);
diff --git a/fs/timerfd.c b/fs/timerfd.c
index b042bd7034b1..1bfc95ad5f71 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -200,7 +200,7 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags)
 	hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS);
 
 	ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx,
-			       flags & TFD_SHARED_FCNTL_FLAGS);
+			       O_RDWR | (flags & TFD_SHARED_FCNTL_FLAGS));
 	if (ufd < 0)
 		kfree(ctx);
 
diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c
index dbc093afd946..90492327b383 100644
--- a/fs/ubifs/debug.c
+++ b/fs/ubifs/debug.c
@@ -350,13 +350,8 @@ void dbg_dump_node(const struct ubifs_info *c, const void *node)
 		       le32_to_cpu(sup->fmt_version));
 		printk(KERN_DEBUG "\ttime_gran      %u\n",
 		       le32_to_cpu(sup->time_gran));
-		printk(KERN_DEBUG "\tUUID           %02X%02X%02X%02X-%02X%02X"
-		       "-%02X%02X-%02X%02X-%02X%02X%02X%02X%02X%02X\n",
-		       sup->uuid[0], sup->uuid[1], sup->uuid[2], sup->uuid[3],
-		       sup->uuid[4], sup->uuid[5], sup->uuid[6], sup->uuid[7],
-		       sup->uuid[8], sup->uuid[9], sup->uuid[10], sup->uuid[11],
-		       sup->uuid[12], sup->uuid[13], sup->uuid[14],
-		       sup->uuid[15]);
+		printk(KERN_DEBUG "\tUUID           %pUB\n",
+		       sup->uuid);
 		break;
 	}
 	case UBIFS_MST_NODE:
@@ -2014,7 +2009,7 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr,
 		inum = key_inum_flash(c, &dent->key);
 		fscki1 = read_add_inode(c, priv, inum);
 		if (IS_ERR(fscki1)) {
-			err = PTR_ERR(fscki);
+			err = PTR_ERR(fscki1);
 			ubifs_err("error %d while processing entry node and "
 				  "trying to find parent inode node %lu",
 				  err, (unsigned long)inum);
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index 1009adc8d602..16a6444330ec 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -45,7 +45,7 @@
  *
  * Similarly, @i_mutex is not always locked in 'ubifs_readpage()', e.g., the
  * read-ahead path does not lock it ("sys_read -> generic_file_aio_read ->
- * ondemand_readahead -> readpage"). In case of readahead, @I_LOCK flag is not
+ * ondemand_readahead -> readpage"). In case of readahead, @I_SYNC flag is not
  * set as well. However, UBIFS disables readahead.
  */
 
@@ -1389,7 +1389,6 @@ static ssize_t ubifs_aio_write(struct kiocb *iocb, const struct iovec *iov,
 			       unsigned long nr_segs, loff_t pos)
 {
 	int err;
-	ssize_t ret;
 	struct inode *inode = iocb->ki_filp->f_mapping->host;
 	struct ubifs_info *c = inode->i_sb->s_fs_info;
 
@@ -1397,17 +1396,7 @@ static ssize_t ubifs_aio_write(struct kiocb *iocb, const struct iovec *iov,
 	if (err)
 		return err;
 
-	ret = generic_file_aio_write(iocb, iov, nr_segs, pos);
-	if (ret < 0)
-		return ret;
-
-	if (ret > 0 && (IS_SYNC(inode) || iocb->ki_filp->f_flags & O_SYNC)) {
-		err = ubifs_sync_wbufs_by_inode(c, inode);
-		if (err)
-			return err;
-	}
-
-	return ret;
+	return generic_file_aio_write(iocb, iov, nr_segs, pos);
 }
 
 static int ubifs_set_page_dirty(struct page *page)
diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c
index 618c2701d3a7..e5a3d8e96bb7 100644
--- a/fs/ubifs/gc.c
+++ b/fs/ubifs/gc.c
@@ -54,6 +54,7 @@
  */
 
 #include <linux/pagemap.h>
+#include <linux/list_sort.h>
 #include "ubifs.h"
 
 /*
@@ -108,101 +109,6 @@ static int switch_gc_head(struct ubifs_info *c)
 }
 
 /**
- * list_sort - sort a list.
- * @priv: private data, passed to @cmp
- * @head: the list to sort
- * @cmp: the elements comparison function
- *
- * This function has been implemented by Mark J Roberts <mjr@znex.org>. It
- * implements "merge sort" which has O(nlog(n)) complexity. The list is sorted
- * in ascending order.
- *
- * The comparison function @cmp is supposed to return a negative value if @a is
- * than @b, and a positive value if @a is greater than @b. If @a and @b are
- * equivalent, then it does not matter what this function returns.
- */
-static void list_sort(void *priv, struct list_head *head,
-		      int (*cmp)(void *priv, struct list_head *a,
-				 struct list_head *b))
-{
-	struct list_head *p, *q, *e, *list, *tail, *oldhead;
-	int insize, nmerges, psize, qsize, i;
-
-	if (list_empty(head))
-		return;
-
-	list = head->next;
-	list_del(head);
-	insize = 1;
-	for (;;) {
-		p = oldhead = list;
-		list = tail = NULL;
-		nmerges = 0;
-
-		while (p) {
-			nmerges++;
-			q = p;
-			psize = 0;
-			for (i = 0; i < insize; i++) {
-				psize++;
-				q = q->next == oldhead ? NULL : q->next;
-				if (!q)
-					break;
-			}
-
-			qsize = insize;
-			while (psize > 0 || (qsize > 0 && q)) {
-				if (!psize) {
-					e = q;
-					q = q->next;
-					qsize--;
-					if (q == oldhead)
-						q = NULL;
-				} else if (!qsize || !q) {
-					e = p;
-					p = p->next;
-					psize--;
-					if (p == oldhead)
-						p = NULL;
-				} else if (cmp(priv, p, q) <= 0) {
-					e = p;
-					p = p->next;
-					psize--;
-					if (p == oldhead)
-						p = NULL;
-				} else {
-					e = q;
-					q = q->next;
-					qsize--;
-					if (q == oldhead)
-						q = NULL;
-				}
-				if (tail)
-					tail->next = e;
-				else
-					list = e;
-				e->prev = tail;
-				tail = e;
-			}
-			p = q;
-		}
-
-		tail->next = list;
-		list->prev = tail;
-
-		if (nmerges <= 1)
-			break;
-
-		insize *= 2;
-	}
-
-	head->next = list;
-	head->prev = list->prev;
-	list->prev->next = head;
-	list->prev = head;
-}
-
-/**
  * data_nodes_cmp - compare 2 data nodes.
  * @priv: UBIFS file-system description object
  * @a: first data node
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 333e181ee987..43f9d19a6f33 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -1393,12 +1393,7 @@ static int mount_ubifs(struct ubifs_info *c)
 		c->leb_size, c->leb_size >> 10);
 	dbg_msg("data journal heads:  %d",
 		c->jhead_cnt - NONDATA_JHEADS_CNT);
-	dbg_msg("UUID:                %02X%02X%02X%02X-%02X%02X"
-	       "-%02X%02X-%02X%02X-%02X%02X%02X%02X%02X%02X",
-	       c->uuid[0], c->uuid[1], c->uuid[2], c->uuid[3],
-	       c->uuid[4], c->uuid[5], c->uuid[6], c->uuid[7],
-	       c->uuid[8], c->uuid[9], c->uuid[10], c->uuid[11],
-	       c->uuid[12], c->uuid[13], c->uuid[14], c->uuid[15]);
+	dbg_msg("UUID:                %pUB", c->uuid);
 	dbg_msg("big_lpt              %d", c->big_lpt);
 	dbg_msg("log LEBs:            %d (%d - %d)",
 		c->log_lebs, UBIFS_LOG_LNUM, c->log_last);
@@ -1842,22 +1837,32 @@ const struct super_operations ubifs_super_operations = {
  * @name: UBI volume name
  * @mode: UBI volume open mode
  *
- * There are several ways to specify UBI volumes when mounting UBIFS:
- * o ubiX_Y    - UBI device number X, volume Y;
- * o ubiY      - UBI device number 0, volume Y;
+ * The primary method of mounting UBIFS is by specifying the UBI volume
+ * character device node path. However, UBIFS may also be mounted withoug any
+ * character device node using one of the following methods:
+ *
+ * o ubiX_Y    - mount UBI device number X, volume Y;
+ * o ubiY      - mount UBI device number 0, volume Y;
  * o ubiX:NAME - mount UBI device X, volume with name NAME;
  * o ubi:NAME  - mount UBI device 0, volume with name NAME.
  *
  * Alternative '!' separator may be used instead of ':' (because some shells
  * like busybox may interpret ':' as an NFS host name separator). This function
- * returns ubi volume object in case of success and a negative error code in
- * case of failure.
+ * returns UBI volume description object in case of success and a negative
+ * error code in case of failure.
  */
 static struct ubi_volume_desc *open_ubi(const char *name, int mode)
 {
+	struct ubi_volume_desc *ubi;
 	int dev, vol;
 	char *endptr;
 
+	/* First, try to open using the device node path method */
+	ubi = ubi_open_volume_path(name, mode);
+	if (!IS_ERR(ubi))
+		return ubi;
+
+	/* Try the "nodev" method */
 	if (name[0] != 'u' || name[1] != 'b' || name[2] != 'i')
 		return ERR_PTR(-EINVAL);
 
diff --git a/fs/udf/balloc.c b/fs/udf/balloc.c
index 1e068535b58b..82372e332f08 100644
--- a/fs/udf/balloc.c
+++ b/fs/udf/balloc.c
@@ -440,7 +440,7 @@ static void udf_table_free_blocks(struct super_block *sb,
 	    (bloc->logicalBlockNum + count) >
 		partmap->s_partition_len) {
 		udf_debug("%d < %d || %d + %d > %d\n",
-			  bloc.logicalBlockNum, 0, bloc.logicalBlockNum, count,
+			  bloc->logicalBlockNum, 0, bloc->logicalBlockNum, count,
 			  partmap->s_partition_len);
 		goto error_return;
 	}
diff --git a/fs/udf/file.c b/fs/udf/file.c
index b80cbd78833c..f311d509b6a3 100644
--- a/fs/udf/file.c
+++ b/fs/udf/file.c
@@ -196,6 +196,7 @@ static int udf_release_file(struct inode *inode, struct file *filp)
 		mutex_lock(&inode->i_mutex);
 		lock_kernel();
 		udf_discard_prealloc(inode);
+		udf_truncate_tail_extent(inode);
 		unlock_kernel();
 		mutex_unlock(&inode->i_mutex);
 	}
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index 6d24c2c63f93..f90231eb2916 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -97,15 +97,17 @@ no_delete:
  */
 void udf_clear_inode(struct inode *inode)
 {
-	struct udf_inode_info *iinfo;
-	if (!(inode->i_sb->s_flags & MS_RDONLY)) {
-		lock_kernel();
-		udf_truncate_tail_extent(inode);
-		unlock_kernel();
-		write_inode_now(inode, 0);
-		invalidate_inode_buffers(inode);
+	struct udf_inode_info *iinfo = UDF_I(inode);
+
+	if (iinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB &&
+	    inode->i_size != iinfo->i_lenExtents) {
+		printk(KERN_WARNING "UDF-fs (%s): Inode %lu (mode %o) has "
+			"inode size %llu different from extent lenght %llu. "
+			"Filesystem need not be standards compliant.\n",
+			inode->i_sb->s_id, inode->i_ino, inode->i_mode,
+			(unsigned long long)inode->i_size,
+			(unsigned long long)iinfo->i_lenExtents);
 	}
-	iinfo = UDF_I(inode);
 	kfree(iinfo->i_ext.i_data);
 	iinfo->i_ext.i_data = NULL;
 }
@@ -198,7 +200,6 @@ struct buffer_head *udf_expand_dir_adinicb(struct inode *inode, int *block,
 	int newblock;
 	struct buffer_head *dbh = NULL;
 	struct kernel_lb_addr eloc;
-	uint32_t elen;
 	uint8_t alloctype;
 	struct extent_position epos;
 
@@ -273,12 +274,11 @@ struct buffer_head *udf_expand_dir_adinicb(struct inode *inode, int *block,
 	eloc.logicalBlockNum = *block;
 	eloc.partitionReferenceNum =
 				iinfo->i_location.partitionReferenceNum;
-	elen = inode->i_sb->s_blocksize;
-	iinfo->i_lenExtents = elen;
+	iinfo->i_lenExtents = inode->i_size;
 	epos.bh = NULL;
 	epos.block = iinfo->i_location;
 	epos.offset = udf_file_entry_alloc_offset(inode);
-	udf_add_aext(inode, &epos, &eloc, elen, 0);
+	udf_add_aext(inode, &epos, &eloc, inode->i_size, 0);
 	/* UniqueID stuff */
 
 	brelse(epos.bh);
diff --git a/fs/udf/namei.c b/fs/udf/namei.c
index 21dad8c608f9..cd2115060fdc 100644
--- a/fs/udf/namei.c
+++ b/fs/udf/namei.c
@@ -408,15 +408,6 @@ static struct fileIdentDesc *udf_add_entry(struct inode *dir,
 	}
 
 add:
-	/* Is there any extent whose size we need to round up? */
-	if (dinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB && elen) {
-		elen = (elen + sb->s_blocksize - 1) & ~(sb->s_blocksize - 1);
-		if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT)
-			epos.offset -= sizeof(struct short_ad);
-		else if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG)
-			epos.offset -= sizeof(struct long_ad);
-		udf_write_aext(dir, &epos, &eloc, elen, 1);
-	}
 	f_pos += nfidlen;
 
 	if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB &&
@@ -439,6 +430,7 @@ add:
 		udf_current_aext(dir, &epos, &eloc, &elen, 1);
 	}
 
+	/* Entry fits into current block? */
 	if (sb->s_blocksize - fibh->eoffset >= nfidlen) {
 		fibh->soffset = fibh->eoffset;
 		fibh->eoffset += nfidlen;
@@ -462,6 +454,16 @@ add:
 				(fibh->sbh->b_data + fibh->soffset);
 		}
 	} else {
+		/* Round up last extent in the file */
+		elen = (elen + sb->s_blocksize - 1) & ~(sb->s_blocksize - 1);
+		if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT)
+			epos.offset -= sizeof(struct short_ad);
+		else if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG)
+			epos.offset -= sizeof(struct long_ad);
+		udf_write_aext(dir, &epos, &eloc, elen, 1);
+		dinfo->i_lenExtents = (dinfo->i_lenExtents + sb->s_blocksize
+					- 1) & ~(sb->s_blocksize - 1);
+
 		fibh->soffset = fibh->eoffset - sb->s_blocksize;
 		fibh->eoffset += nfidlen - sb->s_blocksize;
 		if (fibh->sbh != fibh->ebh) {
@@ -508,6 +510,20 @@ add:
 		dir->i_size += nfidlen;
 		if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)
 			dinfo->i_lenAlloc += nfidlen;
+		else {
+			/* Find the last extent and truncate it to proper size */
+			while (udf_next_aext(dir, &epos, &eloc, &elen, 1) ==
+				(EXT_RECORDED_ALLOCATED >> 30))
+				;
+			elen -= dinfo->i_lenExtents - dir->i_size;
+			if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_SHORT)
+				epos.offset -= sizeof(struct short_ad);
+			else if (dinfo->i_alloc_type == ICBTAG_FLAG_AD_LONG)
+				epos.offset -= sizeof(struct long_ad);
+			udf_write_aext(dir, &epos, &eloc, elen, 1);
+			dinfo->i_lenExtents = dir->i_size;
+		}
+
 		mark_inode_dirty(dir);
 		goto out_ok;
 	} else {
@@ -922,7 +938,7 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry,
 		block = udf_get_pblock(inode->i_sb, block,
 				iinfo->i_location.partitionReferenceNum,
 				0);
-		epos.bh = udf_tread(inode->i_sb, block);
+		epos.bh = udf_tgetblk(inode->i_sb, block);
 		lock_buffer(epos.bh);
 		memset(epos.bh->b_data, 0x00, inode->i_sb->s_blocksize);
 		set_buffer_uptodate(epos.bh);
@@ -999,6 +1015,8 @@ static int udf_symlink(struct inode *dir, struct dentry *dentry,
 	inode->i_size = elen;
 	if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB)
 		iinfo->i_lenAlloc = inode->i_size;
+	else
+		udf_truncate_tail_extent(inode);
 	mark_inode_dirty(inode);
 
 	fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err);
diff --git a/fs/udf/super.c b/fs/udf/super.c
index 9d1b8c2e6c45..1e4543cbcd27 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -1078,21 +1078,39 @@ static int udf_fill_partdesc_info(struct super_block *sb,
 	return 0;
 }
 
-static int udf_load_vat(struct super_block *sb, int p_index, int type1_index)
+static void udf_find_vat_block(struct super_block *sb, int p_index,
+			       int type1_index, sector_t start_block)
 {
 	struct udf_sb_info *sbi = UDF_SB(sb);
 	struct udf_part_map *map = &sbi->s_partmaps[p_index];
+	sector_t vat_block;
 	struct kernel_lb_addr ino;
+
+	/*
+	 * VAT file entry is in the last recorded block. Some broken disks have
+	 * it a few blocks before so try a bit harder...
+	 */
+	ino.partitionReferenceNum = type1_index;
+	for (vat_block = start_block;
+	     vat_block >= map->s_partition_root &&
+	     vat_block >= start_block - 3 &&
+	     !sbi->s_vat_inode; vat_block--) {
+		ino.logicalBlockNum = vat_block - map->s_partition_root;
+		sbi->s_vat_inode = udf_iget(sb, &ino);
+	}
+}
+
+static int udf_load_vat(struct super_block *sb, int p_index, int type1_index)
+{
+	struct udf_sb_info *sbi = UDF_SB(sb);
+	struct udf_part_map *map = &sbi->s_partmaps[p_index];
 	struct buffer_head *bh = NULL;
 	struct udf_inode_info *vati;
 	uint32_t pos;
 	struct virtualAllocationTable20 *vat20;
 	sector_t blocks = sb->s_bdev->bd_inode->i_size >> sb->s_blocksize_bits;
 
-	/* VAT file entry is in the last recorded block */
-	ino.partitionReferenceNum = type1_index;
-	ino.logicalBlockNum = sbi->s_last_block - map->s_partition_root;
-	sbi->s_vat_inode = udf_iget(sb, &ino);
+	udf_find_vat_block(sb, p_index, type1_index, sbi->s_last_block);
 	if (!sbi->s_vat_inode &&
 	    sbi->s_last_block != blocks - 1) {
 		printk(KERN_NOTICE "UDF-fs: Failed to read VAT inode from the"
@@ -1100,9 +1118,7 @@ static int udf_load_vat(struct super_block *sb, int p_index, int type1_index)
 		       "block of the device (%lu).\n",
 		       (unsigned long)sbi->s_last_block,
 		       (unsigned long)blocks - 1);
-		ino.partitionReferenceNum = type1_index;
-		ino.logicalBlockNum = blocks - 1 - map->s_partition_root;
-		sbi->s_vat_inode = udf_iget(sb, &ino);
+		udf_find_vat_block(sb, p_index, type1_index, blocks - 1);
 	}
 	if (!sbi->s_vat_inode)
 		return 1;
diff --git a/fs/ufs/dir.c b/fs/ufs/dir.c
index 6f671f1ac271..22af68f8b682 100644
--- a/fs/ufs/dir.c
+++ b/fs/ufs/dir.c
@@ -70,13 +70,13 @@ static inline unsigned long ufs_dir_pages(struct inode *inode)
 	return (inode->i_size+PAGE_CACHE_SIZE-1)>>PAGE_CACHE_SHIFT;
 }
 
-ino_t ufs_inode_by_name(struct inode *dir, struct dentry *dentry)
+ino_t ufs_inode_by_name(struct inode *dir, struct qstr *qstr)
 {
 	ino_t res = 0;
 	struct ufs_dir_entry *de;
 	struct page *page;
 	
-	de = ufs_find_entry(dir, dentry, &page);
+	de = ufs_find_entry(dir, qstr, &page);
 	if (de) {
 		res = fs32_to_cpu(dir->i_sb, de->d_ino);
 		ufs_put_page(page);
@@ -249,12 +249,12 @@ struct ufs_dir_entry *ufs_dotdot(struct inode *dir, struct page **p)
  * (as a parameter - res_dir). Page is returned mapped and unlocked.
  * Entry is guaranteed to be valid.
  */
-struct ufs_dir_entry *ufs_find_entry(struct inode *dir, struct dentry *dentry,
+struct ufs_dir_entry *ufs_find_entry(struct inode *dir, struct qstr *qstr,
 				     struct page **res_page)
 {
 	struct super_block *sb = dir->i_sb;
-	const char *name = dentry->d_name.name;
-	int namelen = dentry->d_name.len;
+	const char *name = qstr->name;
+	int namelen = qstr->len;
 	unsigned reclen = UFS_DIR_REC_LEN(namelen);
 	unsigned long start, n;
 	unsigned long npages = ufs_dir_pages(dir);
diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
index 23119fe7ad62..4c26d9e8bc94 100644
--- a/fs/ufs/namei.c
+++ b/fs/ufs/namei.c
@@ -56,7 +56,7 @@ static struct dentry *ufs_lookup(struct inode * dir, struct dentry *dentry, stru
 		return ERR_PTR(-ENAMETOOLONG);
 
 	lock_kernel();
-	ino = ufs_inode_by_name(dir, dentry);
+	ino = ufs_inode_by_name(dir, &dentry->d_name);
 	if (ino) {
 		inode = ufs_iget(dir->i_sb, ino);
 		if (IS_ERR(inode)) {
@@ -237,7 +237,7 @@ static int ufs_unlink(struct inode *dir, struct dentry *dentry)
 	struct page *page;
 	int err = -ENOENT;
 
-	de = ufs_find_entry(dir, dentry, &page);
+	de = ufs_find_entry(dir, &dentry->d_name, &page);
 	if (!de)
 		goto out;
 
@@ -281,7 +281,7 @@ static int ufs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct ufs_dir_entry *old_de;
 	int err = -ENOENT;
 
-	old_de = ufs_find_entry(old_dir, old_dentry, &old_page);
+	old_de = ufs_find_entry(old_dir, &old_dentry->d_name, &old_page);
 	if (!old_de)
 		goto out;
 
@@ -301,7 +301,7 @@ static int ufs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			goto out_dir;
 
 		err = -ENOENT;
-		new_de = ufs_find_entry(new_dir, new_dentry, &new_page);
+		new_de = ufs_find_entry(new_dir, &new_dentry->d_name, &new_page);
 		if (!new_de)
 			goto out_dir;
 		inode_inc_link_count(old_inode);
diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index 5faed7954d0a..143c20bfb04b 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -66,6 +66,7 @@
  */
 
 
+#include <linux/exportfs.h>
 #include <linux/module.h>
 #include <linux/bitops.h>
 
@@ -96,6 +97,56 @@
 #include "swab.h"
 #include "util.h"
 
+static struct inode *ufs_nfs_get_inode(struct super_block *sb, u64 ino, u32 generation)
+{
+	struct ufs_sb_private_info *uspi = UFS_SB(sb)->s_uspi;
+	struct inode *inode;
+
+	if (ino < UFS_ROOTINO || ino > uspi->s_ncg * uspi->s_ipg)
+		return ERR_PTR(-ESTALE);
+
+	inode = ufs_iget(sb, ino);
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
+	if (generation && inode->i_generation != generation) {
+		iput(inode);
+		return ERR_PTR(-ESTALE);
+	}
+	return inode;
+}
+
+static struct dentry *ufs_fh_to_dentry(struct super_block *sb, struct fid *fid,
+				       int fh_len, int fh_type)
+{
+	return generic_fh_to_dentry(sb, fid, fh_len, fh_type, ufs_nfs_get_inode);
+}
+
+static struct dentry *ufs_fh_to_parent(struct super_block *sb, struct fid *fid,
+				       int fh_len, int fh_type)
+{
+	return generic_fh_to_parent(sb, fid, fh_len, fh_type, ufs_nfs_get_inode);
+}
+
+static struct dentry *ufs_get_parent(struct dentry *child)
+{
+	struct qstr dot_dot = {
+		.name	= "..",
+		.len	= 2,
+	};
+	ino_t ino;
+
+	ino = ufs_inode_by_name(child->d_inode, &dot_dot);
+	if (!ino)
+		return ERR_PTR(-ENOENT);
+	return d_obtain_alias(ufs_iget(child->d_inode->i_sb, ino));
+}
+
+static const struct export_operations ufs_export_ops = {
+	.fh_to_dentry	= ufs_fh_to_dentry,
+	.fh_to_parent	= ufs_fh_to_parent,
+	.get_parent	= ufs_get_parent,
+};
+
 #ifdef CONFIG_UFS_DEBUG
 /*
  * Print contents of ufs_super_block, useful for debugging
@@ -990,6 +1041,7 @@ magic_found:
 	 * Read ufs_super_block into internal data structures
 	 */
 	sb->s_op = &ufs_super_ops;
+	sb->s_export_op = &ufs_export_ops;
 	sb->dq_op = NULL; /***/
 	sb->s_magic = fs32_to_cpu(sb, usb3->fs_magic);
 
diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h
index 644e77e13599..0b4c39bc0d9e 100644
--- a/fs/ufs/ufs.h
+++ b/fs/ufs/ufs.h
@@ -86,9 +86,9 @@ extern void ufs_put_cylinder (struct super_block *, unsigned);
 /* dir.c */
 extern const struct inode_operations ufs_dir_inode_operations;
 extern int ufs_add_link (struct dentry *, struct inode *);
-extern ino_t ufs_inode_by_name(struct inode *, struct dentry *);
+extern ino_t ufs_inode_by_name(struct inode *, struct qstr *);
 extern int ufs_make_empty(struct inode *, struct inode *);
-extern struct ufs_dir_entry *ufs_find_entry(struct inode *, struct dentry *, struct page **);
+extern struct ufs_dir_entry *ufs_find_entry(struct inode *, struct qstr *, struct page **);
 extern int ufs_delete_entry(struct inode *, struct ufs_dir_entry *, struct page *);
 extern int ufs_empty_dir (struct inode *);
 extern struct ufs_dir_entry *ufs_dotdot(struct inode *, struct page **);
diff --git a/fs/xattr.c b/fs/xattr.c
index 6d4f6d3449fb..46f87e828b48 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -615,12 +615,11 @@ ssize_t
 generic_getxattr(struct dentry *dentry, const char *name, void *buffer, size_t size)
 {
 	struct xattr_handler *handler;
-	struct inode *inode = dentry->d_inode;
 
-	handler = xattr_resolve_name(inode->i_sb->s_xattr, &name);
+	handler = xattr_resolve_name(dentry->d_sb->s_xattr, &name);
 	if (!handler)
 		return -EOPNOTSUPP;
-	return handler->get(inode, name, buffer, size);
+	return handler->get(dentry, name, buffer, size, handler->flags);
 }
 
 /*
@@ -630,18 +629,20 @@ generic_getxattr(struct dentry *dentry, const char *name, void *buffer, size_t s
 ssize_t
 generic_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
-	struct inode *inode = dentry->d_inode;
-	struct xattr_handler *handler, **handlers = inode->i_sb->s_xattr;
+	struct xattr_handler *handler, **handlers = dentry->d_sb->s_xattr;
 	unsigned int size = 0;
 
 	if (!buffer) {
-		for_each_xattr_handler(handlers, handler)
-			size += handler->list(inode, NULL, 0, NULL, 0);
+		for_each_xattr_handler(handlers, handler) {
+			size += handler->list(dentry, NULL, 0, NULL, 0,
+					      handler->flags);
+		}
 	} else {
 		char *buf = buffer;
 
 		for_each_xattr_handler(handlers, handler) {
-			size = handler->list(inode, buf, buffer_size, NULL, 0);
+			size = handler->list(dentry, buf, buffer_size,
+					     NULL, 0, handler->flags);
 			if (size > buffer_size)
 				return -ERANGE;
 			buf += size;
@@ -659,14 +660,13 @@ int
 generic_setxattr(struct dentry *dentry, const char *name, const void *value, size_t size, int flags)
 {
 	struct xattr_handler *handler;
-	struct inode *inode = dentry->d_inode;
 
 	if (size == 0)
 		value = "";  /* empty EA, do not remove */
-	handler = xattr_resolve_name(inode->i_sb->s_xattr, &name);
+	handler = xattr_resolve_name(dentry->d_sb->s_xattr, &name);
 	if (!handler)
 		return -EOPNOTSUPP;
-	return handler->set(inode, name, value, size, flags);
+	return handler->set(dentry, name, value, size, 0, handler->flags);
 }
 
 /*
@@ -677,12 +677,12 @@ int
 generic_removexattr(struct dentry *dentry, const char *name)
 {
 	struct xattr_handler *handler;
-	struct inode *inode = dentry->d_inode;
 
-	handler = xattr_resolve_name(inode->i_sb->s_xattr, &name);
+	handler = xattr_resolve_name(dentry->d_sb->s_xattr, &name);
 	if (!handler)
 		return -EOPNOTSUPP;
-	return handler->set(inode, name, NULL, 0, XATTR_REPLACE);
+	return handler->set(dentry, name, NULL, 0,
+			    XATTR_REPLACE, handler->flags);
 }
 
 EXPORT_SYMBOL(generic_getxattr);
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 7a59daed1782..56641fe52a23 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -26,6 +26,8 @@ endif
 
 obj-$(CONFIG_XFS_FS)		+= xfs.o
 
+xfs-y				+= linux-2.6/xfs_trace.o
+
 xfs-$(CONFIG_XFS_QUOTA)		+= $(addprefix quota/, \
 				   xfs_dquot.o \
 				   xfs_dquot_item.o \
@@ -90,8 +92,7 @@ xfs-y				+= xfs_alloc.o \
 				   xfs_rw.o \
 				   xfs_dmops.o
 
-xfs-$(CONFIG_XFS_TRACE)		+= xfs_btree_trace.o \
-				   xfs_dir2_trace.o
+xfs-$(CONFIG_XFS_TRACE)		+= xfs_btree_trace.o
 
 # Objects in linux/
 xfs-y				+= $(addprefix $(XFS_LINUX)/, \
@@ -113,6 +114,3 @@ xfs-y				+= $(addprefix $(XFS_LINUX)/, \
 xfs-y				+= $(addprefix support/, \
 				   debug.o \
 				   uuid.o)
-
-xfs-$(CONFIG_XFS_TRACE)		+= support/ktrace.o
-
diff --git a/fs/xfs/linux-2.6/xfs_acl.c b/fs/xfs/linux-2.6/xfs_acl.c
index b23a54506446..883ca5ab8af5 100644
--- a/fs/xfs/linux-2.6/xfs_acl.c
+++ b/fs/xfs/linux-2.6/xfs_acl.c
@@ -21,6 +21,7 @@
 #include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 #include <linux/xattr.h>
 #include <linux/posix_acl_xattr.h>
 
@@ -250,8 +251,9 @@ xfs_set_mode(struct inode *inode, mode_t mode)
 	if (mode != inode->i_mode) {
 		struct iattr iattr;
 
-		iattr.ia_valid = ATTR_MODE;
+		iattr.ia_valid = ATTR_MODE | ATTR_CTIME;
 		iattr.ia_mode = mode;
+		iattr.ia_ctime = current_fs_time(inode->i_sb);
 
 		error = -xfs_setattr(XFS_I(inode), &iattr, XFS_ATTR_NOACL);
 	}
@@ -353,37 +355,14 @@ xfs_acl_chmod(struct inode *inode)
 	return error;
 }
 
-/*
- * System xattr handlers.
- *
- * Currently Posix ACLs are the only system namespace extended attribute
- * handlers supported by XFS, so we just implement the handlers here.
- * If we ever support other system extended attributes this will need
- * some refactoring.
- */
-
 static int
-xfs_decode_acl(const char *name)
-{
-	if (strcmp(name, "posix_acl_access") == 0)
-		return ACL_TYPE_ACCESS;
-	else if (strcmp(name, "posix_acl_default") == 0)
-		return ACL_TYPE_DEFAULT;
-	return -EINVAL;
-}
-
-static int
-xfs_xattr_system_get(struct inode *inode, const char *name,
-		void *value, size_t size)
+xfs_xattr_acl_get(struct dentry *dentry, const char *name,
+		void *value, size_t size, int type)
 {
 	struct posix_acl *acl;
-	int type, error;
-
-	type = xfs_decode_acl(name);
-	if (type < 0)
-		return type;
+	int error;
 
-	acl = xfs_get_acl(inode, type);
+	acl = xfs_get_acl(dentry->d_inode, type);
 	if (IS_ERR(acl))
 		return PTR_ERR(acl);
 	if (acl == NULL)
@@ -396,15 +375,13 @@ xfs_xattr_system_get(struct inode *inode, const char *name,
 }
 
 static int
-xfs_xattr_system_set(struct inode *inode, const char *name,
-		const void *value, size_t size, int flags)
+xfs_xattr_acl_set(struct dentry *dentry, const char *name,
+		const void *value, size_t size, int flags, int type)
 {
+	struct inode *inode = dentry->d_inode;
 	struct posix_acl *acl = NULL;
-	int error = 0, type;
+	int error = 0;
 
-	type = xfs_decode_acl(name);
-	if (type < 0)
-		return type;
 	if (flags & XATTR_CREATE)
 		return -EINVAL;
 	if (type == ACL_TYPE_DEFAULT && !S_ISDIR(inode->i_mode))
@@ -461,8 +438,16 @@ xfs_xattr_system_set(struct inode *inode, const char *name,
 	return error;
 }
 
-struct xattr_handler xfs_xattr_system_handler = {
-	.prefix	= XATTR_SYSTEM_PREFIX,
-	.get	= xfs_xattr_system_get,
-	.set	= xfs_xattr_system_set,
+struct xattr_handler xfs_xattr_acl_access_handler = {
+	.prefix	= POSIX_ACL_XATTR_ACCESS,
+	.flags	= ACL_TYPE_ACCESS,
+	.get	= xfs_xattr_acl_get,
+	.set	= xfs_xattr_acl_set,
+};
+
+struct xattr_handler xfs_xattr_acl_default_handler = {
+	.prefix	= POSIX_ACL_XATTR_DEFAULT,
+	.flags	= ACL_TYPE_DEFAULT,
+	.get	= xfs_xattr_acl_get,
+	.set	= xfs_xattr_acl_set,
 };
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 70f989895d15..66abe36c1213 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -38,6 +38,7 @@
 #include "xfs_rw.h"
 #include "xfs_iomap.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 #include <linux/mpage.h>
 #include <linux/pagevec.h>
 #include <linux/writeback.h>
@@ -76,7 +77,7 @@ xfs_ioend_wake(
 		wake_up(to_ioend_wq(ip));
 }
 
-STATIC void
+void
 xfs_count_page_state(
 	struct page		*page,
 	int			*delalloc,
@@ -98,48 +99,6 @@ xfs_count_page_state(
 	} while ((bh = bh->b_this_page) != head);
 }
 
-#if defined(XFS_RW_TRACE)
-void
-xfs_page_trace(
-	int		tag,
-	struct inode	*inode,
-	struct page	*page,
-	unsigned long	pgoff)
-{
-	xfs_inode_t	*ip;
-	loff_t		isize = i_size_read(inode);
-	loff_t		offset = page_offset(page);
-	int		delalloc = -1, unmapped = -1, unwritten = -1;
-
-	if (page_has_buffers(page))
-		xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
-
-	ip = XFS_I(inode);
-	if (!ip->i_rwtrace)
-		return;
-
-	ktrace_enter(ip->i_rwtrace,
-		(void *)((unsigned long)tag),
-		(void *)ip,
-		(void *)inode,
-		(void *)page,
-		(void *)pgoff,
-		(void *)((unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(ip->i_d.di_size & 0xffffffff)),
-		(void *)((unsigned long)((isize >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(isize & 0xffffffff)),
-		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(offset & 0xffffffff)),
-		(void *)((unsigned long)delalloc),
-		(void *)((unsigned long)unmapped),
-		(void *)((unsigned long)unwritten),
-		(void *)((unsigned long)current_pid()),
-		(void *)NULL);
-}
-#else
-#define xfs_page_trace(tag, inode, page, pgoff)
-#endif
-
 STATIC struct block_device *
 xfs_find_bdev_for_inode(
 	struct xfs_inode	*ip)
@@ -235,71 +194,36 @@ xfs_setfilesize(
 }
 
 /*
- * Buffered IO write completion for delayed allocate extents.
- */
-STATIC void
-xfs_end_bio_delalloc(
-	struct work_struct	*work)
-{
-	xfs_ioend_t		*ioend =
-		container_of(work, xfs_ioend_t, io_work);
-
-	xfs_setfilesize(ioend);
-	xfs_destroy_ioend(ioend);
-}
-
-/*
- * Buffered IO write completion for regular, written extents.
+ * IO write completion.
  */
 STATIC void
-xfs_end_bio_written(
-	struct work_struct	*work)
-{
-	xfs_ioend_t		*ioend =
-		container_of(work, xfs_ioend_t, io_work);
-
-	xfs_setfilesize(ioend);
-	xfs_destroy_ioend(ioend);
-}
-
-/*
- * IO write completion for unwritten extents.
- *
- * Issue transactions to convert a buffer range from unwritten
- * to written extents.
- */
-STATIC void
-xfs_end_bio_unwritten(
+xfs_end_io(
 	struct work_struct	*work)
 {
 	xfs_ioend_t		*ioend =
 		container_of(work, xfs_ioend_t, io_work);
 	struct xfs_inode	*ip = XFS_I(ioend->io_inode);
-	xfs_off_t		offset = ioend->io_offset;
-	size_t			size = ioend->io_size;
-
-	if (likely(!ioend->io_error)) {
-		if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-			int error;
-			error = xfs_iomap_write_unwritten(ip, offset, size);
-			if (error)
-				ioend->io_error = error;
-		}
-		xfs_setfilesize(ioend);
-	}
-	xfs_destroy_ioend(ioend);
-}
 
-/*
- * IO read completion for regular, written extents.
- */
-STATIC void
-xfs_end_bio_read(
-	struct work_struct	*work)
-{
-	xfs_ioend_t		*ioend =
-		container_of(work, xfs_ioend_t, io_work);
+	/*
+	 * For unwritten extents we need to issue transactions to convert a
+	 * range to normal written extens after the data I/O has finished.
+	 */
+	if (ioend->io_type == IOMAP_UNWRITTEN &&
+	    likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) {
+		int error;
+
+		error = xfs_iomap_write_unwritten(ip, ioend->io_offset,
+						 ioend->io_size);
+		if (error)
+			ioend->io_error = error;
+	}
 
+	/*
+	 * We might have to update the on-disk file size after extending
+	 * writes.
+	 */
+	if (ioend->io_type != IOMAP_READ)
+		xfs_setfilesize(ioend);
 	xfs_destroy_ioend(ioend);
 }
 
@@ -314,10 +238,10 @@ xfs_finish_ioend(
 	int		wait)
 {
 	if (atomic_dec_and_test(&ioend->io_remaining)) {
-		struct workqueue_struct *wq = xfsdatad_workqueue;
-		if (ioend->io_work.func == xfs_end_bio_unwritten)
-			wq = xfsconvertd_workqueue;
+		struct workqueue_struct *wq;
 
+		wq = (ioend->io_type == IOMAP_UNWRITTEN) ?
+			xfsconvertd_workqueue : xfsdatad_workqueue;
 		queue_work(wq, &ioend->io_work);
 		if (wait)
 			flush_workqueue(wq);
@@ -355,15 +279,7 @@ xfs_alloc_ioend(
 	ioend->io_offset = 0;
 	ioend->io_size = 0;
 
-	if (type == IOMAP_UNWRITTEN)
-		INIT_WORK(&ioend->io_work, xfs_end_bio_unwritten);
-	else if (type == IOMAP_DELAY)
-		INIT_WORK(&ioend->io_work, xfs_end_bio_delalloc);
-	else if (type == IOMAP_READ)
-		INIT_WORK(&ioend->io_work, xfs_end_bio_read);
-	else
-		INIT_WORK(&ioend->io_work, xfs_end_bio_written);
-
+	INIT_WORK(&ioend->io_work, xfs_end_io);
 	return ioend;
 }
 
@@ -380,7 +296,7 @@ xfs_map_blocks(
 	return -xfs_iomap(XFS_I(inode), offset, count, flags, mapp, &nmaps);
 }
 
-STATIC_INLINE int
+STATIC int
 xfs_iomap_valid(
 	xfs_iomap_t		*iomapp,
 	loff_t			offset)
@@ -412,8 +328,9 @@ xfs_end_bio(
 
 STATIC void
 xfs_submit_ioend_bio(
-	xfs_ioend_t	*ioend,
-	struct bio	*bio)
+	struct writeback_control *wbc,
+	xfs_ioend_t		*ioend,
+	struct bio		*bio)
 {
 	atomic_inc(&ioend->io_remaining);
 	bio->bi_private = ioend;
@@ -426,7 +343,8 @@ xfs_submit_ioend_bio(
 	if (xfs_ioend_new_eof(ioend))
 		xfs_mark_inode_dirty_sync(XFS_I(ioend->io_inode));
 
-	submit_bio(WRITE, bio);
+	submit_bio(wbc->sync_mode == WB_SYNC_ALL ?
+		   WRITE_SYNC_PLUG : WRITE, bio);
 	ASSERT(!bio_flagged(bio, BIO_EOPNOTSUPP));
 	bio_put(bio);
 }
@@ -505,6 +423,7 @@ static inline int bio_add_buffer(struct bio *bio, struct buffer_head *bh)
  */
 STATIC void
 xfs_submit_ioend(
+	struct writeback_control *wbc,
 	xfs_ioend_t		*ioend)
 {
 	xfs_ioend_t		*head = ioend;
@@ -533,19 +452,19 @@ xfs_submit_ioend(
  retry:
 				bio = xfs_alloc_ioend_bio(bh);
 			} else if (bh->b_blocknr != lastblock + 1) {
-				xfs_submit_ioend_bio(ioend, bio);
+				xfs_submit_ioend_bio(wbc, ioend, bio);
 				goto retry;
 			}
 
 			if (bio_add_buffer(bio, bh) != bh->b_size) {
-				xfs_submit_ioend_bio(ioend, bio);
+				xfs_submit_ioend_bio(wbc, ioend, bio);
 				goto retry;
 			}
 
 			lastblock = bh->b_blocknr;
 		}
 		if (bio)
-			xfs_submit_ioend_bio(ioend, bio);
+			xfs_submit_ioend_bio(wbc, ioend, bio);
 		xfs_finish_ioend(ioend, 0);
 	} while ((ioend = next) != NULL);
 }
@@ -1191,7 +1110,7 @@ xfs_page_state_convert(
 	}
 
 	if (iohead)
-		xfs_submit_ioend(iohead);
+		xfs_submit_ioend(wbc, iohead);
 
 	return page_dirty;
 
@@ -1242,7 +1161,7 @@ xfs_vm_writepage(
 	int			delalloc, unmapped, unwritten;
 	struct inode		*inode = page->mapping->host;
 
-	xfs_page_trace(XFS_WRITEPAGE_ENTER, inode, page, 0);
+	trace_xfs_writepage(inode, page, 0);
 
 	/*
 	 * We need a transaction if:
@@ -1347,7 +1266,7 @@ xfs_vm_releasepage(
 		.nr_to_write = 1,
 	};
 
-	xfs_page_trace(XFS_RELEASEPAGE_ENTER, inode, page, 0);
+	trace_xfs_releasepage(inode, page, 0);
 
 	if (!page_has_buffers(page))
 		return 0;
@@ -1528,7 +1447,7 @@ xfs_end_io_direct(
 		 * didn't map an unwritten extent so switch it's completion
 		 * handler.
 		 */
-		INIT_WORK(&ioend->io_work, xfs_end_bio_written);
+		ioend->io_type = IOMAP_NEW;
 		xfs_finish_ioend(ioend, 0);
 	}
 
@@ -1555,19 +1474,13 @@ xfs_vm_direct_IO(
 
 	bdev = xfs_find_bdev_for_inode(XFS_I(inode));
 
-	if (rw == WRITE) {
-		iocb->private = xfs_alloc_ioend(inode, IOMAP_UNWRITTEN);
-		ret = blockdev_direct_IO_own_locking(rw, iocb, inode,
-			bdev, iov, offset, nr_segs,
-			xfs_get_blocks_direct,
-			xfs_end_io_direct);
-	} else {
-		iocb->private = xfs_alloc_ioend(inode, IOMAP_READ);
-		ret = blockdev_direct_IO_no_locking(rw, iocb, inode,
-			bdev, iov, offset, nr_segs,
-			xfs_get_blocks_direct,
-			xfs_end_io_direct);
-	}
+	iocb->private = xfs_alloc_ioend(inode, rw == WRITE ?
+					IOMAP_UNWRITTEN : IOMAP_READ);
+
+	ret = blockdev_direct_IO_no_locking(rw, iocb, inode, bdev, iov,
+					    offset, nr_segs,
+					    xfs_get_blocks_direct,
+					    xfs_end_io_direct);
 
 	if (unlikely(ret != -EIOCBQUEUED && iocb->private))
 		xfs_destroy_ioend(iocb->private);
@@ -1627,8 +1540,7 @@ xfs_vm_invalidatepage(
 	struct page		*page,
 	unsigned long		offset)
 {
-	xfs_page_trace(XFS_INVALIDPAGE_ENTER,
-			page->mapping->host, page, offset);
+	trace_xfs_invalidatepage(page->mapping->host, page, offset);
 	block_invalidatepage(page, offset);
 }
 
diff --git a/fs/xfs/linux-2.6/xfs_aops.h b/fs/xfs/linux-2.6/xfs_aops.h
index 221b3e66ceef..4cfc6ea87df8 100644
--- a/fs/xfs/linux-2.6/xfs_aops.h
+++ b/fs/xfs/linux-2.6/xfs_aops.h
@@ -45,4 +45,6 @@ extern int xfs_get_blocks(struct inode *, sector_t, struct buffer_head *, int);
 extern void xfs_ioend_init(void);
 extern void xfs_ioend_wait(struct xfs_inode *);
 
+extern void xfs_count_page_state(struct page *, int *, int *, int *);
+
 #endif /* __XFS_AOPS_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 965df1227d64..77b8be81c769 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -39,6 +39,7 @@
 #include "xfs_ag.h"
 #include "xfs_dmapi.h"
 #include "xfs_mount.h"
+#include "xfs_trace.h"
 
 static kmem_zone_t *xfs_buf_zone;
 STATIC int xfsbufd(void *);
@@ -53,34 +54,6 @@ static struct workqueue_struct *xfslogd_workqueue;
 struct workqueue_struct *xfsdatad_workqueue;
 struct workqueue_struct *xfsconvertd_workqueue;
 
-#ifdef XFS_BUF_TRACE
-void
-xfs_buf_trace(
-	xfs_buf_t	*bp,
-	char		*id,
-	void		*data,
-	void		*ra)
-{
-	ktrace_enter(xfs_buf_trace_buf,
-		bp, id,
-		(void *)(unsigned long)bp->b_flags,
-		(void *)(unsigned long)bp->b_hold.counter,
-		(void *)(unsigned long)bp->b_sema.count,
-		(void *)current,
-		data, ra,
-		(void *)(unsigned long)((bp->b_file_offset>>32) & 0xffffffff),
-		(void *)(unsigned long)(bp->b_file_offset & 0xffffffff),
-		(void *)(unsigned long)bp->b_buffer_length,
-		NULL, NULL, NULL, NULL, NULL);
-}
-ktrace_t *xfs_buf_trace_buf;
-#define XFS_BUF_TRACE_SIZE	4096
-#define XB_TRACE(bp, id, data)	\
-	xfs_buf_trace(bp, id, (void *)data, (void *)__builtin_return_address(0))
-#else
-#define XB_TRACE(bp, id, data)	do { } while (0)
-#endif
-
 #ifdef XFS_BUF_LOCK_TRACKING
 # define XB_SET_OWNER(bp)	((bp)->b_last_holder = current->pid)
 # define XB_CLEAR_OWNER(bp)	((bp)->b_last_holder = -1)
@@ -149,7 +122,7 @@ page_region_mask(
 	return mask;
 }
 
-STATIC_INLINE void
+STATIC void
 set_page_region(
 	struct page	*page,
 	size_t		offset,
@@ -161,7 +134,7 @@ set_page_region(
 		SetPageUptodate(page);
 }
 
-STATIC_INLINE int
+STATIC int
 test_page_region(
 	struct page	*page,
 	size_t		offset,
@@ -279,7 +252,8 @@ _xfs_buf_initialize(
 	init_waitqueue_head(&bp->b_waiters);
 
 	XFS_STATS_INC(xb_create);
-	XB_TRACE(bp, "initialize", target);
+
+	trace_xfs_buf_init(bp, _RET_IP_);
 }
 
 /*
@@ -318,6 +292,7 @@ _xfs_buf_free_pages(
 {
 	if (bp->b_pages != bp->b_page_array) {
 		kmem_free(bp->b_pages);
+		bp->b_pages = NULL;
 	}
 }
 
@@ -332,7 +307,7 @@ void
 xfs_buf_free(
 	xfs_buf_t		*bp)
 {
-	XB_TRACE(bp, "free", 0);
+	trace_xfs_buf_free(bp, _RET_IP_);
 
 	ASSERT(list_empty(&bp->b_hash_list));
 
@@ -349,9 +324,8 @@ xfs_buf_free(
 				ASSERT(!PagePrivate(page));
 			page_cache_release(page);
 		}
-		_xfs_buf_free_pages(bp);
 	}
-
+	_xfs_buf_free_pages(bp);
 	xfs_buf_deallocate(bp);
 }
 
@@ -445,7 +419,6 @@ _xfs_buf_lookup_pages(
 	if (page_count == bp->b_page_count)
 		bp->b_flags |= XBF_DONE;
 
-	XB_TRACE(bp, "lookup_pages", (long)page_count);
 	return error;
 }
 
@@ -548,7 +521,6 @@ found:
 	if (down_trylock(&bp->b_sema)) {
 		if (!(flags & XBF_TRYLOCK)) {
 			/* wait for buffer ownership */
-			XB_TRACE(bp, "get_lock", 0);
 			xfs_buf_lock(bp);
 			XFS_STATS_INC(xb_get_locked_waited);
 		} else {
@@ -571,7 +543,8 @@ found:
 		ASSERT((bp->b_flags & _XBF_DELWRI_Q) == 0);
 		bp->b_flags &= XBF_MAPPED;
 	}
-	XB_TRACE(bp, "got_lock", 0);
+
+	trace_xfs_buf_find(bp, flags, _RET_IP_);
 	XFS_STATS_INC(xb_get_locked);
 	return bp;
 }
@@ -582,7 +555,7 @@ found:
  *	although backing storage may not be.
  */
 xfs_buf_t *
-xfs_buf_get_flags(
+xfs_buf_get(
 	xfs_buftarg_t		*target,/* target for buffer		*/
 	xfs_off_t		ioff,	/* starting offset of range	*/
 	size_t			isize,	/* length of range		*/
@@ -627,7 +600,7 @@ xfs_buf_get_flags(
 	bp->b_bn = ioff;
 	bp->b_count_desired = bp->b_buffer_length;
 
-	XB_TRACE(bp, "get", (unsigned long)flags);
+	trace_xfs_buf_get(bp, flags, _RET_IP_);
 	return bp;
 
  no_buffer:
@@ -644,8 +617,6 @@ _xfs_buf_read(
 {
 	int			status;
 
-	XB_TRACE(bp, "_xfs_buf_read", (unsigned long)flags);
-
 	ASSERT(!(flags & (XBF_DELWRI|XBF_WRITE)));
 	ASSERT(bp->b_bn != XFS_BUF_DADDR_NULL);
 
@@ -661,7 +632,7 @@ _xfs_buf_read(
 }
 
 xfs_buf_t *
-xfs_buf_read_flags(
+xfs_buf_read(
 	xfs_buftarg_t		*target,
 	xfs_off_t		ioff,
 	size_t			isize,
@@ -671,21 +642,20 @@ xfs_buf_read_flags(
 
 	flags |= XBF_READ;
 
-	bp = xfs_buf_get_flags(target, ioff, isize, flags);
+	bp = xfs_buf_get(target, ioff, isize, flags);
 	if (bp) {
+		trace_xfs_buf_read(bp, flags, _RET_IP_);
+
 		if (!XFS_BUF_ISDONE(bp)) {
-			XB_TRACE(bp, "read", (unsigned long)flags);
 			XFS_STATS_INC(xb_get_read);
 			_xfs_buf_read(bp, flags);
 		} else if (flags & XBF_ASYNC) {
-			XB_TRACE(bp, "read_async", (unsigned long)flags);
 			/*
 			 * Read ahead call which is already satisfied,
 			 * drop the buffer
 			 */
 			goto no_buffer;
 		} else {
-			XB_TRACE(bp, "read_done", (unsigned long)flags);
 			/* We do not want read in the flags */
 			bp->b_flags &= ~XBF_READ;
 		}
@@ -718,7 +688,7 @@ xfs_buf_readahead(
 		return;
 
 	flags |= (XBF_TRYLOCK|XBF_ASYNC|XBF_READ_AHEAD);
-	xfs_buf_read_flags(target, ioff, isize, flags);
+	xfs_buf_read(target, ioff, isize, flags);
 }
 
 xfs_buf_t *
@@ -823,7 +793,7 @@ xfs_buf_get_noaddr(
 
 	xfs_buf_unlock(bp);
 
-	XB_TRACE(bp, "no_daddr", len);
+	trace_xfs_buf_get_noaddr(bp, _RET_IP_);
 	return bp;
 
  fail_free_mem:
@@ -845,8 +815,8 @@ void
 xfs_buf_hold(
 	xfs_buf_t		*bp)
 {
+	trace_xfs_buf_hold(bp, _RET_IP_);
 	atomic_inc(&bp->b_hold);
-	XB_TRACE(bp, "hold", 0);
 }
 
 /*
@@ -859,7 +829,7 @@ xfs_buf_rele(
 {
 	xfs_bufhash_t		*hash = bp->b_hash;
 
-	XB_TRACE(bp, "rele", bp->b_relse);
+	trace_xfs_buf_rele(bp, _RET_IP_);
 
 	if (unlikely(!hash)) {
 		ASSERT(!bp->b_relse);
@@ -909,21 +879,19 @@ xfs_buf_cond_lock(
 	int			locked;
 
 	locked = down_trylock(&bp->b_sema) == 0;
-	if (locked) {
+	if (locked)
 		XB_SET_OWNER(bp);
-	}
-	XB_TRACE(bp, "cond_lock", (long)locked);
+
+	trace_xfs_buf_cond_lock(bp, _RET_IP_);
 	return locked ? 0 : -EBUSY;
 }
 
-#if defined(DEBUG) || defined(XFS_BLI_TRACE)
 int
 xfs_buf_lock_value(
 	xfs_buf_t		*bp)
 {
 	return bp->b_sema.count;
 }
-#endif
 
 /*
  *	Locks a buffer object.
@@ -935,12 +903,14 @@ void
 xfs_buf_lock(
 	xfs_buf_t		*bp)
 {
-	XB_TRACE(bp, "lock", 0);
+	trace_xfs_buf_lock(bp, _RET_IP_);
+
 	if (atomic_read(&bp->b_io_remaining))
 		blk_run_address_space(bp->b_target->bt_mapping);
 	down(&bp->b_sema);
 	XB_SET_OWNER(bp);
-	XB_TRACE(bp, "locked", 0);
+
+	trace_xfs_buf_lock_done(bp, _RET_IP_);
 }
 
 /*
@@ -962,7 +932,8 @@ xfs_buf_unlock(
 
 	XB_CLEAR_OWNER(bp);
 	up(&bp->b_sema);
-	XB_TRACE(bp, "unlock", 0);
+
+	trace_xfs_buf_unlock(bp, _RET_IP_);
 }
 
 
@@ -974,17 +945,18 @@ void
 xfs_buf_pin(
 	xfs_buf_t		*bp)
 {
+	trace_xfs_buf_pin(bp, _RET_IP_);
 	atomic_inc(&bp->b_pin_count);
-	XB_TRACE(bp, "pin", (long)bp->b_pin_count.counter);
 }
 
 void
 xfs_buf_unpin(
 	xfs_buf_t		*bp)
 {
+	trace_xfs_buf_unpin(bp, _RET_IP_);
+
 	if (atomic_dec_and_test(&bp->b_pin_count))
 		wake_up_all(&bp->b_waiters);
-	XB_TRACE(bp, "unpin", (long)bp->b_pin_count.counter);
 }
 
 int
@@ -1035,7 +1007,7 @@ xfs_buf_iodone_work(
 	 */
 	if ((bp->b_error == EOPNOTSUPP) &&
 	    (bp->b_flags & (XBF_ORDERED|XBF_ASYNC)) == (XBF_ORDERED|XBF_ASYNC)) {
-		XB_TRACE(bp, "ordered_retry", bp->b_iodone);
+		trace_xfs_buf_ordered_retry(bp, _RET_IP_);
 		bp->b_flags &= ~XBF_ORDERED;
 		bp->b_flags |= _XFS_BARRIER_FAILED;
 		xfs_buf_iorequest(bp);
@@ -1050,12 +1022,12 @@ xfs_buf_ioend(
 	xfs_buf_t		*bp,
 	int			schedule)
 {
+	trace_xfs_buf_iodone(bp, _RET_IP_);
+
 	bp->b_flags &= ~(XBF_READ | XBF_WRITE | XBF_READ_AHEAD);
 	if (bp->b_error == 0)
 		bp->b_flags |= XBF_DONE;
 
-	XB_TRACE(bp, "iodone", bp->b_iodone);
-
 	if ((bp->b_iodone) || (bp->b_flags & XBF_ASYNC)) {
 		if (schedule) {
 			INIT_WORK(&bp->b_iodone_work, xfs_buf_iodone_work);
@@ -1075,7 +1047,7 @@ xfs_buf_ioerror(
 {
 	ASSERT(error >= 0 && error <= 0xffff);
 	bp->b_error = (unsigned short)error;
-	XB_TRACE(bp, "ioerror", (unsigned long)error);
+	trace_xfs_buf_ioerror(bp, error, _RET_IP_);
 }
 
 int
@@ -1083,7 +1055,7 @@ xfs_bawrite(
 	void			*mp,
 	struct xfs_buf		*bp)
 {
-	XB_TRACE(bp, "bawrite", 0);
+	trace_xfs_buf_bawrite(bp, _RET_IP_);
 
 	ASSERT(bp->b_bn != XFS_BUF_DADDR_NULL);
 
@@ -1102,7 +1074,7 @@ xfs_bdwrite(
 	void			*mp,
 	struct xfs_buf		*bp)
 {
-	XB_TRACE(bp, "bdwrite", 0);
+	trace_xfs_buf_bdwrite(bp, _RET_IP_);
 
 	bp->b_strat = xfs_bdstrat_cb;
 	bp->b_mount = mp;
@@ -1113,7 +1085,7 @@ xfs_bdwrite(
 	xfs_buf_delwri_queue(bp, 1);
 }
 
-STATIC_INLINE void
+STATIC void
 _xfs_buf_ioend(
 	xfs_buf_t		*bp,
 	int			schedule)
@@ -1177,10 +1149,14 @@ _xfs_buf_ioapply(
 	if (bp->b_flags & XBF_ORDERED) {
 		ASSERT(!(bp->b_flags & XBF_READ));
 		rw = WRITE_BARRIER;
-	} else if (bp->b_flags & _XBF_RUN_QUEUES) {
+	} else if (bp->b_flags & XBF_LOG_BUFFER) {
 		ASSERT(!(bp->b_flags & XBF_READ_AHEAD));
 		bp->b_flags &= ~_XBF_RUN_QUEUES;
 		rw = (bp->b_flags & XBF_WRITE) ? WRITE_SYNC : READ_SYNC;
+	} else if (bp->b_flags & _XBF_RUN_QUEUES) {
+		ASSERT(!(bp->b_flags & XBF_READ_AHEAD));
+		bp->b_flags &= ~_XBF_RUN_QUEUES;
+		rw = (bp->b_flags & XBF_WRITE) ? WRITE_META : READ_META;
 	} else {
 		rw = (bp->b_flags & XBF_WRITE) ? WRITE :
 		     (bp->b_flags & XBF_READ_AHEAD) ? READA : READ;
@@ -1253,7 +1229,7 @@ int
 xfs_buf_iorequest(
 	xfs_buf_t		*bp)
 {
-	XB_TRACE(bp, "iorequest", 0);
+	trace_xfs_buf_iorequest(bp, _RET_IP_);
 
 	if (bp->b_flags & XBF_DELWRI) {
 		xfs_buf_delwri_queue(bp, 1);
@@ -1287,11 +1263,13 @@ int
 xfs_buf_iowait(
 	xfs_buf_t		*bp)
 {
-	XB_TRACE(bp, "iowait", 0);
+	trace_xfs_buf_iowait(bp, _RET_IP_);
+
 	if (atomic_read(&bp->b_io_remaining))
 		blk_run_address_space(bp->b_target->bt_mapping);
 	wait_for_completion(&bp->b_iowait);
-	XB_TRACE(bp, "iowaited", (long)bp->b_error);
+
+	trace_xfs_buf_iowait_done(bp, _RET_IP_);
 	return bp->b_error;
 }
 
@@ -1604,7 +1582,8 @@ xfs_buf_delwri_queue(
 	struct list_head	*dwq = &bp->b_target->bt_delwrite_queue;
 	spinlock_t		*dwlk = &bp->b_target->bt_delwrite_lock;
 
-	XB_TRACE(bp, "delwri_q", (long)unlock);
+	trace_xfs_buf_delwri_queue(bp, _RET_IP_);
+
 	ASSERT((bp->b_flags&(XBF_DELWRI|XBF_ASYNC)) == (XBF_DELWRI|XBF_ASYNC));
 
 	spin_lock(dwlk);
@@ -1644,7 +1623,7 @@ xfs_buf_delwri_dequeue(
 	if (dequeued)
 		xfs_buf_rele(bp);
 
-	XB_TRACE(bp, "delwri_dq", (long)dequeued);
+	trace_xfs_buf_delwri_dequeue(bp, _RET_IP_);
 }
 
 STATIC void
@@ -1692,7 +1671,7 @@ xfs_buf_delwri_split(
 	INIT_LIST_HEAD(list);
 	spin_lock(dwlk);
 	list_for_each_entry_safe(bp, n, dwq, b_list) {
-		XB_TRACE(bp, "walkq1", (long)xfs_buf_ispin(bp));
+		trace_xfs_buf_delwri_split(bp, _RET_IP_);
 		ASSERT(bp->b_flags & XBF_DELWRI);
 
 		if (!xfs_buf_ispin(bp) && !xfs_buf_cond_lock(bp)) {
@@ -1816,14 +1795,10 @@ xfs_flush_buftarg(
 int __init
 xfs_buf_init(void)
 {
-#ifdef XFS_BUF_TRACE
-	xfs_buf_trace_buf = ktrace_alloc(XFS_BUF_TRACE_SIZE, KM_NOFS);
-#endif
-
 	xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf",
 						KM_ZONE_HWALIGN, NULL);
 	if (!xfs_buf_zone)
-		goto out_free_trace_buf;
+		goto out;
 
 	xfslogd_workqueue = create_workqueue("xfslogd");
 	if (!xfslogd_workqueue)
@@ -1846,10 +1821,7 @@ xfs_buf_init(void)
 	destroy_workqueue(xfslogd_workqueue);
  out_free_buf_zone:
 	kmem_zone_destroy(xfs_buf_zone);
- out_free_trace_buf:
-#ifdef XFS_BUF_TRACE
-	ktrace_free(xfs_buf_trace_buf);
-#endif
+ out:
 	return -ENOMEM;
 }
 
@@ -1861,9 +1833,6 @@ xfs_buf_terminate(void)
 	destroy_workqueue(xfsdatad_workqueue);
 	destroy_workqueue(xfslogd_workqueue);
 	kmem_zone_destroy(xfs_buf_zone);
-#ifdef XFS_BUF_TRACE
-	ktrace_free(xfs_buf_trace_buf);
-#endif
 }
 
 #ifdef CONFIG_KDB_MODULES
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index 9b4d666ad31f..a34c7b54822d 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -55,6 +55,7 @@ typedef enum {
 	XBF_FS_MANAGED = (1 << 8),  /* filesystem controls freeing memory  */
  	XBF_ORDERED = (1 << 11),    /* use ordered writes		   */
 	XBF_READ_AHEAD = (1 << 12), /* asynchronous read-ahead		   */
+	XBF_LOG_BUFFER = (1 << 13), /* this is a buffer used for the log   */
 
 	/* flags used only as arguments to access routines */
 	XBF_LOCK = (1 << 14),       /* lock requested			   */
@@ -95,6 +96,28 @@ typedef enum {
 	_XFS_BARRIER_FAILED = (1 << 23),
 } xfs_buf_flags_t;
 
+#define XFS_BUF_FLAGS \
+	{ XBF_READ,		"READ" }, \
+	{ XBF_WRITE,		"WRITE" }, \
+	{ XBF_MAPPED,		"MAPPED" }, \
+	{ XBF_ASYNC,		"ASYNC" }, \
+	{ XBF_DONE,		"DONE" }, \
+	{ XBF_DELWRI,		"DELWRI" }, \
+	{ XBF_STALE,		"STALE" }, \
+	{ XBF_FS_MANAGED,	"FS_MANAGED" }, \
+	{ XBF_ORDERED,		"ORDERED" }, \
+	{ XBF_READ_AHEAD,	"READ_AHEAD" }, \
+	{ XBF_LOCK,		"LOCK" },  	/* should never be set */\
+	{ XBF_TRYLOCK,		"TRYLOCK" }, 	/* ditto */\
+	{ XBF_DONT_BLOCK,	"DONT_BLOCK" },	/* ditto */\
+	{ _XBF_PAGE_CACHE,	"PAGE_CACHE" }, \
+	{ _XBF_PAGES,		"PAGES" }, \
+	{ _XBF_RUN_QUEUES,	"RUN_QUEUES" }, \
+	{ _XBF_DELWRI_Q,	"DELWRI_Q" }, \
+	{ _XBF_PAGE_LOCKED,	"PAGE_LOCKED" }, \
+	{ _XFS_BARRIER_FAILED,	"BARRIER_FAILED" }
+
+
 typedef enum {
 	XBT_FORCE_SLEEP = 0,
 	XBT_FORCE_FLUSH = 1,
@@ -186,15 +209,10 @@ extern xfs_buf_t *_xfs_buf_find(xfs_buftarg_t *, xfs_off_t, size_t,
 #define xfs_incore(buftarg,blkno,len,lockit) \
 	_xfs_buf_find(buftarg, blkno ,len, lockit, NULL)
 
-extern xfs_buf_t *xfs_buf_get_flags(xfs_buftarg_t *, xfs_off_t, size_t,
+extern xfs_buf_t *xfs_buf_get(xfs_buftarg_t *, xfs_off_t, size_t,
 				xfs_buf_flags_t);
-#define xfs_buf_get(target, blkno, len, flags) \
-	xfs_buf_get_flags((target), (blkno), (len), XBF_LOCK | XBF_MAPPED)
-
-extern xfs_buf_t *xfs_buf_read_flags(xfs_buftarg_t *, xfs_off_t, size_t,
+extern xfs_buf_t *xfs_buf_read(xfs_buftarg_t *, xfs_off_t, size_t,
 				xfs_buf_flags_t);
-#define xfs_buf_read(target, blkno, len, flags) \
-	xfs_buf_read_flags((target), (blkno), (len), XBF_LOCK | XBF_MAPPED)
 
 extern xfs_buf_t *xfs_buf_get_empty(size_t, xfs_buftarg_t *);
 extern xfs_buf_t *xfs_buf_get_noaddr(size_t, xfs_buftarg_t *);
@@ -248,13 +266,6 @@ extern void xfs_buf_delwri_dequeue(xfs_buf_t *);
 extern int xfs_buf_init(void);
 extern void xfs_buf_terminate(void);
 
-#ifdef XFS_BUF_TRACE
-extern ktrace_t *xfs_buf_trace_buf;
-extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *);
-#else
-#define xfs_buf_trace(bp,id,ptr,ra)	do { } while (0)
-#endif
-
 #define xfs_buf_target_name(target)	\
 	({ char __b[BDEVNAME_SIZE]; bdevname((target)->bt_bdev, __b); __b; })
 
@@ -370,10 +381,6 @@ static inline void xfs_buf_relse(xfs_buf_t *bp)
 
 #define xfs_bpin(bp)		xfs_buf_pin(bp)
 #define xfs_bunpin(bp)		xfs_buf_unpin(bp)
-
-#define xfs_buftrace(id, bp)	\
-	    xfs_buf_trace(bp, id, NULL, (void *)__builtin_return_address(0))
-
 #define xfs_biodone(bp)		xfs_buf_ioend(bp, 0)
 
 #define xfs_biomove(bp, off, len, data, rw) \
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index eff61e2732af..e4caeb28ce2e 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -52,7 +52,7 @@ xfs_file_aio_read(
 	loff_t			pos)
 {
 	struct file		*file = iocb->ki_filp;
-	int			ioflags = IO_ISAIO;
+	int			ioflags = 0;
 
 	BUG_ON(iocb->ki_pos != pos);
 	if (unlikely(file->f_flags & O_DIRECT))
@@ -71,7 +71,7 @@ xfs_file_aio_write(
 	loff_t			pos)
 {
 	struct file		*file = iocb->ki_filp;
-	int			ioflags = IO_ISAIO;
+	int			ioflags = 0;
 
 	BUG_ON(iocb->ki_pos != pos);
 	if (unlikely(file->f_flags & O_DIRECT))
diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c
index 08be36d7326c..7501b85fd860 100644
--- a/fs/xfs/linux-2.6/xfs_fs_subr.c
+++ b/fs/xfs/linux-2.6/xfs_fs_subr.c
@@ -19,6 +19,7 @@
 #include "xfs_vnodeops.h"
 #include "xfs_bmap_btree.h"
 #include "xfs_inode.h"
+#include "xfs_trace.h"
 
 int  fs_noerr(void) { return 0; }
 int  fs_nosys(void) { return ENOSYS; }
@@ -51,6 +52,8 @@ xfs_flushinval_pages(
 	struct address_space *mapping = VFS_I(ip)->i_mapping;
 	int		ret = 0;
 
+	trace_xfs_pagecache_inval(ip, first, last);
+
 	if (mapping->nrpages) {
 		xfs_iflags_clear(ip, XFS_ITRUNCATED);
 		ret = filemap_write_and_wait(mapping);
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index 5bb523d7f37e..a034cf624437 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -51,6 +51,7 @@
 #include "xfs_quota.h"
 #include "xfs_inode_item.h"
 #include "xfs_export.h"
+#include "xfs_trace.h"
 
 #include <linux/capability.h>
 #include <linux/dcache.h>
diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c
index eafcc7c18706..be1527b1670c 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl32.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl32.c
@@ -46,6 +46,7 @@
 #include "xfs_attr.h"
 #include "xfs_ioctl.h"
 #include "xfs_ioctl32.h"
+#include "xfs_trace.h"
 
 #define  _NATIVE_IOC(cmd, type) \
 	  _IOC(_IOC_DIR(cmd), _IOC_TYPE(cmd), _IOC_NR(cmd), sizeof(type))
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index cd42ef78f6b5..225946012d0b 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -47,6 +47,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_utils.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 #include <linux/capability.h>
 #include <linux/xattr.h>
@@ -573,8 +574,8 @@ xfs_vn_fallocate(
 	bf.l_len = len;
 
 	xfs_ilock(ip, XFS_IOLOCK_EXCL);
-	error = xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
-				      0, XFS_ATTR_NOLOCK);
+	error = -xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
+				       0, XFS_ATTR_NOLOCK);
 	if (!error && !(mode & FALLOC_FL_KEEP_SIZE) &&
 	    offset + len > i_size_read(inode))
 		new_size = offset + len;
@@ -585,7 +586,7 @@ xfs_vn_fallocate(
 
 		iattr.ia_valid = ATTR_SIZE;
 		iattr.ia_size = new_size;
-		error = xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
+		error = -xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
 	}
 
 	xfs_iunlock(ip, XFS_IOLOCK_EXCL);
@@ -793,7 +794,7 @@ xfs_setup_inode(
 	struct inode		*inode = &ip->i_vnode;
 
 	inode->i_ino = ip->i_ino;
-	inode->i_state = I_NEW|I_LOCK;
+	inode->i_state = I_NEW;
 	inode_add_to_lists(ip->i_mount->m_super, inode);
 
 	inode->i_mode	= ip->i_d.di_mode;
diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h
index 6127e24062d0..5af0c81ca1ae 100644
--- a/fs/xfs/linux-2.6/xfs_linux.h
+++ b/fs/xfs/linux-2.6/xfs_linux.h
@@ -40,7 +40,6 @@
 #include <sv.h>
 #include <time.h>
 
-#include <support/ktrace.h>
 #include <support/debug.h>
 #include <support/uuid.h>
 
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index 072050f8d346..0d32457abef1 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -48,73 +48,12 @@
 #include "xfs_utils.h"
 #include "xfs_iomap.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 #include <linux/capability.h>
 #include <linux/writeback.h>
 
 
-#if defined(XFS_RW_TRACE)
-void
-xfs_rw_enter_trace(
-	int			tag,
-	xfs_inode_t		*ip,
-	void			*data,
-	size_t			segs,
-	loff_t			offset,
-	int			ioflags)
-{
-	if (ip->i_rwtrace == NULL)
-		return;
-	ktrace_enter(ip->i_rwtrace,
-		(void *)(unsigned long)tag,
-		(void *)ip,
-		(void *)((unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(ip->i_d.di_size & 0xffffffff)),
-		(void *)data,
-		(void *)((unsigned long)segs),
-		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(offset & 0xffffffff)),
-		(void *)((unsigned long)ioflags),
-		(void *)((unsigned long)((ip->i_new_size >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(ip->i_new_size & 0xffffffff)),
-		(void *)((unsigned long)current_pid()),
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL);
-}
-
-void
-xfs_inval_cached_trace(
-	xfs_inode_t	*ip,
-	xfs_off_t	offset,
-	xfs_off_t	len,
-	xfs_off_t	first,
-	xfs_off_t	last)
-{
-
-	if (ip->i_rwtrace == NULL)
-		return;
-	ktrace_enter(ip->i_rwtrace,
-		(void *)(__psint_t)XFS_INVAL_CACHED,
-		(void *)ip,
-		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(offset & 0xffffffff)),
-		(void *)((unsigned long)((len >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(len & 0xffffffff)),
-		(void *)((unsigned long)((first >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(first & 0xffffffff)),
-		(void *)((unsigned long)((last >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(last & 0xffffffff)),
-		(void *)((unsigned long)current_pid()),
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL);
-}
-#endif
-
 /*
  *	xfs_iozero
  *
@@ -250,13 +189,10 @@ xfs_read(
 		}
 	}
 
-	xfs_rw_enter_trace(XFS_READ_ENTER, ip,
-				(void *)iovp, segs, *offset, ioflags);
+	trace_xfs_file_read(ip, size, *offset, ioflags);
 
 	iocb->ki_pos = *offset;
 	ret = generic_file_aio_read(iocb, iovp, segs, *offset);
-	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
-		ret = wait_on_sync_kiocb(iocb);
 	if (ret > 0)
 		XFS_STATS_ADD(xs_read_bytes, ret);
 
@@ -294,8 +230,9 @@ xfs_splice_read(
 			return -error;
 		}
 	}
-	xfs_rw_enter_trace(XFS_SPLICE_READ_ENTER, ip,
-			   pipe, count, *ppos, ioflags);
+
+	trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
+
 	ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
 	if (ret > 0)
 		XFS_STATS_ADD(xs_read_bytes, ret);
@@ -344,8 +281,8 @@ xfs_splice_write(
 		ip->i_new_size = new_size;
 	xfs_iunlock(ip, XFS_ILOCK_EXCL);
 
-	xfs_rw_enter_trace(XFS_SPLICE_WRITE_ENTER, ip,
-			   pipe, count, *ppos, ioflags);
+	trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
+
 	ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
 	if (ret > 0)
 		XFS_STATS_ADD(xs_write_bytes, ret);
@@ -712,8 +649,6 @@ start:
 	if ((ioflags & IO_ISDIRECT)) {
 		if (mapping->nrpages) {
 			WARN_ON(need_i_mutex == 0);
-			xfs_inval_cached_trace(xip, pos, -1,
-					(pos & PAGE_CACHE_MASK), -1);
 			error = xfs_flushinval_pages(xip,
 					(pos & PAGE_CACHE_MASK),
 					-1, FI_REMAPF_LOCKED);
@@ -730,8 +665,7 @@ start:
 			need_i_mutex = 0;
 		}
 
- 		xfs_rw_enter_trace(XFS_DIOWR_ENTER, xip, (void *)iovp, segs,
-				*offset, ioflags);
+		trace_xfs_file_direct_write(xip, count, *offset, ioflags);
 		ret = generic_file_direct_write(iocb, iovp,
 				&segs, pos, offset, count, ocount);
 
@@ -754,8 +688,7 @@ start:
 		ssize_t ret2 = 0;
 
 write_retry:
-		xfs_rw_enter_trace(XFS_WRITE_ENTER, xip, (void *)iovp, segs,
-				*offset, ioflags);
+		trace_xfs_file_buffered_write(xip, count, *offset, ioflags);
 		ret2 = generic_file_buffered_write(iocb, iovp, segs,
 				pos, offset, count, ret);
 		/*
@@ -774,9 +707,6 @@ write_retry:
 
 	current->backing_dev_info = NULL;
 
-	if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
-		ret = wait_on_sync_kiocb(iocb);
-
 	isize = i_size_read(inode);
 	if (unlikely(ret < 0 && ret != -EFAULT && *offset > isize))
 		*offset = isize;
@@ -811,7 +741,7 @@ write_retry:
 	XFS_STATS_ADD(xs_write_bytes, ret);
 
 	/* Handle various SYNC-type writes */
-	if ((file->f_flags & O_SYNC) || IS_SYNC(inode)) {
+	if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
 		loff_t end = pos + ret - 1;
 		int error2;
 
@@ -863,7 +793,7 @@ int
 xfs_bdstrat_cb(struct xfs_buf *bp)
 {
 	if (XFS_FORCED_SHUTDOWN(bp->b_mount)) {
-		xfs_buftrace("XFS__BDSTRAT IOERROR", bp);
+		trace_xfs_bdstrat_shut(bp, _RET_IP_);
 		/*
 		 * Metadata write that didn't get logged but
 		 * written delayed anyway. These aren't associated
@@ -896,7 +826,7 @@ xfsbdstrat(
 		return;
 	}
 
-	xfs_buftrace("XFSBDSTRAT IOERROR", bp);
+	trace_xfs_bdstrat_shut(bp, _RET_IP_);
 	xfs_bioerror_relse(bp);
 }
 
diff --git a/fs/xfs/linux-2.6/xfs_lrw.h b/fs/xfs/linux-2.6/xfs_lrw.h
index e6be37dbd0e9..d1f7789c7ffb 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.h
+++ b/fs/xfs/linux-2.6/xfs_lrw.h
@@ -20,52 +20,7 @@
 
 struct xfs_mount;
 struct xfs_inode;
-struct xfs_bmbt_irec;
 struct xfs_buf;
-struct xfs_iomap;
-
-#if defined(XFS_RW_TRACE)
-/*
- * Defines for the trace mechanisms in xfs_lrw.c.
- */
-#define	XFS_RW_KTRACE_SIZE	128
-
-#define	XFS_READ_ENTER		1
-#define	XFS_WRITE_ENTER		2
-#define XFS_IOMAP_READ_ENTER	3
-#define	XFS_IOMAP_WRITE_ENTER	4
-#define	XFS_IOMAP_READ_MAP	5
-#define	XFS_IOMAP_WRITE_MAP	6
-#define	XFS_IOMAP_WRITE_NOSPACE	7
-#define	XFS_ITRUNC_START	8
-#define	XFS_ITRUNC_FINISH1	9
-#define	XFS_ITRUNC_FINISH2	10
-#define	XFS_CTRUNC1		11
-#define	XFS_CTRUNC2		12
-#define	XFS_CTRUNC3		13
-#define	XFS_CTRUNC4		14
-#define	XFS_CTRUNC5		15
-#define	XFS_CTRUNC6		16
-#define	XFS_BUNMAP		17
-#define	XFS_INVAL_CACHED	18
-#define	XFS_DIORD_ENTER		19
-#define	XFS_DIOWR_ENTER		20
-#define	XFS_WRITEPAGE_ENTER	22
-#define	XFS_RELEASEPAGE_ENTER	23
-#define	XFS_INVALIDPAGE_ENTER	24
-#define	XFS_IOMAP_ALLOC_ENTER	25
-#define	XFS_IOMAP_ALLOC_MAP	26
-#define	XFS_IOMAP_UNWRITTEN	27
-#define XFS_SPLICE_READ_ENTER	28
-#define XFS_SPLICE_WRITE_ENTER	29
-extern void xfs_rw_enter_trace(int, struct xfs_inode *,
-		void *, size_t, loff_t, int);
-extern void xfs_inval_cached_trace(struct xfs_inode *,
-		xfs_off_t, xfs_off_t, xfs_off_t, xfs_off_t);
-#else
-#define xfs_rw_enter_trace(tag, ip, data, size, offset, ioflags)
-#define xfs_inval_cached_trace(ip, offset, len, first, last)
-#endif
 
 /* errors from xfsbdstrat() must be extracted from the buffer */
 extern void xfsbdstrat(struct xfs_mount *, struct xfs_buf *);
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 18a4b8e11df2..77414db10dc2 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -15,6 +15,7 @@
  * along with this program; if not, write the Free Software Foundation,
  * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
  */
+
 #include "xfs.h"
 #include "xfs_bit.h"
 #include "xfs_log.h"
@@ -52,11 +53,11 @@
 #include "xfs_trans_priv.h"
 #include "xfs_filestream.h"
 #include "xfs_da_btree.h"
-#include "xfs_dir2_trace.h"
 #include "xfs_extfree_item.h"
 #include "xfs_mru_cache.h"
 #include "xfs_inode_item.h"
 #include "xfs_sync.h"
+#include "xfs_trace.h"
 
 #include <linux/namei.h>
 #include <linux/init.h>
@@ -930,13 +931,37 @@ xfs_fs_alloc_inode(
  */
 STATIC void
 xfs_fs_destroy_inode(
-	struct inode	*inode)
+	struct inode		*inode)
 {
-	xfs_inode_t		*ip = XFS_I(inode);
+	struct xfs_inode	*ip = XFS_I(inode);
+
+	xfs_itrace_entry(ip);
 
 	XFS_STATS_INC(vn_reclaim);
-	if (xfs_reclaim(ip))
-		panic("%s: cannot reclaim 0x%p\n", __func__, inode);
+
+	/* bad inode, get out here ASAP */
+	if (is_bad_inode(inode))
+		goto out_reclaim;
+
+	xfs_ioend_wait(ip);
+
+	ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0);
+
+	/*
+	 * We should never get here with one of the reclaim flags already set.
+	 */
+	ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIMABLE));
+	ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIM));
+
+	/*
+	 * We always use background reclaim here because even if the
+	 * inode is clean, it still may be under IO and hence we have
+	 * to take the flush lock. The background reclaim path handles
+	 * this more efficiently than we can here, so simply let background
+	 * reclaim tear down all inodes.
+	 */
+out_reclaim:
+	xfs_inode_set_reclaim_tag(ip);
 }
 
 /*
@@ -973,7 +998,6 @@ xfs_fs_inode_init_once(
 
 	mrlock_init(&ip->i_lock, MRLOCK_ALLOW_EQUAL_PRI|MRLOCK_BARRIER,
 		     "xfsino", ip->i_ino);
-	mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
 }
 
 /*
@@ -1075,6 +1099,20 @@ xfs_fs_clear_inode(
 	XFS_STATS_INC(vn_remove);
 	XFS_STATS_DEC(vn_active);
 
+	/*
+	 * The iolock is used by the file system to coordinate reads,
+	 * writes, and block truncates.  Up to this point the lock
+	 * protected concurrent accesses by users of the inode.  But
+	 * from here forward we're doing some final processing of the
+	 * inode because we're done with it, and although we reuse the
+	 * iolock for protection it is really a distinct lock class
+	 * (in the lockdep sense) from before.  To keep lockdep happy
+	 * (and basically indicate what we are doing), we explicitly
+	 * re-init the iolock here.
+	 */
+	ASSERT(!rwsem_is_locked(&ip->i_iolock.mr_lock));
+	mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
+
 	xfs_inactive(ip);
 }
 
@@ -1092,8 +1130,6 @@ xfs_fs_put_super(
 	struct super_block	*sb)
 {
 	struct xfs_mount	*mp = XFS_M(sb);
-	struct xfs_inode	*rip = mp->m_rootip;
-	int			unmount_event_flags = 0;
 
 	xfs_syncd_stop(mp);
 
@@ -1109,20 +1145,7 @@ xfs_fs_put_super(
 		xfs_sync_attr(mp, 0);
 	}
 
-#ifdef HAVE_DMAPI
-	if (mp->m_flags & XFS_MOUNT_DMAPI) {
-		unmount_event_flags =
-			(mp->m_dmevmask & (1 << DM_EVENT_UNMOUNT)) ?
-				0 : DM_FLAGS_UNWANTED;
-		/*
-		 * Ignore error from dmapi here, first unmount is not allowed
-		 * to fail anyway, and second we wouldn't want to fail a
-		 * unmount because of dmapi.
-		 */
-		XFS_SEND_PREUNMOUNT(mp, rip, DM_RIGHT_NULL, rip, DM_RIGHT_NULL,
-				NULL, NULL, 0, 0, unmount_event_flags);
-	}
-#endif
+	XFS_SEND_PREUNMOUNT(mp);
 
 	/*
 	 * Blow away any referenced inode in the filestreams cache.
@@ -1133,10 +1156,7 @@ xfs_fs_put_super(
 
 	XFS_bflush(mp->m_ddev_targp);
 
-	if (mp->m_flags & XFS_MOUNT_DMAPI) {
-		XFS_SEND_UNMOUNT(mp, rip, DM_RIGHT_NULL, 0, 0,
-				unmount_event_flags);
-	}
+	XFS_SEND_UNMOUNT(mp);
 
 	xfs_unmountfs(mp);
 	xfs_freesb(mp);
@@ -1504,8 +1524,6 @@ xfs_fs_fill_super(
 		goto fail_vnrele;
 
 	kfree(mtpt);
-
-	xfs_itrace_exit(XFS_I(sb->s_root->d_inode));
 	return 0;
 
  out_filestream_unmount:
@@ -1581,94 +1599,6 @@ static struct file_system_type xfs_fs_type = {
 };
 
 STATIC int __init
-xfs_alloc_trace_bufs(void)
-{
-#ifdef XFS_ALLOC_TRACE
-	xfs_alloc_trace_buf = ktrace_alloc(XFS_ALLOC_TRACE_SIZE, KM_MAYFAIL);
-	if (!xfs_alloc_trace_buf)
-		goto out;
-#endif
-#ifdef XFS_BMAP_TRACE
-	xfs_bmap_trace_buf = ktrace_alloc(XFS_BMAP_TRACE_SIZE, KM_MAYFAIL);
-	if (!xfs_bmap_trace_buf)
-		goto out_free_alloc_trace;
-#endif
-#ifdef XFS_BTREE_TRACE
-	xfs_allocbt_trace_buf = ktrace_alloc(XFS_ALLOCBT_TRACE_SIZE,
-					     KM_MAYFAIL);
-	if (!xfs_allocbt_trace_buf)
-		goto out_free_bmap_trace;
-
-	xfs_inobt_trace_buf = ktrace_alloc(XFS_INOBT_TRACE_SIZE, KM_MAYFAIL);
-	if (!xfs_inobt_trace_buf)
-		goto out_free_allocbt_trace;
-
-	xfs_bmbt_trace_buf = ktrace_alloc(XFS_BMBT_TRACE_SIZE, KM_MAYFAIL);
-	if (!xfs_bmbt_trace_buf)
-		goto out_free_inobt_trace;
-#endif
-#ifdef XFS_ATTR_TRACE
-	xfs_attr_trace_buf = ktrace_alloc(XFS_ATTR_TRACE_SIZE, KM_MAYFAIL);
-	if (!xfs_attr_trace_buf)
-		goto out_free_bmbt_trace;
-#endif
-#ifdef XFS_DIR2_TRACE
-	xfs_dir2_trace_buf = ktrace_alloc(XFS_DIR2_GTRACE_SIZE, KM_MAYFAIL);
-	if (!xfs_dir2_trace_buf)
-		goto out_free_attr_trace;
-#endif
-
-	return 0;
-
-#ifdef XFS_DIR2_TRACE
- out_free_attr_trace:
-#endif
-#ifdef XFS_ATTR_TRACE
-	ktrace_free(xfs_attr_trace_buf);
- out_free_bmbt_trace:
-#endif
-#ifdef XFS_BTREE_TRACE
-	ktrace_free(xfs_bmbt_trace_buf);
- out_free_inobt_trace:
-	ktrace_free(xfs_inobt_trace_buf);
- out_free_allocbt_trace:
-	ktrace_free(xfs_allocbt_trace_buf);
- out_free_bmap_trace:
-#endif
-#ifdef XFS_BMAP_TRACE
-	ktrace_free(xfs_bmap_trace_buf);
- out_free_alloc_trace:
-#endif
-#ifdef XFS_ALLOC_TRACE
-	ktrace_free(xfs_alloc_trace_buf);
- out:
-#endif
-	return -ENOMEM;
-}
-
-STATIC void
-xfs_free_trace_bufs(void)
-{
-#ifdef XFS_DIR2_TRACE
-	ktrace_free(xfs_dir2_trace_buf);
-#endif
-#ifdef XFS_ATTR_TRACE
-	ktrace_free(xfs_attr_trace_buf);
-#endif
-#ifdef XFS_BTREE_TRACE
-	ktrace_free(xfs_bmbt_trace_buf);
-	ktrace_free(xfs_inobt_trace_buf);
-	ktrace_free(xfs_allocbt_trace_buf);
-#endif
-#ifdef XFS_BMAP_TRACE
-	ktrace_free(xfs_bmap_trace_buf);
-#endif
-#ifdef XFS_ALLOC_TRACE
-	ktrace_free(xfs_alloc_trace_buf);
-#endif
-}
-
-STATIC int __init
 xfs_init_zones(void)
 {
 
@@ -1809,7 +1739,6 @@ init_xfs_fs(void)
 	printk(KERN_INFO XFS_VERSION_STRING " with "
 			 XFS_BUILD_OPTIONS " enabled\n");
 
-	ktrace_init(64);
 	xfs_ioend_init();
 	xfs_dir_startup();
 
@@ -1817,13 +1746,9 @@ init_xfs_fs(void)
 	if (error)
 		goto out;
 
-	error = xfs_alloc_trace_bufs();
-	if (error)
-		goto out_destroy_zones;
-
 	error = xfs_mru_cache_init();
 	if (error)
-		goto out_free_trace_buffers;
+		goto out_destroy_zones;
 
 	error = xfs_filestream_init();
 	if (error)
@@ -1858,8 +1783,6 @@ init_xfs_fs(void)
 	xfs_filestream_uninit();
  out_mru_cache_uninit:
 	xfs_mru_cache_uninit();
- out_free_trace_buffers:
-	xfs_free_trace_bufs();
  out_destroy_zones:
 	xfs_destroy_zones();
  out:
@@ -1876,9 +1799,7 @@ exit_xfs_fs(void)
 	xfs_buf_terminate();
 	xfs_filestream_uninit();
 	xfs_mru_cache_uninit();
-	xfs_free_trace_bufs();
 	xfs_destroy_zones();
-	ktrace_uninit();
 }
 
 module_init(init_xfs_fs);
diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h
index 18175ebd58ed..233d4b9881b1 100644
--- a/fs/xfs/linux-2.6/xfs_super.h
+++ b/fs/xfs/linux-2.6/xfs_super.h
@@ -56,12 +56,6 @@ extern void xfs_qm_exit(void);
 # define XFS_BIGFS_STRING
 #endif
 
-#ifdef CONFIG_XFS_TRACE
-# define XFS_TRACE_STRING	"tracing, "
-#else
-# define XFS_TRACE_STRING
-#endif
-
 #ifdef CONFIG_XFS_DMAPI
 # define XFS_DMAPI_STRING	"dmapi support, "
 #else
@@ -78,7 +72,6 @@ extern void xfs_qm_exit(void);
 				XFS_SECURITY_STRING \
 				XFS_REALTIME_STRING \
 				XFS_BIGFS_STRING \
-				XFS_TRACE_STRING \
 				XFS_DMAPI_STRING \
 				XFS_DBG_STRING /* DBG must be last */
 
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index 961df0a22c78..1f5e4bb5e970 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -44,6 +44,7 @@
 #include "xfs_inode_item.h"
 #include "xfs_rw.h"
 #include "xfs_quota.h"
+#include "xfs_trace.h"
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
@@ -64,7 +65,6 @@ xfs_inode_ag_lookup(
 	 * as the tree is sparse and a gang lookup walks to find
 	 * the number of objects requested.
 	 */
-	read_lock(&pag->pag_ici_lock);
 	if (tag == XFS_ICI_NO_TAG) {
 		nr_found = radix_tree_gang_lookup(&pag->pag_ici_root,
 				(void **)&ip, *first_index, 1);
@@ -73,7 +73,7 @@ xfs_inode_ag_lookup(
 				(void **)&ip, *first_index, 1, tag);
 	}
 	if (!nr_found)
-		goto unlock;
+		return NULL;
 
 	/*
 	 * Update the index for the next lookup. Catch overflows
@@ -83,13 +83,8 @@ xfs_inode_ag_lookup(
 	 */
 	*first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
 	if (*first_index < XFS_INO_TO_AGINO(mp, ip->i_ino))
-		goto unlock;
-
+		return NULL;
 	return ip;
-
-unlock:
-	read_unlock(&pag->pag_ici_lock);
-	return NULL;
 }
 
 STATIC int
@@ -99,7 +94,8 @@ xfs_inode_ag_walk(
 	int			(*execute)(struct xfs_inode *ip,
 					   struct xfs_perag *pag, int flags),
 	int			flags,
-	int			tag)
+	int			tag,
+	int			exclusive)
 {
 	struct xfs_perag	*pag = &mp->m_perag[ag];
 	uint32_t		first_index;
@@ -113,10 +109,20 @@ restart:
 		int		error = 0;
 		xfs_inode_t	*ip;
 
+		if (exclusive)
+			write_lock(&pag->pag_ici_lock);
+		else
+			read_lock(&pag->pag_ici_lock);
 		ip = xfs_inode_ag_lookup(mp, pag, &first_index, tag);
-		if (!ip)
+		if (!ip) {
+			if (exclusive)
+				write_unlock(&pag->pag_ici_lock);
+			else
+				read_unlock(&pag->pag_ici_lock);
 			break;
+		}
 
+		/* execute releases pag->pag_ici_lock */
 		error = execute(ip, pag, flags);
 		if (error == EAGAIN) {
 			skipped++;
@@ -124,9 +130,8 @@ restart:
 		}
 		if (error)
 			last_error = error;
-		/*
-		 * bail out if the filesystem is corrupted.
-		 */
+
+		/* bail out if the filesystem is corrupted.  */
 		if (error == EFSCORRUPTED)
 			break;
 
@@ -147,7 +152,8 @@ xfs_inode_ag_iterator(
 	int			(*execute)(struct xfs_inode *ip,
 					   struct xfs_perag *pag, int flags),
 	int			flags,
-	int			tag)
+	int			tag,
+	int			exclusive)
 {
 	int			error = 0;
 	int			last_error = 0;
@@ -156,7 +162,8 @@ xfs_inode_ag_iterator(
 	for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
 		if (!mp->m_perag[ag].pag_ici_init)
 			continue;
-		error = xfs_inode_ag_walk(mp, ag, execute, flags, tag);
+		error = xfs_inode_ag_walk(mp, ag, execute, flags, tag,
+						exclusive);
 		if (error) {
 			last_error = error;
 			if (error == EFSCORRUPTED)
@@ -173,30 +180,31 @@ xfs_sync_inode_valid(
 	struct xfs_perag	*pag)
 {
 	struct inode		*inode = VFS_I(ip);
+	int			error = EFSCORRUPTED;
 
 	/* nothing to sync during shutdown */
-	if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
-		read_unlock(&pag->pag_ici_lock);
-		return EFSCORRUPTED;
-	}
+	if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+		goto out_unlock;
 
-	/*
-	 * If we can't get a reference on the inode, it must be in reclaim.
-	 * Leave it for the reclaim code to flush. Also avoid inodes that
-	 * haven't been fully initialised.
-	 */
-	if (!igrab(inode)) {
-		read_unlock(&pag->pag_ici_lock);
-		return ENOENT;
-	}
-	read_unlock(&pag->pag_ici_lock);
+	/* avoid new or reclaimable inodes. Leave for reclaim code to flush */
+	error = ENOENT;
+	if (xfs_iflags_test(ip, XFS_INEW | XFS_IRECLAIMABLE | XFS_IRECLAIM))
+		goto out_unlock;
 
-	if (is_bad_inode(inode) || xfs_iflags_test(ip, XFS_INEW)) {
+	/* If we can't grab the inode, it must on it's way to reclaim. */
+	if (!igrab(inode))
+		goto out_unlock;
+
+	if (is_bad_inode(inode)) {
 		IRELE(ip);
-		return ENOENT;
+		goto out_unlock;
 	}
 
-	return 0;
+	/* inode is valid */
+	error = 0;
+out_unlock:
+	read_unlock(&pag->pag_ici_lock);
+	return error;
 }
 
 STATIC int
@@ -281,7 +289,7 @@ xfs_sync_data(
 	ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0);
 
 	error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags,
-				      XFS_ICI_NO_TAG);
+				      XFS_ICI_NO_TAG, 0);
 	if (error)
 		return XFS_ERROR(error);
 
@@ -303,7 +311,7 @@ xfs_sync_attr(
 	ASSERT((flags & ~SYNC_WAIT) == 0);
 
 	return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags,
-				     XFS_ICI_NO_TAG);
+				     XFS_ICI_NO_TAG, 0);
 }
 
 STATIC int
@@ -663,67 +671,6 @@ xfs_syncd_stop(
 	kthread_stop(mp->m_sync_task);
 }
 
-int
-xfs_reclaim_inode(
-	xfs_inode_t	*ip,
-	int		locked,
-	int		sync_mode)
-{
-	xfs_perag_t	*pag = xfs_get_perag(ip->i_mount, ip->i_ino);
-
-	/* The hash lock here protects a thread in xfs_iget_core from
-	 * racing with us on linking the inode back with a vnode.
-	 * Once we have the XFS_IRECLAIM flag set it will not touch
-	 * us.
-	 */
-	write_lock(&pag->pag_ici_lock);
-	spin_lock(&ip->i_flags_lock);
-	if (__xfs_iflags_test(ip, XFS_IRECLAIM) ||
-	    !__xfs_iflags_test(ip, XFS_IRECLAIMABLE)) {
-		spin_unlock(&ip->i_flags_lock);
-		write_unlock(&pag->pag_ici_lock);
-		if (locked) {
-			xfs_ifunlock(ip);
-			xfs_iunlock(ip, XFS_ILOCK_EXCL);
-		}
-		return -EAGAIN;
-	}
-	__xfs_iflags_set(ip, XFS_IRECLAIM);
-	spin_unlock(&ip->i_flags_lock);
-	write_unlock(&pag->pag_ici_lock);
-	xfs_put_perag(ip->i_mount, pag);
-
-	/*
-	 * If the inode is still dirty, then flush it out.  If the inode
-	 * is not in the AIL, then it will be OK to flush it delwri as
-	 * long as xfs_iflush() does not keep any references to the inode.
-	 * We leave that decision up to xfs_iflush() since it has the
-	 * knowledge of whether it's OK to simply do a delwri flush of
-	 * the inode or whether we need to wait until the inode is
-	 * pulled from the AIL.
-	 * We get the flush lock regardless, though, just to make sure
-	 * we don't free it while it is being flushed.
-	 */
-	if (!locked) {
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		xfs_iflock(ip);
-	}
-
-	/*
-	 * In the case of a forced shutdown we rely on xfs_iflush() to
-	 * wait for the inode to be unpinned before returning an error.
-	 */
-	if (!is_bad_inode(VFS_I(ip)) && xfs_iflush(ip, sync_mode) == 0) {
-		/* synchronize with xfs_iflush_done */
-		xfs_iflock(ip);
-		xfs_ifunlock(ip);
-	}
-
-	xfs_iunlock(ip, XFS_ILOCK_EXCL);
-	xfs_ireclaim(ip);
-	return 0;
-}
-
 void
 __xfs_inode_set_reclaim_tag(
 	struct xfs_perag	*pag,
@@ -766,19 +713,55 @@ __xfs_inode_clear_reclaim_tag(
 }
 
 STATIC int
-xfs_reclaim_inode_now(
+xfs_reclaim_inode(
 	struct xfs_inode	*ip,
 	struct xfs_perag	*pag,
-	int			flags)
+	int			sync_mode)
 {
-	/* ignore if already under reclaim */
-	if (xfs_iflags_test(ip, XFS_IRECLAIM)) {
-		read_unlock(&pag->pag_ici_lock);
+	/*
+	 * The radix tree lock here protects a thread in xfs_iget from racing
+	 * with us starting reclaim on the inode.  Once we have the
+	 * XFS_IRECLAIM flag set it will not touch us.
+	 */
+	spin_lock(&ip->i_flags_lock);
+	ASSERT_ALWAYS(__xfs_iflags_test(ip, XFS_IRECLAIMABLE));
+	if (__xfs_iflags_test(ip, XFS_IRECLAIM)) {
+		/* ignore as it is already under reclaim */
+		spin_unlock(&ip->i_flags_lock);
+		write_unlock(&pag->pag_ici_lock);
 		return 0;
 	}
-	read_unlock(&pag->pag_ici_lock);
+	__xfs_iflags_set(ip, XFS_IRECLAIM);
+	spin_unlock(&ip->i_flags_lock);
+	write_unlock(&pag->pag_ici_lock);
 
-	return xfs_reclaim_inode(ip, 0, flags);
+	/*
+	 * If the inode is still dirty, then flush it out.  If the inode
+	 * is not in the AIL, then it will be OK to flush it delwri as
+	 * long as xfs_iflush() does not keep any references to the inode.
+	 * We leave that decision up to xfs_iflush() since it has the
+	 * knowledge of whether it's OK to simply do a delwri flush of
+	 * the inode or whether we need to wait until the inode is
+	 * pulled from the AIL.
+	 * We get the flush lock regardless, though, just to make sure
+	 * we don't free it while it is being flushed.
+	 */
+	xfs_ilock(ip, XFS_ILOCK_EXCL);
+	xfs_iflock(ip);
+
+	/*
+	 * In the case of a forced shutdown we rely on xfs_iflush() to
+	 * wait for the inode to be unpinned before returning an error.
+	 */
+	if (!is_bad_inode(VFS_I(ip)) && xfs_iflush(ip, sync_mode) == 0) {
+		/* synchronize with xfs_iflush_done */
+		xfs_iflock(ip);
+		xfs_ifunlock(ip);
+	}
+
+	xfs_iunlock(ip, XFS_ILOCK_EXCL);
+	xfs_ireclaim(ip);
+	return 0;
 }
 
 int
@@ -786,6 +769,6 @@ xfs_reclaim_inodes(
 	xfs_mount_t	*mp,
 	int		mode)
 {
-	return xfs_inode_ag_iterator(mp, xfs_reclaim_inode_now, mode,
-					XFS_ICI_RECLAIM_TAG);
+	return xfs_inode_ag_iterator(mp, xfs_reclaim_inode, mode,
+					XFS_ICI_RECLAIM_TAG, 1);
 }
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index 27920eb7a820..ea932b43335d 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -44,7 +44,6 @@ void xfs_quiesce_attr(struct xfs_mount *mp);
 
 void xfs_flush_inodes(struct xfs_inode *ip);
 
-int xfs_reclaim_inode(struct xfs_inode *ip, int locked, int sync_mode);
 int xfs_reclaim_inodes(struct xfs_mount *mp, int mode);
 
 void xfs_inode_set_reclaim_tag(struct xfs_inode *ip);
@@ -55,6 +54,6 @@ void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag,
 int xfs_sync_inode_valid(struct xfs_inode *ip, struct xfs_perag *pag);
 int xfs_inode_ag_iterator(struct xfs_mount *mp,
 	int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags),
-	int flags, int tag);
+	int flags, int tag, int write_lock);
 
 #endif
diff --git a/fs/xfs/linux-2.6/xfs_trace.c b/fs/xfs/linux-2.6/xfs_trace.c
new file mode 100644
index 000000000000..856eb3c8d605
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_trace.c
@@ -0,0 +1,75 @@
+/*
+ * Copyright (c) 2009, Christoph Hellwig
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#include "xfs.h"
+#include "xfs_fs.h"
+#include "xfs_types.h"
+#include "xfs_bit.h"
+#include "xfs_log.h"
+#include "xfs_inum.h"
+#include "xfs_trans.h"
+#include "xfs_sb.h"
+#include "xfs_ag.h"
+#include "xfs_dir2.h"
+#include "xfs_da_btree.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_alloc_btree.h"
+#include "xfs_ialloc_btree.h"
+#include "xfs_dir2_sf.h"
+#include "xfs_attr_sf.h"
+#include "xfs_dinode.h"
+#include "xfs_inode.h"
+#include "xfs_btree.h"
+#include "xfs_dmapi.h"
+#include "xfs_mount.h"
+#include "xfs_ialloc.h"
+#include "xfs_itable.h"
+#include "xfs_alloc.h"
+#include "xfs_bmap.h"
+#include "xfs_attr.h"
+#include "xfs_attr_sf.h"
+#include "xfs_attr_leaf.h"
+#include "xfs_log_priv.h"
+#include "xfs_buf_item.h"
+#include "xfs_quota.h"
+#include "xfs_iomap.h"
+#include "xfs_aops.h"
+#include "quota/xfs_dquot_item.h"
+#include "quota/xfs_dquot.h"
+
+/*
+ * Format fsblock number into a static buffer & return it.
+ */
+STATIC char *xfs_fmtfsblock(xfs_fsblock_t bno)
+{
+	static char rval[50];
+
+	if (bno == NULLFSBLOCK)
+		sprintf(rval, "NULLFSBLOCK");
+	else if (isnullstartblock(bno))
+		sprintf(rval, "NULLSTARTBLOCK(%lld)", startblockval(bno));
+	else
+		sprintf(rval, "%lld", (xfs_dfsbno_t)bno);
+	return rval;
+}
+
+/*
+ * We include this last to have the helpers above available for the trace
+ * event implementations.
+ */
+#define CREATE_TRACE_POINTS
+#include "xfs_trace.h"
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
new file mode 100644
index 000000000000..c22a608321a3
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -0,0 +1,1422 @@
+/*
+ * Copyright (c) 2009, Christoph Hellwig
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM xfs
+
+#if !defined(_TRACE_XFS_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_XFS_H
+
+#include <linux/tracepoint.h>
+
+struct xfs_agf;
+struct xfs_alloc_arg;
+struct xfs_attr_list_context;
+struct xfs_buf_log_item;
+struct xfs_da_args;
+struct xfs_da_node_entry;
+struct xfs_dquot;
+struct xlog_ticket;
+struct log;
+
+DECLARE_EVENT_CLASS(xfs_attr_list_class,
+	TP_PROTO(struct xfs_attr_list_context *ctx),
+	TP_ARGS(ctx),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(u32, hashval)
+		__field(u32, blkno)
+		__field(u32, offset)
+		__field(void *, alist)
+		__field(int, bufsize)
+		__field(int, count)
+		__field(int, firstu)
+		__field(int, dupcnt)
+		__field(int, flags)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ctx->dp)->i_sb->s_dev;
+		__entry->ino = ctx->dp->i_ino;
+		__entry->hashval = ctx->cursor->hashval;
+		__entry->blkno = ctx->cursor->blkno;
+		__entry->offset = ctx->cursor->offset;
+		__entry->alist = ctx->alist;
+		__entry->bufsize = ctx->bufsize;
+		__entry->count = ctx->count;
+		__entry->firstu = ctx->firstu;
+		__entry->flags = ctx->flags;
+	),
+	TP_printk("dev %d:%d ino 0x%llx cursor h/b/o 0x%x/0x%x/%u dupcnt %u "
+		  "alist 0x%p size %u count %u firstu %u flags %d %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		   __entry->ino,
+		   __entry->hashval,
+		   __entry->blkno,
+		   __entry->offset,
+		   __entry->dupcnt,
+		   __entry->alist,
+		   __entry->bufsize,
+		   __entry->count,
+		   __entry->firstu,
+		   __entry->flags,
+		   __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS)
+	)
+)
+
+#define DEFINE_ATTR_LIST_EVENT(name) \
+DEFINE_EVENT(xfs_attr_list_class, name, \
+	TP_PROTO(struct xfs_attr_list_context *ctx), \
+	TP_ARGS(ctx))
+DEFINE_ATTR_LIST_EVENT(xfs_attr_list_sf);
+DEFINE_ATTR_LIST_EVENT(xfs_attr_list_sf_all);
+DEFINE_ATTR_LIST_EVENT(xfs_attr_list_leaf);
+DEFINE_ATTR_LIST_EVENT(xfs_attr_list_leaf_end);
+DEFINE_ATTR_LIST_EVENT(xfs_attr_list_full);
+DEFINE_ATTR_LIST_EVENT(xfs_attr_list_add);
+DEFINE_ATTR_LIST_EVENT(xfs_attr_list_wrong_blk);
+DEFINE_ATTR_LIST_EVENT(xfs_attr_list_notfound);
+
+TRACE_EVENT(xfs_attr_list_node_descend,
+	TP_PROTO(struct xfs_attr_list_context *ctx,
+		 struct xfs_da_node_entry *btree),
+	TP_ARGS(ctx, btree),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(u32, hashval)
+		__field(u32, blkno)
+		__field(u32, offset)
+		__field(void *, alist)
+		__field(int, bufsize)
+		__field(int, count)
+		__field(int, firstu)
+		__field(int, dupcnt)
+		__field(int, flags)
+		__field(u32, bt_hashval)
+		__field(u32, bt_before)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ctx->dp)->i_sb->s_dev;
+		__entry->ino = ctx->dp->i_ino;
+		__entry->hashval = ctx->cursor->hashval;
+		__entry->blkno = ctx->cursor->blkno;
+		__entry->offset = ctx->cursor->offset;
+		__entry->alist = ctx->alist;
+		__entry->bufsize = ctx->bufsize;
+		__entry->count = ctx->count;
+		__entry->firstu = ctx->firstu;
+		__entry->flags = ctx->flags;
+		__entry->bt_hashval = be32_to_cpu(btree->hashval);
+		__entry->bt_before = be32_to_cpu(btree->before);
+	),
+	TP_printk("dev %d:%d ino 0x%llx cursor h/b/o 0x%x/0x%x/%u dupcnt %u "
+		  "alist 0x%p size %u count %u firstu %u flags %d %s "
+		  "node hashval %u, node before %u",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		   __entry->ino,
+		   __entry->hashval,
+		   __entry->blkno,
+		   __entry->offset,
+		   __entry->dupcnt,
+		   __entry->alist,
+		   __entry->bufsize,
+		   __entry->count,
+		   __entry->firstu,
+		   __entry->flags,
+		   __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS),
+		   __entry->bt_hashval,
+		   __entry->bt_before)
+);
+
+TRACE_EVENT(xfs_iext_insert,
+	TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx,
+		 struct xfs_bmbt_irec *r, int state, unsigned long caller_ip),
+	TP_ARGS(ip, idx, r, state, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(xfs_extnum_t, idx)
+		__field(xfs_fileoff_t, startoff)
+		__field(xfs_fsblock_t, startblock)
+		__field(xfs_filblks_t, blockcount)
+		__field(xfs_exntst_t, state)
+		__field(int, bmap_state)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->idx = idx;
+		__entry->startoff = r->br_startoff;
+		__entry->startblock = r->br_startblock;
+		__entry->blockcount = r->br_blockcount;
+		__entry->state = r->br_state;
+		__entry->bmap_state = state;
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d ino 0x%llx state %s idx %ld "
+		  "offset %lld block %s count %lld flag %d caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS),
+		  (long)__entry->idx,
+		  __entry->startoff,
+		  xfs_fmtfsblock(__entry->startblock),
+		  __entry->blockcount,
+		  __entry->state,
+		  (char *)__entry->caller_ip)
+);
+
+DECLARE_EVENT_CLASS(xfs_bmap_class,
+	TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, int state,
+		 unsigned long caller_ip),
+	TP_ARGS(ip, idx, state, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(xfs_extnum_t, idx)
+		__field(xfs_fileoff_t, startoff)
+		__field(xfs_fsblock_t, startblock)
+		__field(xfs_filblks_t, blockcount)
+		__field(xfs_exntst_t, state)
+		__field(int, bmap_state)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		struct xfs_ifork	*ifp = (state & BMAP_ATTRFORK) ?
+						ip->i_afp : &ip->i_df;
+		struct xfs_bmbt_irec	r;
+
+		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx), &r);
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->idx = idx;
+		__entry->startoff = r.br_startoff;
+		__entry->startblock = r.br_startblock;
+		__entry->blockcount = r.br_blockcount;
+		__entry->state = r.br_state;
+		__entry->bmap_state = state;
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d ino 0x%llx state %s idx %ld "
+		  "offset %lld block %s count %lld flag %d caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS),
+		  (long)__entry->idx,
+		  __entry->startoff,
+		  xfs_fmtfsblock(__entry->startblock),
+		  __entry->blockcount,
+		  __entry->state,
+		  (char *)__entry->caller_ip)
+)
+
+#define DEFINE_BMAP_EVENT(name) \
+DEFINE_EVENT(xfs_bmap_class, name, \
+	TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, int state, \
+		 unsigned long caller_ip), \
+	TP_ARGS(ip, idx, state, caller_ip))
+DEFINE_BMAP_EVENT(xfs_iext_remove);
+DEFINE_BMAP_EVENT(xfs_bmap_pre_update);
+DEFINE_BMAP_EVENT(xfs_bmap_post_update);
+DEFINE_BMAP_EVENT(xfs_extlist);
+
+DECLARE_EVENT_CLASS(xfs_buf_class,
+	TP_PROTO(struct xfs_buf *bp, unsigned long caller_ip),
+	TP_ARGS(bp, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_daddr_t, bno)
+		__field(size_t, buffer_length)
+		__field(int, hold)
+		__field(int, pincount)
+		__field(unsigned, lockval)
+		__field(unsigned, flags)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		__entry->dev = bp->b_target->bt_dev;
+		__entry->bno = bp->b_bn;
+		__entry->buffer_length = bp->b_buffer_length;
+		__entry->hold = atomic_read(&bp->b_hold);
+		__entry->pincount = atomic_read(&bp->b_pin_count);
+		__entry->lockval = xfs_buf_lock_value(bp);
+		__entry->flags = bp->b_flags;
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d "
+		  "lock %d flags %s caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  (unsigned long long)__entry->bno,
+		  __entry->buffer_length,
+		  __entry->hold,
+		  __entry->pincount,
+		  __entry->lockval,
+		  __print_flags(__entry->flags, "|", XFS_BUF_FLAGS),
+		  (void *)__entry->caller_ip)
+)
+
+#define DEFINE_BUF_EVENT(name) \
+DEFINE_EVENT(xfs_buf_class, name, \
+	TP_PROTO(struct xfs_buf *bp, unsigned long caller_ip), \
+	TP_ARGS(bp, caller_ip))
+DEFINE_BUF_EVENT(xfs_buf_init);
+DEFINE_BUF_EVENT(xfs_buf_free);
+DEFINE_BUF_EVENT(xfs_buf_hold);
+DEFINE_BUF_EVENT(xfs_buf_rele);
+DEFINE_BUF_EVENT(xfs_buf_pin);
+DEFINE_BUF_EVENT(xfs_buf_unpin);
+DEFINE_BUF_EVENT(xfs_buf_iodone);
+DEFINE_BUF_EVENT(xfs_buf_iorequest);
+DEFINE_BUF_EVENT(xfs_buf_bawrite);
+DEFINE_BUF_EVENT(xfs_buf_bdwrite);
+DEFINE_BUF_EVENT(xfs_buf_lock);
+DEFINE_BUF_EVENT(xfs_buf_lock_done);
+DEFINE_BUF_EVENT(xfs_buf_cond_lock);
+DEFINE_BUF_EVENT(xfs_buf_unlock);
+DEFINE_BUF_EVENT(xfs_buf_ordered_retry);
+DEFINE_BUF_EVENT(xfs_buf_iowait);
+DEFINE_BUF_EVENT(xfs_buf_iowait_done);
+DEFINE_BUF_EVENT(xfs_buf_delwri_queue);
+DEFINE_BUF_EVENT(xfs_buf_delwri_dequeue);
+DEFINE_BUF_EVENT(xfs_buf_delwri_split);
+DEFINE_BUF_EVENT(xfs_buf_get_noaddr);
+DEFINE_BUF_EVENT(xfs_bdstrat_shut);
+DEFINE_BUF_EVENT(xfs_buf_item_relse);
+DEFINE_BUF_EVENT(xfs_buf_item_iodone);
+DEFINE_BUF_EVENT(xfs_buf_item_iodone_async);
+DEFINE_BUF_EVENT(xfs_buf_error_relse);
+DEFINE_BUF_EVENT(xfs_trans_read_buf_io);
+DEFINE_BUF_EVENT(xfs_trans_read_buf_shut);
+
+/* not really buffer traces, but the buf provides useful information */
+DEFINE_BUF_EVENT(xfs_btree_corrupt);
+DEFINE_BUF_EVENT(xfs_da_btree_corrupt);
+DEFINE_BUF_EVENT(xfs_reset_dqcounts);
+DEFINE_BUF_EVENT(xfs_inode_item_push);
+
+/* pass flags explicitly */
+DECLARE_EVENT_CLASS(xfs_buf_flags_class,
+	TP_PROTO(struct xfs_buf *bp, unsigned flags, unsigned long caller_ip),
+	TP_ARGS(bp, flags, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_daddr_t, bno)
+		__field(size_t, buffer_length)
+		__field(int, hold)
+		__field(int, pincount)
+		__field(unsigned, lockval)
+		__field(unsigned, flags)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		__entry->dev = bp->b_target->bt_dev;
+		__entry->bno = bp->b_bn;
+		__entry->buffer_length = bp->b_buffer_length;
+		__entry->flags = flags;
+		__entry->hold = atomic_read(&bp->b_hold);
+		__entry->pincount = atomic_read(&bp->b_pin_count);
+		__entry->lockval = xfs_buf_lock_value(bp);
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d "
+		  "lock %d flags %s caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  (unsigned long long)__entry->bno,
+		  __entry->buffer_length,
+		  __entry->hold,
+		  __entry->pincount,
+		  __entry->lockval,
+		  __print_flags(__entry->flags, "|", XFS_BUF_FLAGS),
+		  (void *)__entry->caller_ip)
+)
+
+#define DEFINE_BUF_FLAGS_EVENT(name) \
+DEFINE_EVENT(xfs_buf_flags_class, name, \
+	TP_PROTO(struct xfs_buf *bp, unsigned flags, unsigned long caller_ip), \
+	TP_ARGS(bp, flags, caller_ip))
+DEFINE_BUF_FLAGS_EVENT(xfs_buf_find);
+DEFINE_BUF_FLAGS_EVENT(xfs_buf_get);
+DEFINE_BUF_FLAGS_EVENT(xfs_buf_read);
+
+TRACE_EVENT(xfs_buf_ioerror,
+	TP_PROTO(struct xfs_buf *bp, int error, unsigned long caller_ip),
+	TP_ARGS(bp, error, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_daddr_t, bno)
+		__field(size_t, buffer_length)
+		__field(unsigned, flags)
+		__field(int, hold)
+		__field(int, pincount)
+		__field(unsigned, lockval)
+		__field(int, error)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		__entry->dev = bp->b_target->bt_dev;
+		__entry->bno = bp->b_bn;
+		__entry->buffer_length = bp->b_buffer_length;
+		__entry->hold = atomic_read(&bp->b_hold);
+		__entry->pincount = atomic_read(&bp->b_pin_count);
+		__entry->lockval = xfs_buf_lock_value(bp);
+		__entry->error = error;
+		__entry->flags = bp->b_flags;
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d "
+		  "lock %d error %d flags %s caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  (unsigned long long)__entry->bno,
+		  __entry->buffer_length,
+		  __entry->hold,
+		  __entry->pincount,
+		  __entry->lockval,
+		  __entry->error,
+		  __print_flags(__entry->flags, "|", XFS_BUF_FLAGS),
+		  (void *)__entry->caller_ip)
+);
+
+DECLARE_EVENT_CLASS(xfs_buf_item_class,
+	TP_PROTO(struct xfs_buf_log_item *bip),
+	TP_ARGS(bip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_daddr_t, buf_bno)
+		__field(size_t, buf_len)
+		__field(int, buf_hold)
+		__field(int, buf_pincount)
+		__field(int, buf_lockval)
+		__field(unsigned, buf_flags)
+		__field(unsigned, bli_recur)
+		__field(int, bli_refcount)
+		__field(unsigned, bli_flags)
+		__field(void *, li_desc)
+		__field(unsigned, li_flags)
+	),
+	TP_fast_assign(
+		__entry->dev = bip->bli_buf->b_target->bt_dev;
+		__entry->bli_flags = bip->bli_flags;
+		__entry->bli_recur = bip->bli_recur;
+		__entry->bli_refcount = atomic_read(&bip->bli_refcount);
+		__entry->buf_bno = bip->bli_buf->b_bn;
+		__entry->buf_len = bip->bli_buf->b_buffer_length;
+		__entry->buf_flags = bip->bli_buf->b_flags;
+		__entry->buf_hold = atomic_read(&bip->bli_buf->b_hold);
+		__entry->buf_pincount = atomic_read(&bip->bli_buf->b_pin_count);
+		__entry->buf_lockval = xfs_buf_lock_value(bip->bli_buf);
+		__entry->li_desc = bip->bli_item.li_desc;
+		__entry->li_flags = bip->bli_item.li_flags;
+	),
+	TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d "
+		  "lock %d flags %s recur %d refcount %d bliflags %s "
+		  "lidesc 0x%p liflags %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  (unsigned long long)__entry->buf_bno,
+		  __entry->buf_len,
+		  __entry->buf_hold,
+		  __entry->buf_pincount,
+		  __entry->buf_lockval,
+		  __print_flags(__entry->buf_flags, "|", XFS_BUF_FLAGS),
+		  __entry->bli_recur,
+		  __entry->bli_refcount,
+		  __print_flags(__entry->bli_flags, "|", XFS_BLI_FLAGS),
+		  __entry->li_desc,
+		  __print_flags(__entry->li_flags, "|", XFS_LI_FLAGS))
+)
+
+#define DEFINE_BUF_ITEM_EVENT(name) \
+DEFINE_EVENT(xfs_buf_item_class, name, \
+	TP_PROTO(struct xfs_buf_log_item *bip), \
+	TP_ARGS(bip))
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_size);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_size_stale);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_format);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_format_stale);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_pin);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unpin);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unpin_stale);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_trylock);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock_stale);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_committed);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_push);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf_recur);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_getsb);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_getsb_recur);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_read_buf);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_read_buf_recur);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_log_buf);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_brelse);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_bjoin);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold_release);
+DEFINE_BUF_ITEM_EVENT(xfs_trans_binval);
+
+DECLARE_EVENT_CLASS(xfs_lock_class,
+	TP_PROTO(struct xfs_inode *ip, unsigned lock_flags,
+		 unsigned long caller_ip),
+	TP_ARGS(ip,  lock_flags, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(int, lock_flags)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->lock_flags = lock_flags;
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d ino 0x%llx flags %s caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __print_flags(__entry->lock_flags, "|", XFS_LOCK_FLAGS),
+		  (void *)__entry->caller_ip)
+)
+
+#define DEFINE_LOCK_EVENT(name) \
+DEFINE_EVENT(xfs_lock_class, name, \
+	TP_PROTO(struct xfs_inode *ip, unsigned lock_flags, \
+		 unsigned long caller_ip), \
+	TP_ARGS(ip,  lock_flags, caller_ip))
+DEFINE_LOCK_EVENT(xfs_ilock);
+DEFINE_LOCK_EVENT(xfs_ilock_nowait);
+DEFINE_LOCK_EVENT(xfs_ilock_demote);
+DEFINE_LOCK_EVENT(xfs_iunlock);
+
+DECLARE_EVENT_CLASS(xfs_iget_class,
+	TP_PROTO(struct xfs_inode *ip),
+	TP_ARGS(ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+	),
+	TP_printk("dev %d:%d ino 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino)
+)
+
+#define DEFINE_IGET_EVENT(name) \
+DEFINE_EVENT(xfs_iget_class, name, \
+	TP_PROTO(struct xfs_inode *ip), \
+	TP_ARGS(ip))
+DEFINE_IGET_EVENT(xfs_iget_skip);
+DEFINE_IGET_EVENT(xfs_iget_reclaim);
+DEFINE_IGET_EVENT(xfs_iget_found);
+DEFINE_IGET_EVENT(xfs_iget_alloc);
+
+DECLARE_EVENT_CLASS(xfs_inode_class,
+	TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip),
+	TP_ARGS(ip, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(int, count)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->count = atomic_read(&VFS_I(ip)->i_count);
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d ino 0x%llx count %d caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->count,
+		  (char *)__entry->caller_ip)
+)
+
+#define DEFINE_INODE_EVENT(name) \
+DEFINE_EVENT(xfs_inode_class, name, \
+	TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip), \
+	TP_ARGS(ip, caller_ip))
+DEFINE_INODE_EVENT(xfs_ihold);
+DEFINE_INODE_EVENT(xfs_irele);
+/* the old xfs_itrace_entry tracer - to be replaced by s.th. in the VFS */
+DEFINE_INODE_EVENT(xfs_inode);
+#define xfs_itrace_entry(ip)    \
+	trace_xfs_inode(ip, _THIS_IP_)
+
+DECLARE_EVENT_CLASS(xfs_dquot_class,
+	TP_PROTO(struct xfs_dquot *dqp),
+	TP_ARGS(dqp),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(__be32, id)
+		__field(unsigned, flags)
+		__field(unsigned, nrefs)
+		__field(unsigned long long, res_bcount)
+		__field(unsigned long long, bcount)
+		__field(unsigned long long, icount)
+		__field(unsigned long long, blk_hardlimit)
+		__field(unsigned long long, blk_softlimit)
+		__field(unsigned long long, ino_hardlimit)
+		__field(unsigned long long, ino_softlimit)
+	), \
+	TP_fast_assign(
+		__entry->dev = dqp->q_mount->m_super->s_dev;
+		__entry->id = dqp->q_core.d_id;
+		__entry->flags = dqp->dq_flags;
+		__entry->nrefs = dqp->q_nrefs;
+		__entry->res_bcount = dqp->q_res_bcount;
+		__entry->bcount = be64_to_cpu(dqp->q_core.d_bcount);
+		__entry->icount = be64_to_cpu(dqp->q_core.d_icount);
+		__entry->blk_hardlimit =
+			be64_to_cpu(dqp->q_core.d_blk_hardlimit);
+		__entry->blk_softlimit =
+			be64_to_cpu(dqp->q_core.d_blk_softlimit);
+		__entry->ino_hardlimit =
+			be64_to_cpu(dqp->q_core.d_ino_hardlimit);
+		__entry->ino_softlimit =
+			be64_to_cpu(dqp->q_core.d_ino_softlimit);
+	),
+	TP_printk("dev %d:%d id 0x%x flags %s nrefs %u res_bc 0x%llx "
+		  "bcnt 0x%llx [hard 0x%llx | soft 0x%llx] "
+		  "icnt 0x%llx [hard 0x%llx | soft 0x%llx]",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  be32_to_cpu(__entry->id),
+		  __print_flags(__entry->flags, "|", XFS_DQ_FLAGS),
+		  __entry->nrefs,
+		  __entry->res_bcount,
+		  __entry->bcount,
+		  __entry->blk_hardlimit,
+		  __entry->blk_softlimit,
+		  __entry->icount,
+		  __entry->ino_hardlimit,
+		  __entry->ino_softlimit)
+)
+
+#define DEFINE_DQUOT_EVENT(name) \
+DEFINE_EVENT(xfs_dquot_class, name, \
+	TP_PROTO(struct xfs_dquot *dqp), \
+	TP_ARGS(dqp))
+DEFINE_DQUOT_EVENT(xfs_dqadjust);
+DEFINE_DQUOT_EVENT(xfs_dqshake_dirty);
+DEFINE_DQUOT_EVENT(xfs_dqshake_unlink);
+DEFINE_DQUOT_EVENT(xfs_dqreclaim_want);
+DEFINE_DQUOT_EVENT(xfs_dqreclaim_dirty);
+DEFINE_DQUOT_EVENT(xfs_dqreclaim_unlink);
+DEFINE_DQUOT_EVENT(xfs_dqattach_found);
+DEFINE_DQUOT_EVENT(xfs_dqattach_get);
+DEFINE_DQUOT_EVENT(xfs_dqinit);
+DEFINE_DQUOT_EVENT(xfs_dqreuse);
+DEFINE_DQUOT_EVENT(xfs_dqalloc);
+DEFINE_DQUOT_EVENT(xfs_dqtobp_read);
+DEFINE_DQUOT_EVENT(xfs_dqread);
+DEFINE_DQUOT_EVENT(xfs_dqread_fail);
+DEFINE_DQUOT_EVENT(xfs_dqlookup_found);
+DEFINE_DQUOT_EVENT(xfs_dqlookup_want);
+DEFINE_DQUOT_EVENT(xfs_dqlookup_freelist);
+DEFINE_DQUOT_EVENT(xfs_dqlookup_move);
+DEFINE_DQUOT_EVENT(xfs_dqlookup_done);
+DEFINE_DQUOT_EVENT(xfs_dqget_hit);
+DEFINE_DQUOT_EVENT(xfs_dqget_miss);
+DEFINE_DQUOT_EVENT(xfs_dqput);
+DEFINE_DQUOT_EVENT(xfs_dqput_wait);
+DEFINE_DQUOT_EVENT(xfs_dqput_free);
+DEFINE_DQUOT_EVENT(xfs_dqrele);
+DEFINE_DQUOT_EVENT(xfs_dqflush);
+DEFINE_DQUOT_EVENT(xfs_dqflush_force);
+DEFINE_DQUOT_EVENT(xfs_dqflush_done);
+/* not really iget events, but we re-use the format */
+DEFINE_IGET_EVENT(xfs_dquot_dqalloc);
+DEFINE_IGET_EVENT(xfs_dquot_dqdetach);
+
+DECLARE_EVENT_CLASS(xfs_loggrant_class,
+	TP_PROTO(struct log *log, struct xlog_ticket *tic),
+	TP_ARGS(log, tic),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(unsigned, trans_type)
+		__field(char, ocnt)
+		__field(char, cnt)
+		__field(int, curr_res)
+		__field(int, unit_res)
+		__field(unsigned int, flags)
+		__field(void *, reserve_headq)
+		__field(void *, write_headq)
+		__field(int, grant_reserve_cycle)
+		__field(int, grant_reserve_bytes)
+		__field(int, grant_write_cycle)
+		__field(int, grant_write_bytes)
+		__field(int, curr_cycle)
+		__field(int, curr_block)
+		__field(xfs_lsn_t, tail_lsn)
+	),
+	TP_fast_assign(
+		__entry->dev = log->l_mp->m_super->s_dev;
+		__entry->trans_type = tic->t_trans_type;
+		__entry->ocnt = tic->t_ocnt;
+		__entry->cnt = tic->t_cnt;
+		__entry->curr_res = tic->t_curr_res;
+		__entry->unit_res = tic->t_unit_res;
+		__entry->flags = tic->t_flags;
+		__entry->reserve_headq = log->l_reserve_headq;
+		__entry->write_headq = log->l_write_headq;
+		__entry->grant_reserve_cycle = log->l_grant_reserve_cycle;
+		__entry->grant_reserve_bytes = log->l_grant_reserve_bytes;
+		__entry->grant_write_cycle = log->l_grant_write_cycle;
+		__entry->grant_write_bytes = log->l_grant_write_bytes;
+		__entry->curr_cycle = log->l_curr_cycle;
+		__entry->curr_block = log->l_curr_block;
+		__entry->tail_lsn = log->l_tail_lsn;
+	),
+	TP_printk("dev %d:%d type %s t_ocnt %u t_cnt %u t_curr_res %u "
+		  "t_unit_res %u t_flags %s reserve_headq 0x%p "
+		  "write_headq 0x%p grant_reserve_cycle %d "
+		  "grant_reserve_bytes %d grant_write_cycle %d "
+		  "grant_write_bytes %d curr_cycle %d curr_block %d "
+		  "tail_cycle %d tail_block %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __print_symbolic(__entry->trans_type, XFS_TRANS_TYPES),
+		  __entry->ocnt,
+		  __entry->cnt,
+		  __entry->curr_res,
+		  __entry->unit_res,
+		  __print_flags(__entry->flags, "|", XLOG_TIC_FLAGS),
+		  __entry->reserve_headq,
+		  __entry->write_headq,
+		  __entry->grant_reserve_cycle,
+		  __entry->grant_reserve_bytes,
+		  __entry->grant_write_cycle,
+		  __entry->grant_write_bytes,
+		  __entry->curr_cycle,
+		  __entry->curr_block,
+		  CYCLE_LSN(__entry->tail_lsn),
+		  BLOCK_LSN(__entry->tail_lsn)
+	)
+)
+
+#define DEFINE_LOGGRANT_EVENT(name) \
+DEFINE_EVENT(xfs_loggrant_class, name, \
+	TP_PROTO(struct log *log, struct xlog_ticket *tic), \
+	TP_ARGS(log, tic))
+DEFINE_LOGGRANT_EVENT(xfs_log_done_nonperm);
+DEFINE_LOGGRANT_EVENT(xfs_log_done_perm);
+DEFINE_LOGGRANT_EVENT(xfs_log_reserve);
+DEFINE_LOGGRANT_EVENT(xfs_log_umount_write);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_enter);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_exit);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_error);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep1);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake1);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep2);
+DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake2);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_enter);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_exit);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_error);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep1);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake1);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep2);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake2);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_enter);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_exit);
+DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_sub);
+DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_enter);
+DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_exit);
+DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_sub);
+
+#define DEFINE_RW_EVENT(name) \
+TRACE_EVENT(name, \
+	TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags), \
+	TP_ARGS(ip, count, offset, flags), \
+	TP_STRUCT__entry( \
+		__field(dev_t, dev) \
+		__field(xfs_ino_t, ino) \
+		__field(xfs_fsize_t, size) \
+		__field(xfs_fsize_t, new_size) \
+		__field(loff_t, offset) \
+		__field(size_t, count) \
+		__field(int, flags) \
+	), \
+	TP_fast_assign( \
+		__entry->dev = VFS_I(ip)->i_sb->s_dev; \
+		__entry->ino = ip->i_ino; \
+		__entry->size = ip->i_d.di_size; \
+		__entry->new_size = ip->i_new_size; \
+		__entry->offset = offset; \
+		__entry->count = count; \
+		__entry->flags = flags; \
+	), \
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " \
+		  "offset 0x%llx count 0x%zx ioflags %s", \
+		  MAJOR(__entry->dev), MINOR(__entry->dev), \
+		  __entry->ino, \
+		  __entry->size, \
+		  __entry->new_size, \
+		  __entry->offset, \
+		  __entry->count, \
+		  __print_flags(__entry->flags, "|", XFS_IO_FLAGS)) \
+)
+DEFINE_RW_EVENT(xfs_file_read);
+DEFINE_RW_EVENT(xfs_file_buffered_write);
+DEFINE_RW_EVENT(xfs_file_direct_write);
+DEFINE_RW_EVENT(xfs_file_splice_read);
+DEFINE_RW_EVENT(xfs_file_splice_write);
+
+
+#define DEFINE_PAGE_EVENT(name) \
+TRACE_EVENT(name, \
+	TP_PROTO(struct inode *inode, struct page *page, unsigned long off), \
+	TP_ARGS(inode, page, off), \
+	TP_STRUCT__entry( \
+		__field(dev_t, dev) \
+		__field(xfs_ino_t, ino) \
+		__field(pgoff_t, pgoff) \
+		__field(loff_t, size) \
+		__field(unsigned long, offset) \
+		__field(int, delalloc) \
+		__field(int, unmapped) \
+		__field(int, unwritten) \
+	), \
+	TP_fast_assign( \
+		int delalloc = -1, unmapped = -1, unwritten = -1; \
+	\
+		if (page_has_buffers(page)) \
+			xfs_count_page_state(page, &delalloc, \
+					     &unmapped, &unwritten); \
+		__entry->dev = inode->i_sb->s_dev; \
+		__entry->ino = XFS_I(inode)->i_ino; \
+		__entry->pgoff = page_offset(page); \
+		__entry->size = i_size_read(inode); \
+		__entry->offset = off; \
+		__entry->delalloc = delalloc; \
+		__entry->unmapped = unmapped; \
+		__entry->unwritten = unwritten; \
+	), \
+	TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx " \
+		  "delalloc %d unmapped %d unwritten %d", \
+		  MAJOR(__entry->dev), MINOR(__entry->dev), \
+		  __entry->ino, \
+		  __entry->pgoff, \
+		  __entry->size, \
+		  __entry->offset, \
+		  __entry->delalloc, \
+		  __entry->unmapped, \
+		  __entry->unwritten) \
+)
+DEFINE_PAGE_EVENT(xfs_writepage);
+DEFINE_PAGE_EVENT(xfs_releasepage);
+DEFINE_PAGE_EVENT(xfs_invalidatepage);
+
+#define DEFINE_IOMAP_EVENT(name) \
+TRACE_EVENT(name, \
+	TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, \
+		 int flags, struct xfs_bmbt_irec *irec), \
+	TP_ARGS(ip, offset, count, flags, irec), \
+	TP_STRUCT__entry( \
+		__field(dev_t, dev) \
+		__field(xfs_ino_t, ino) \
+		__field(loff_t, size) \
+		__field(loff_t, new_size) \
+		__field(loff_t, offset) \
+		__field(size_t, count) \
+		__field(int, flags) \
+		__field(xfs_fileoff_t, startoff) \
+		__field(xfs_fsblock_t, startblock) \
+		__field(xfs_filblks_t, blockcount) \
+	), \
+	TP_fast_assign( \
+		__entry->dev = VFS_I(ip)->i_sb->s_dev; \
+		__entry->ino = ip->i_ino; \
+		__entry->size = ip->i_d.di_size; \
+		__entry->new_size = ip->i_new_size; \
+		__entry->offset = offset; \
+		__entry->count = count; \
+		__entry->flags = flags; \
+		__entry->startoff = irec ? irec->br_startoff : 0; \
+		__entry->startblock = irec ? irec->br_startblock : 0; \
+		__entry->blockcount = irec ? irec->br_blockcount : 0; \
+	), \
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " \
+		  "offset 0x%llx count %zd flags %s " \
+		  "startoff 0x%llx startblock %s blockcount 0x%llx", \
+		  MAJOR(__entry->dev), MINOR(__entry->dev), \
+		  __entry->ino, \
+		  __entry->size, \
+		  __entry->new_size, \
+		  __entry->offset, \
+		  __entry->count, \
+		  __print_flags(__entry->flags, "|", BMAPI_FLAGS), \
+		  __entry->startoff, \
+		  xfs_fmtfsblock(__entry->startblock), \
+		  __entry->blockcount) \
+)
+DEFINE_IOMAP_EVENT(xfs_iomap_enter);
+DEFINE_IOMAP_EVENT(xfs_iomap_found);
+DEFINE_IOMAP_EVENT(xfs_iomap_alloc);
+
+#define DEFINE_SIMPLE_IO_EVENT(name) \
+TRACE_EVENT(name, \
+	TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count), \
+	TP_ARGS(ip, offset, count), \
+	TP_STRUCT__entry( \
+		__field(dev_t, dev) \
+		__field(xfs_ino_t, ino) \
+		__field(loff_t, size) \
+		__field(loff_t, new_size) \
+		__field(loff_t, offset) \
+		__field(size_t, count) \
+	), \
+	TP_fast_assign( \
+		__entry->dev = VFS_I(ip)->i_sb->s_dev; \
+		__entry->ino = ip->i_ino; \
+		__entry->size = ip->i_d.di_size; \
+		__entry->new_size = ip->i_new_size; \
+		__entry->offset = offset; \
+		__entry->count = count; \
+	), \
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " \
+		  "offset 0x%llx count %zd", \
+		  MAJOR(__entry->dev), MINOR(__entry->dev), \
+		  __entry->ino, \
+		  __entry->size, \
+		  __entry->new_size, \
+		  __entry->offset, \
+		  __entry->count) \
+);
+DEFINE_SIMPLE_IO_EVENT(xfs_delalloc_enospc);
+DEFINE_SIMPLE_IO_EVENT(xfs_unwritten_convert);
+
+
+TRACE_EVENT(xfs_itruncate_start,
+	TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size, int flag,
+		 xfs_off_t toss_start, xfs_off_t toss_finish),
+	TP_ARGS(ip, new_size, flag, toss_start, toss_finish),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(xfs_fsize_t, size)
+		__field(xfs_fsize_t, new_size)
+		__field(xfs_off_t, toss_start)
+		__field(xfs_off_t, toss_finish)
+		__field(int, flag)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->size = ip->i_d.di_size;
+		__entry->new_size = new_size;
+		__entry->toss_start = toss_start;
+		__entry->toss_finish = toss_finish;
+		__entry->flag = flag;
+	),
+	TP_printk("dev %d:%d ino 0x%llx %s size 0x%llx new_size 0x%llx "
+		  "toss start 0x%llx toss finish 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __print_flags(__entry->flag, "|", XFS_ITRUNC_FLAGS),
+		  __entry->size,
+		  __entry->new_size,
+		  __entry->toss_start,
+		  __entry->toss_finish)
+);
+
+DECLARE_EVENT_CLASS(xfs_itrunc_class,
+	TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size),
+	TP_ARGS(ip, new_size),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(xfs_fsize_t, size)
+		__field(xfs_fsize_t, new_size)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->size = ip->i_d.di_size;
+		__entry->new_size = new_size;
+	),
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->size,
+		  __entry->new_size)
+)
+
+#define DEFINE_ITRUNC_EVENT(name) \
+DEFINE_EVENT(xfs_itrunc_class, name, \
+	TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size), \
+	TP_ARGS(ip, new_size))
+DEFINE_ITRUNC_EVENT(xfs_itruncate_finish_start);
+DEFINE_ITRUNC_EVENT(xfs_itruncate_finish_end);
+
+TRACE_EVENT(xfs_pagecache_inval,
+	TP_PROTO(struct xfs_inode *ip, xfs_off_t start, xfs_off_t finish),
+	TP_ARGS(ip, start, finish),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(xfs_fsize_t, size)
+		__field(xfs_off_t, start)
+		__field(xfs_off_t, finish)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->size = ip->i_d.di_size;
+		__entry->start = start;
+		__entry->finish = finish;
+	),
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx start 0x%llx finish 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->size,
+		  __entry->start,
+		  __entry->finish)
+);
+
+TRACE_EVENT(xfs_bunmap,
+	TP_PROTO(struct xfs_inode *ip, xfs_fileoff_t bno, xfs_filblks_t len,
+		 int flags, unsigned long caller_ip),
+	TP_ARGS(ip, bno, len, flags, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(xfs_fsize_t, size)
+		__field(xfs_fileoff_t, bno)
+		__field(xfs_filblks_t, len)
+		__field(unsigned long, caller_ip)
+		__field(int, flags)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(ip)->i_sb->s_dev;
+		__entry->ino = ip->i_ino;
+		__entry->size = ip->i_d.di_size;
+		__entry->bno = bno;
+		__entry->len = len;
+		__entry->caller_ip = caller_ip;
+		__entry->flags = flags;
+	),
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx bno 0x%llx len 0x%llx"
+		  "flags %s caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->size,
+		  __entry->bno,
+		  __entry->len,
+		  __print_flags(__entry->flags, "|", XFS_BMAPI_FLAGS),
+		  (void *)__entry->caller_ip)
+
+);
+
+TRACE_EVENT(xfs_alloc_busy,
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
+		 xfs_extlen_t len, int slot),
+	TP_ARGS(mp, agno, agbno, len, slot),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, agbno)
+		__field(xfs_extlen_t, len)
+		__field(int, slot)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->agno = agno;
+		__entry->agbno = agbno;
+		__entry->len = len;
+		__entry->slot = slot;
+	),
+	TP_printk("dev %d:%d agno %u agbno %u len %u slot %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno,
+		  __entry->agbno,
+		  __entry->len,
+		  __entry->slot)
+
+);
+
+#define XFS_BUSY_STATES \
+	{ 0,	"found" }, \
+	{ 1,	"missing" }
+
+TRACE_EVENT(xfs_alloc_unbusy,
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
+		 int slot, int found),
+	TP_ARGS(mp, agno, slot, found),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+		__field(int, slot)
+		__field(int, found)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->agno = agno;
+		__entry->slot = slot;
+		__entry->found = found;
+	),
+	TP_printk("dev %d:%d agno %u slot %d %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno,
+		  __entry->slot,
+		  __print_symbolic(__entry->found, XFS_BUSY_STATES))
+);
+
+TRACE_EVENT(xfs_alloc_busysearch,
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
+		 xfs_extlen_t len, xfs_lsn_t lsn),
+	TP_ARGS(mp, agno, agbno, len, lsn),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, agbno)
+		__field(xfs_extlen_t, len)
+		__field(xfs_lsn_t, lsn)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->agno = agno;
+		__entry->agbno = agbno;
+		__entry->len = len;
+		__entry->lsn = lsn;
+	),
+	TP_printk("dev %d:%d agno %u agbno %u len %u force lsn 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno,
+		  __entry->agbno,
+		  __entry->len,
+		  __entry->lsn)
+);
+
+TRACE_EVENT(xfs_agf,
+	TP_PROTO(struct xfs_mount *mp, struct xfs_agf *agf, int flags,
+		 unsigned long caller_ip),
+	TP_ARGS(mp, agf, flags, caller_ip),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+		__field(int, flags)
+		__field(__u32, length)
+		__field(__u32, bno_root)
+		__field(__u32, cnt_root)
+		__field(__u32, bno_level)
+		__field(__u32, cnt_level)
+		__field(__u32, flfirst)
+		__field(__u32, fllast)
+		__field(__u32, flcount)
+		__field(__u32, freeblks)
+		__field(__u32, longest)
+		__field(unsigned long, caller_ip)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->agno = be32_to_cpu(agf->agf_seqno),
+		__entry->flags = flags;
+		__entry->length = be32_to_cpu(agf->agf_length),
+		__entry->bno_root = be32_to_cpu(agf->agf_roots[XFS_BTNUM_BNO]),
+		__entry->cnt_root = be32_to_cpu(agf->agf_roots[XFS_BTNUM_CNT]),
+		__entry->bno_level =
+				be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]),
+		__entry->cnt_level =
+				be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]),
+		__entry->flfirst = be32_to_cpu(agf->agf_flfirst),
+		__entry->fllast = be32_to_cpu(agf->agf_fllast),
+		__entry->flcount = be32_to_cpu(agf->agf_flcount),
+		__entry->freeblks = be32_to_cpu(agf->agf_freeblks),
+		__entry->longest = be32_to_cpu(agf->agf_longest);
+		__entry->caller_ip = caller_ip;
+	),
+	TP_printk("dev %d:%d agno %u flags %s length %u roots b %u c %u "
+		  "levels b %u c %u flfirst %u fllast %u flcount %u "
+		  "freeblks %u longest %u caller %pf",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno,
+		  __print_flags(__entry->flags, "|", XFS_AGF_FLAGS),
+		  __entry->length,
+		  __entry->bno_root,
+		  __entry->cnt_root,
+		  __entry->bno_level,
+		  __entry->cnt_level,
+		  __entry->flfirst,
+		  __entry->fllast,
+		  __entry->flcount,
+		  __entry->freeblks,
+		  __entry->longest,
+		  (void *)__entry->caller_ip)
+);
+
+TRACE_EVENT(xfs_free_extent,
+	TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
+		 xfs_extlen_t len, bool isfl, int haveleft, int haveright),
+	TP_ARGS(mp, agno, agbno, len, isfl, haveleft, haveright),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, agbno)
+		__field(xfs_extlen_t, len)
+		__field(int, isfl)
+		__field(int, haveleft)
+		__field(int, haveright)
+	),
+	TP_fast_assign(
+		__entry->dev = mp->m_super->s_dev;
+		__entry->agno = agno;
+		__entry->agbno = agbno;
+		__entry->len = len;
+		__entry->isfl = isfl;
+		__entry->haveleft = haveleft;
+		__entry->haveright = haveright;
+	),
+	TP_printk("dev %d:%d agno %u agbno %u len %u isfl %d %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno,
+		  __entry->agbno,
+		  __entry->len,
+		  __entry->isfl,
+		  __entry->haveleft ?
+			(__entry->haveright ? "both" : "left") :
+			(__entry->haveright ? "right" : "none"))
+
+);
+
+DECLARE_EVENT_CLASS(xfs_alloc_class,
+	TP_PROTO(struct xfs_alloc_arg *args),
+	TP_ARGS(args),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_agnumber_t, agno)
+		__field(xfs_agblock_t, agbno)
+		__field(xfs_extlen_t, minlen)
+		__field(xfs_extlen_t, maxlen)
+		__field(xfs_extlen_t, mod)
+		__field(xfs_extlen_t, prod)
+		__field(xfs_extlen_t, minleft)
+		__field(xfs_extlen_t, total)
+		__field(xfs_extlen_t, alignment)
+		__field(xfs_extlen_t, minalignslop)
+		__field(xfs_extlen_t, len)
+		__field(short, type)
+		__field(short, otype)
+		__field(char, wasdel)
+		__field(char, wasfromfl)
+		__field(char, isfl)
+		__field(char, userdata)
+		__field(xfs_fsblock_t, firstblock)
+	),
+	TP_fast_assign(
+		__entry->dev = args->mp->m_super->s_dev;
+		__entry->agno = args->agno;
+		__entry->agbno = args->agbno;
+		__entry->minlen = args->minlen;
+		__entry->maxlen = args->maxlen;
+		__entry->mod = args->mod;
+		__entry->prod = args->prod;
+		__entry->minleft = args->minleft;
+		__entry->total = args->total;
+		__entry->alignment = args->alignment;
+		__entry->minalignslop = args->minalignslop;
+		__entry->len = args->len;
+		__entry->type = args->type;
+		__entry->otype = args->otype;
+		__entry->wasdel = args->wasdel;
+		__entry->wasfromfl = args->wasfromfl;
+		__entry->isfl = args->isfl;
+		__entry->userdata = args->userdata;
+		__entry->firstblock = args->firstblock;
+	),
+	TP_printk("dev %d:%d agno %u agbno %u minlen %u maxlen %u mod %u "
+		  "prod %u minleft %u total %u alignment %u minalignslop %u "
+		  "len %u type %s otype %s wasdel %d wasfromfl %d isfl %d "
+		  "userdata %d firstblock 0x%llx",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->agno,
+		  __entry->agbno,
+		  __entry->minlen,
+		  __entry->maxlen,
+		  __entry->mod,
+		  __entry->prod,
+		  __entry->minleft,
+		  __entry->total,
+		  __entry->alignment,
+		  __entry->minalignslop,
+		  __entry->len,
+		  __print_symbolic(__entry->type, XFS_ALLOC_TYPES),
+		  __print_symbolic(__entry->otype, XFS_ALLOC_TYPES),
+		  __entry->wasdel,
+		  __entry->wasfromfl,
+		  __entry->isfl,
+		  __entry->userdata,
+		  __entry->firstblock)
+)
+
+#define DEFINE_ALLOC_EVENT(name) \
+DEFINE_EVENT(xfs_alloc_class, name, \
+	TP_PROTO(struct xfs_alloc_arg *args), \
+	TP_ARGS(args))
+DEFINE_ALLOC_EVENT(xfs_alloc_exact_done);
+DEFINE_ALLOC_EVENT(xfs_alloc_exact_error);
+DEFINE_ALLOC_EVENT(xfs_alloc_near_nominleft);
+DEFINE_ALLOC_EVENT(xfs_alloc_near_first);
+DEFINE_ALLOC_EVENT(xfs_alloc_near_greater);
+DEFINE_ALLOC_EVENT(xfs_alloc_near_lesser);
+DEFINE_ALLOC_EVENT(xfs_alloc_near_error);
+DEFINE_ALLOC_EVENT(xfs_alloc_size_neither);
+DEFINE_ALLOC_EVENT(xfs_alloc_size_noentry);
+DEFINE_ALLOC_EVENT(xfs_alloc_size_nominleft);
+DEFINE_ALLOC_EVENT(xfs_alloc_size_done);
+DEFINE_ALLOC_EVENT(xfs_alloc_size_error);
+DEFINE_ALLOC_EVENT(xfs_alloc_small_freelist);
+DEFINE_ALLOC_EVENT(xfs_alloc_small_notenough);
+DEFINE_ALLOC_EVENT(xfs_alloc_small_done);
+DEFINE_ALLOC_EVENT(xfs_alloc_small_error);
+DEFINE_ALLOC_EVENT(xfs_alloc_vextent_badargs);
+DEFINE_ALLOC_EVENT(xfs_alloc_vextent_nofix);
+DEFINE_ALLOC_EVENT(xfs_alloc_vextent_noagbp);
+DEFINE_ALLOC_EVENT(xfs_alloc_vextent_loopfailed);
+DEFINE_ALLOC_EVENT(xfs_alloc_vextent_allfailed);
+
+DECLARE_EVENT_CLASS(xfs_dir2_class,
+	TP_PROTO(struct xfs_da_args *args),
+	TP_ARGS(args),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__dynamic_array(char, name, args->namelen)
+		__field(int, namelen)
+		__field(xfs_dahash_t, hashval)
+		__field(xfs_ino_t, inumber)
+		__field(int, op_flags)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(args->dp)->i_sb->s_dev;
+		__entry->ino = args->dp->i_ino;
+		if (args->namelen)
+			memcpy(__get_str(name), args->name, args->namelen);
+		__entry->namelen = args->namelen;
+		__entry->hashval = args->hashval;
+		__entry->inumber = args->inumber;
+		__entry->op_flags = args->op_flags;
+	),
+	TP_printk("dev %d:%d ino 0x%llx name %.*s namelen %d hashval 0x%x "
+		  "inumber 0x%llx op_flags %s",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __entry->namelen,
+		  __entry->namelen ? __get_str(name) : NULL,
+		  __entry->namelen,
+		  __entry->hashval,
+		  __entry->inumber,
+		  __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS))
+)
+
+#define DEFINE_DIR2_EVENT(name) \
+DEFINE_EVENT(xfs_dir2_class, name, \
+	TP_PROTO(struct xfs_da_args *args), \
+	TP_ARGS(args))
+DEFINE_DIR2_EVENT(xfs_dir2_sf_addname);
+DEFINE_DIR2_EVENT(xfs_dir2_sf_create);
+DEFINE_DIR2_EVENT(xfs_dir2_sf_lookup);
+DEFINE_DIR2_EVENT(xfs_dir2_sf_replace);
+DEFINE_DIR2_EVENT(xfs_dir2_sf_removename);
+DEFINE_DIR2_EVENT(xfs_dir2_sf_toino4);
+DEFINE_DIR2_EVENT(xfs_dir2_sf_toino8);
+DEFINE_DIR2_EVENT(xfs_dir2_sf_to_block);
+DEFINE_DIR2_EVENT(xfs_dir2_block_addname);
+DEFINE_DIR2_EVENT(xfs_dir2_block_lookup);
+DEFINE_DIR2_EVENT(xfs_dir2_block_replace);
+DEFINE_DIR2_EVENT(xfs_dir2_block_removename);
+DEFINE_DIR2_EVENT(xfs_dir2_block_to_sf);
+DEFINE_DIR2_EVENT(xfs_dir2_block_to_leaf);
+DEFINE_DIR2_EVENT(xfs_dir2_leaf_addname);
+DEFINE_DIR2_EVENT(xfs_dir2_leaf_lookup);
+DEFINE_DIR2_EVENT(xfs_dir2_leaf_replace);
+DEFINE_DIR2_EVENT(xfs_dir2_leaf_removename);
+DEFINE_DIR2_EVENT(xfs_dir2_leaf_to_block);
+DEFINE_DIR2_EVENT(xfs_dir2_leaf_to_node);
+DEFINE_DIR2_EVENT(xfs_dir2_node_addname);
+DEFINE_DIR2_EVENT(xfs_dir2_node_lookup);
+DEFINE_DIR2_EVENT(xfs_dir2_node_replace);
+DEFINE_DIR2_EVENT(xfs_dir2_node_removename);
+DEFINE_DIR2_EVENT(xfs_dir2_node_to_leaf);
+
+DECLARE_EVENT_CLASS(xfs_dir2_space_class,
+	TP_PROTO(struct xfs_da_args *args, int idx),
+	TP_ARGS(args, idx),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(int, op_flags)
+		__field(int, idx)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(args->dp)->i_sb->s_dev;
+		__entry->ino = args->dp->i_ino;
+		__entry->op_flags = args->op_flags;
+		__entry->idx = idx;
+	),
+	TP_printk("dev %d:%d ino 0x%llx op_flags %s index %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS),
+		  __entry->idx)
+)
+
+#define DEFINE_DIR2_SPACE_EVENT(name) \
+DEFINE_EVENT(xfs_dir2_space_class, name, \
+	TP_PROTO(struct xfs_da_args *args, int idx), \
+	TP_ARGS(args, idx))
+DEFINE_DIR2_SPACE_EVENT(xfs_dir2_leafn_add);
+DEFINE_DIR2_SPACE_EVENT(xfs_dir2_leafn_remove);
+DEFINE_DIR2_SPACE_EVENT(xfs_dir2_grow_inode);
+DEFINE_DIR2_SPACE_EVENT(xfs_dir2_shrink_inode);
+
+TRACE_EVENT(xfs_dir2_leafn_moveents,
+	TP_PROTO(struct xfs_da_args *args, int src_idx, int dst_idx, int count),
+	TP_ARGS(args, src_idx, dst_idx, count),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(xfs_ino_t, ino)
+		__field(int, op_flags)
+		__field(int, src_idx)
+		__field(int, dst_idx)
+		__field(int, count)
+	),
+	TP_fast_assign(
+		__entry->dev = VFS_I(args->dp)->i_sb->s_dev;
+		__entry->ino = args->dp->i_ino;
+		__entry->op_flags = args->op_flags;
+		__entry->src_idx = src_idx;
+		__entry->dst_idx = dst_idx;
+		__entry->count = count;
+	),
+	TP_printk("dev %d:%d ino 0x%llx op_flags %s "
+		  "src_idx %d dst_idx %d count %d",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  __entry->ino,
+		  __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS),
+		  __entry->src_idx,
+		  __entry->dst_idx,
+		  __entry->count)
+);
+
+#endif /* _TRACE_XFS_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE xfs_trace
+#include <trace/define_trace.h>
diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h
index ad7fbead4c97..7c220b4227bc 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.h
+++ b/fs/xfs/linux-2.6/xfs_vnode.h
@@ -36,10 +36,13 @@ struct attrlist_cursor_kern;
 /*
  * Flags for read/write calls - same values as IRIX
  */
-#define IO_ISAIO	0x00001		/* don't wait for completion */
 #define IO_ISDIRECT	0x00004		/* bypass page cache */
 #define IO_INVIS	0x00020		/* don't update inode timestamps */
 
+#define XFS_IO_FLAGS \
+	{ IO_ISDIRECT,	"DIRECT" }, \
+	{ IO_INVIS,	"INVIS"}
+
 /*
  * Flush/Invalidate options for vop_toss/flush/flushinval_pages.
  */
diff --git a/fs/xfs/linux-2.6/xfs_xattr.c b/fs/xfs/linux-2.6/xfs_xattr.c
index 497c7fb75cc1..0b1878857fc3 100644
--- a/fs/xfs/linux-2.6/xfs_xattr.c
+++ b/fs/xfs/linux-2.6/xfs_xattr.c
@@ -30,10 +30,10 @@
 
 
 static int
-__xfs_xattr_get(struct inode *inode, const char *name,
+xfs_xattr_get(struct dentry *dentry, const char *name,
 		void *value, size_t size, int xflags)
 {
-	struct xfs_inode *ip = XFS_I(inode);
+	struct xfs_inode *ip = XFS_I(dentry->d_inode);
 	int error, asize = size;
 
 	if (strcmp(name, "") == 0)
@@ -52,10 +52,10 @@ __xfs_xattr_get(struct inode *inode, const char *name,
 }
 
 static int
-__xfs_xattr_set(struct inode *inode, const char *name, const void *value,
+xfs_xattr_set(struct dentry *dentry, const char *name, const void *value,
 		size_t size, int flags, int xflags)
 {
-	struct xfs_inode *ip = XFS_I(inode);
+	struct xfs_inode *ip = XFS_I(dentry->d_inode);
 
 	if (strcmp(name, "") == 0)
 		return -EINVAL;
@@ -71,75 +71,34 @@ __xfs_xattr_set(struct inode *inode, const char *name, const void *value,
 	return -xfs_attr_set(ip, name, (void *)value, size, xflags);
 }
 
-static int
-xfs_xattr_user_get(struct inode *inode, const char *name,
-		void *value, size_t size)
-{
-	return __xfs_xattr_get(inode, name, value, size, 0);
-}
-
-static int
-xfs_xattr_user_set(struct inode *inode, const char *name,
-		const void *value, size_t size, int flags)
-{
-	return __xfs_xattr_set(inode, name, value, size, flags, 0);
-}
-
 static struct xattr_handler xfs_xattr_user_handler = {
 	.prefix	= XATTR_USER_PREFIX,
-	.get	= xfs_xattr_user_get,
-	.set	= xfs_xattr_user_set,
+	.flags	= 0, /* no flags implies user namespace */
+	.get	= xfs_xattr_get,
+	.set	= xfs_xattr_set,
 };
 
-
-static int
-xfs_xattr_trusted_get(struct inode *inode, const char *name,
-		void *value, size_t size)
-{
-	return __xfs_xattr_get(inode, name, value, size, ATTR_ROOT);
-}
-
-static int
-xfs_xattr_trusted_set(struct inode *inode, const char *name,
-		const void *value, size_t size, int flags)
-{
-	return __xfs_xattr_set(inode, name, value, size, flags, ATTR_ROOT);
-}
-
 static struct xattr_handler xfs_xattr_trusted_handler = {
 	.prefix	= XATTR_TRUSTED_PREFIX,
-	.get	= xfs_xattr_trusted_get,
-	.set	= xfs_xattr_trusted_set,
+	.flags	= ATTR_ROOT,
+	.get	= xfs_xattr_get,
+	.set	= xfs_xattr_set,
 };
 
-
-static int
-xfs_xattr_secure_get(struct inode *inode, const char *name,
-		void *value, size_t size)
-{
-	return __xfs_xattr_get(inode, name, value, size, ATTR_SECURE);
-}
-
-static int
-xfs_xattr_secure_set(struct inode *inode, const char *name,
-		const void *value, size_t size, int flags)
-{
-	return __xfs_xattr_set(inode, name, value, size, flags, ATTR_SECURE);
-}
-
 static struct xattr_handler xfs_xattr_security_handler = {
 	.prefix	= XATTR_SECURITY_PREFIX,
-	.get	= xfs_xattr_secure_get,
-	.set	= xfs_xattr_secure_set,
+	.flags	= ATTR_SECURE,
+	.get	= xfs_xattr_get,
+	.set	= xfs_xattr_set,
 };
 
-
 struct xattr_handler *xfs_xattr_handlers[] = {
 	&xfs_xattr_user_handler,
 	&xfs_xattr_trusted_handler,
 	&xfs_xattr_security_handler,
 #ifdef CONFIG_XFS_POSIX_ACL
-	&xfs_xattr_system_handler,
+	&xfs_xattr_acl_access_handler,
+	&xfs_xattr_acl_default_handler,
 #endif
 	NULL
 };
diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c
index 2f3f2229eaaf..d7c7eea09fc2 100644
--- a/fs/xfs/quota/xfs_dquot.c
+++ b/fs/xfs/quota/xfs_dquot.c
@@ -47,6 +47,7 @@
 #include "xfs_trans_space.h"
 #include "xfs_trans_priv.h"
 #include "xfs_qm.h"
+#include "xfs_trace.h"
 
 
 /*
@@ -112,10 +113,7 @@ xfs_qm_dqinit(
 		init_completion(&dqp->q_flush);
 		complete(&dqp->q_flush);
 
-#ifdef XFS_DQUOT_TRACE
-		dqp->q_trace = ktrace_alloc(DQUOT_TRACE_SIZE, KM_NOFS);
-		xfs_dqtrace_entry(dqp, "DQINIT");
-#endif
+		trace_xfs_dqinit(dqp);
 	} else {
 		/*
 		 * Only the q_core portion was zeroed in dqreclaim_one().
@@ -136,10 +134,7 @@ xfs_qm_dqinit(
 		 dqp->q_hash = NULL;
 		 ASSERT(dqp->dq_flnext == dqp->dq_flprev);
 
-#ifdef XFS_DQUOT_TRACE
-		 ASSERT(dqp->q_trace);
-		 xfs_dqtrace_entry(dqp, "DQRECLAIMED_INIT");
-#endif
+		trace_xfs_dqreuse(dqp);
 	}
 
 	/*
@@ -167,13 +162,8 @@ xfs_qm_dqdestroy(
 
 	mutex_destroy(&dqp->q_qlock);
 	sv_destroy(&dqp->q_pinwait);
-
-#ifdef XFS_DQUOT_TRACE
-	if (dqp->q_trace)
-	     ktrace_free(dqp->q_trace);
-	dqp->q_trace = NULL;
-#endif
 	kmem_zone_free(xfs_Gqm->qm_dqzone, dqp);
+
 	atomic_dec(&xfs_Gqm->qm_totaldquots);
 }
 
@@ -195,49 +185,6 @@ xfs_qm_dqinit_core(
 	d->dd_diskdq.d_flags = type;
 }
 
-
-#ifdef XFS_DQUOT_TRACE
-/*
- * Dquot tracing for debugging.
- */
-/* ARGSUSED */
-void
-__xfs_dqtrace_entry(
-	xfs_dquot_t	*dqp,
-	char		*func,
-	void		*retaddr,
-	xfs_inode_t	*ip)
-{
-	xfs_dquot_t	*udqp = NULL;
-	xfs_ino_t	ino = 0;
-
-	ASSERT(dqp->q_trace);
-	if (ip) {
-		ino = ip->i_ino;
-		udqp = ip->i_udquot;
-	}
-	ktrace_enter(dqp->q_trace,
-		     (void *)(__psint_t)DQUOT_KTRACE_ENTRY,
-		     (void *)func,
-		     (void *)(__psint_t)dqp->q_nrefs,
-		     (void *)(__psint_t)dqp->dq_flags,
-		     (void *)(__psint_t)dqp->q_res_bcount,
-		     (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_bcount),
-		     (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_icount),
-		     (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_blk_hardlimit),
-		     (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_blk_softlimit),
-		     (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_ino_hardlimit),
-		     (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_ino_softlimit),
-		     (void *)(__psint_t)be32_to_cpu(dqp->q_core.d_id),
-		     (void *)(__psint_t)current_pid(),
-		     (void *)(__psint_t)ino,
-		     (void *)(__psint_t)retaddr,
-		     (void *)(__psint_t)udqp);
-	return;
-}
-#endif
-
-
 /*
  * If default limits are in force, push them into the dquot now.
  * We overwrite the dquot limits only if they are zero and this
@@ -425,7 +372,8 @@ xfs_qm_dqalloc(
 	xfs_trans_t	*tp = *tpp;
 
 	ASSERT(tp != NULL);
-	xfs_dqtrace_entry(dqp, "DQALLOC");
+
+	trace_xfs_dqalloc(dqp);
 
 	/*
 	 * Initialize the bmap freelist prior to calling bmapi code.
@@ -612,7 +560,8 @@ xfs_qm_dqtobp(
 	 * (in which case we already have the buf).
 	 */
 	if (! newdquot) {
-		xfs_dqtrace_entry(dqp, "DQTOBP READBUF");
+		trace_xfs_dqtobp_read(dqp);
+
 		if ((error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
 					       dqp->q_blkno,
 					       XFS_QI_DQCHUNKLEN(mp),
@@ -670,11 +619,12 @@ xfs_qm_dqread(
 
 	ASSERT(tpp);
 
+	trace_xfs_dqread(dqp);
+
 	/*
 	 * get a pointer to the on-disk dquot and the buffer containing it
 	 * dqp already knows its own type (GROUP/USER).
 	 */
-	xfs_dqtrace_entry(dqp, "DQREAD");
 	if ((error = xfs_qm_dqtobp(tpp, dqp, &ddqp, &bp, flags))) {
 		return (error);
 	}
@@ -763,7 +713,7 @@ xfs_qm_idtodq(
 		 * or if the dquot didn't exist on disk and we ask to
 		 * allocate (ENOENT).
 		 */
-		xfs_dqtrace_entry(dqp, "DQREAD FAIL");
+		trace_xfs_dqread_fail(dqp);
 		cancelflags |= XFS_TRANS_ABORT;
 		goto error0;
 	}
@@ -817,7 +767,8 @@ xfs_qm_dqlookup(
 		 * id can't be modified without the hashlock anyway.
 		 */
 		if (be32_to_cpu(dqp->q_core.d_id) == id && dqp->q_mount == mp) {
-			xfs_dqtrace_entry(dqp, "DQFOUND BY LOOKUP");
+			trace_xfs_dqlookup_found(dqp);
+
 			/*
 			 * All in core dquots must be on the dqlist of mp
 			 */
@@ -827,7 +778,7 @@ xfs_qm_dqlookup(
 			if (dqp->q_nrefs == 0) {
 				ASSERT (XFS_DQ_IS_ON_FREELIST(dqp));
 				if (! xfs_qm_freelist_lock_nowait(xfs_Gqm)) {
-					xfs_dqtrace_entry(dqp, "DQLOOKUP: WANT");
+					trace_xfs_dqlookup_want(dqp);
 
 					/*
 					 * We may have raced with dqreclaim_one()
@@ -857,8 +808,7 @@ xfs_qm_dqlookup(
 					/*
 					 * take it off the freelist
 					 */
-					xfs_dqtrace_entry(dqp,
-							"DQLOOKUP: TAKEOFF FL");
+					trace_xfs_dqlookup_freelist(dqp);
 					XQM_FREELIST_REMOVE(dqp);
 					/* xfs_qm_freelist_print(&(xfs_Gqm->
 							qm_dqfreelist),
@@ -878,8 +828,7 @@ xfs_qm_dqlookup(
 			 */
 			ASSERT(mutex_is_locked(&qh->qh_lock));
 			if (dqp->HL_PREVP != &qh->qh_next) {
-				xfs_dqtrace_entry(dqp,
-						  "DQLOOKUP: HASH MOVETOFRONT");
+				trace_xfs_dqlookup_move(dqp);
 				if ((d = dqp->HL_NEXT))
 					d->HL_PREVP = dqp->HL_PREVP;
 				*(dqp->HL_PREVP) = d;
@@ -889,7 +838,7 @@ xfs_qm_dqlookup(
 				dqp->HL_PREVP = &qh->qh_next;
 				qh->qh_next = dqp;
 			}
-			xfs_dqtrace_entry(dqp, "LOOKUP END");
+			trace_xfs_dqlookup_done(dqp);
 			*O_dqpp = dqp;
 			ASSERT(mutex_is_locked(&qh->qh_lock));
 			return (0);
@@ -971,7 +920,7 @@ xfs_qm_dqget(
 		ASSERT(*O_dqpp);
 		ASSERT(XFS_DQ_IS_LOCKED(*O_dqpp));
 		mutex_unlock(&h->qh_lock);
-		xfs_dqtrace_entry(*O_dqpp, "DQGET DONE (FROM CACHE)");
+		trace_xfs_dqget_hit(*O_dqpp);
 		return (0);	/* success */
 	}
 	XQM_STATS_INC(xqmstats.xs_qm_dqcachemisses);
@@ -1104,7 +1053,7 @@ xfs_qm_dqget(
 	mutex_unlock(&h->qh_lock);
  dqret:
 	ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
-	xfs_dqtrace_entry(dqp, "DQGET DONE");
+	trace_xfs_dqget_miss(dqp);
 	*O_dqpp = dqp;
 	return (0);
 }
@@ -1124,7 +1073,8 @@ xfs_qm_dqput(
 
 	ASSERT(dqp->q_nrefs > 0);
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
-	xfs_dqtrace_entry(dqp, "DQPUT");
+
+	trace_xfs_dqput(dqp);
 
 	if (dqp->q_nrefs != 1) {
 		dqp->q_nrefs--;
@@ -1137,7 +1087,7 @@ xfs_qm_dqput(
 	 * in the right order; but try to get it out-of-order first
 	 */
 	if (! xfs_qm_freelist_lock_nowait(xfs_Gqm)) {
-		xfs_dqtrace_entry(dqp, "DQPUT: FLLOCK-WAIT");
+		trace_xfs_dqput_wait(dqp);
 		xfs_dqunlock(dqp);
 		xfs_qm_freelist_lock(xfs_Gqm);
 		xfs_dqlock(dqp);
@@ -1148,7 +1098,8 @@ xfs_qm_dqput(
 
 		/* We can't depend on nrefs being == 1 here */
 		if (--dqp->q_nrefs == 0) {
-			xfs_dqtrace_entry(dqp, "DQPUT: ON FREELIST");
+			trace_xfs_dqput_free(dqp);
+
 			/*
 			 * insert at end of the freelist.
 			 */
@@ -1196,7 +1147,7 @@ xfs_qm_dqrele(
 	if (!dqp)
 		return;
 
-	xfs_dqtrace_entry(dqp, "DQRELE");
+	trace_xfs_dqrele(dqp);
 
 	xfs_dqlock(dqp);
 	/*
@@ -1229,7 +1180,7 @@ xfs_qm_dqflush(
 
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
 	ASSERT(!completion_done(&dqp->q_flush));
-	xfs_dqtrace_entry(dqp, "DQFLUSH");
+	trace_xfs_dqflush(dqp);
 
 	/*
 	 * If not dirty, or it's pinned and we are not supposed to
@@ -1259,7 +1210,6 @@ xfs_qm_dqflush(
 	 * the ondisk-dquot has already been allocated for.
 	 */
 	if ((error = xfs_qm_dqtobp(NULL, dqp, &ddqp, &bp, XFS_QMOPT_DOWARN))) {
-		xfs_dqtrace_entry(dqp, "DQTOBP FAIL");
 		ASSERT(error != ENOENT);
 		/*
 		 * Quotas could have gotten turned off (ESRCH)
@@ -1297,7 +1247,7 @@ xfs_qm_dqflush(
 	 * get stuck waiting in the write for too long.
 	 */
 	if (XFS_BUF_ISPINNED(bp)) {
-		xfs_dqtrace_entry(dqp, "DQFLUSH LOG FORCE");
+		trace_xfs_dqflush_force(dqp);
 		xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE);
 	}
 
@@ -1308,7 +1258,9 @@ xfs_qm_dqflush(
 	} else {
 		error = xfs_bwrite(mp, bp);
 	}
-	xfs_dqtrace_entry(dqp, "DQFLUSH END");
+
+	trace_xfs_dqflush_done(dqp);
+
 	/*
 	 * dqp is still locked, but caller is free to unlock it now.
 	 */
@@ -1483,7 +1435,7 @@ xfs_qm_dqpurge(
 	 */
 	if (XFS_DQ_IS_DIRTY(dqp)) {
 		int	error;
-		xfs_dqtrace_entry(dqp, "DQPURGE ->DQFLUSH: DQDIRTY");
+
 		/* dqflush unlocks dqflock */
 		/*
 		 * Given that dqpurge is a very rare occurrence, it is OK
diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h
index a2c16bcee90b..a0f7da586d1b 100644
--- a/fs/xfs/quota/xfs_dquot.h
+++ b/fs/xfs/quota/xfs_dquot.h
@@ -85,9 +85,6 @@ typedef struct xfs_dquot {
 	struct completion q_flush;	/* flush completion queue */
 	atomic_t          q_pincount;	/* dquot pin count */
 	wait_queue_head_t q_pinwait;	/* dquot pinning wait queue */
-#ifdef XFS_DQUOT_TRACE
-	struct ktrace	*q_trace;	/* trace header structure */
-#endif
 } xfs_dquot_t;
 
 
@@ -144,24 +141,6 @@ static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
 				     (XFS_IS_UQUOTA_ON((d)->q_mount)) : \
 				     (XFS_IS_OQUOTA_ON((d)->q_mount))))
 
-#ifdef XFS_DQUOT_TRACE
-/*
- * Dquot Tracing stuff.
- */
-#define DQUOT_TRACE_SIZE	64
-#define DQUOT_KTRACE_ENTRY	1
-
-extern void		__xfs_dqtrace_entry(xfs_dquot_t *dqp, char *func,
-					    void *, xfs_inode_t *);
-#define xfs_dqtrace_entry_ino(a,b,ip) \
-		__xfs_dqtrace_entry((a), (b), (void*)__return_address, (ip))
-#define xfs_dqtrace_entry(a,b) \
-		__xfs_dqtrace_entry((a), (b), (void*)__return_address, NULL)
-#else
-#define xfs_dqtrace_entry(a,b)
-#define xfs_dqtrace_entry_ino(a,b,ip)
-#endif
-
 #ifdef QUOTADEBUG
 extern void		xfs_qm_dqprint(xfs_dquot_t *);
 #else
diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index 45b1bfef7388..9e627a8b5b0e 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -47,6 +47,7 @@
 #include "xfs_trans_space.h"
 #include "xfs_utils.h"
 #include "xfs_qm.h"
+#include "xfs_trace.h"
 
 /*
  * The global quota manager. There is only one of these for the entire
@@ -453,7 +454,7 @@ again:
 			xfs_dqunlock(dqp);
 			continue;
 		}
-		xfs_dqtrace_entry(dqp, "FLUSHALL: DQDIRTY");
+
 		/* XXX a sentinel would be better */
 		recl = XFS_QI_MPLRECLAIMS(mp);
 		if (!xfs_dqflock_nowait(dqp)) {
@@ -651,7 +652,7 @@ xfs_qm_dqattach_one(
 	 */
 	dqp = *IO_idqpp;
 	if (dqp) {
-		xfs_dqtrace_entry(dqp, "DQATTACH: found in ip");
+		trace_xfs_dqattach_found(dqp);
 		return 0;
 	}
 
@@ -704,7 +705,7 @@ xfs_qm_dqattach_one(
 	if (error)
 		return error;
 
-	xfs_dqtrace_entry(dqp, "DQATTACH: found by dqget");
+	trace_xfs_dqattach_get(dqp);
 
 	/*
 	 * dqget may have dropped and re-acquired the ilock, but it guarantees
@@ -890,15 +891,15 @@ xfs_qm_dqdetach(
 	if (!(ip->i_udquot || ip->i_gdquot))
 		return;
 
+	trace_xfs_dquot_dqdetach(ip);
+
 	ASSERT(ip->i_ino != ip->i_mount->m_sb.sb_uquotino);
 	ASSERT(ip->i_ino != ip->i_mount->m_sb.sb_gquotino);
 	if (ip->i_udquot) {
-		xfs_dqtrace_entry_ino(ip->i_udquot, "DQDETTACH", ip);
 		xfs_qm_dqrele(ip->i_udquot);
 		ip->i_udquot = NULL;
 	}
 	if (ip->i_gdquot) {
-		xfs_dqtrace_entry_ino(ip->i_gdquot, "DQDETTACH", ip);
 		xfs_qm_dqrele(ip->i_gdquot);
 		ip->i_gdquot = NULL;
 	}
@@ -977,7 +978,6 @@ xfs_qm_sync(
 		 * across a disk write
 		 */
 		xfs_qm_mplist_unlock(mp);
-		xfs_dqtrace_entry(dqp, "XQM_SYNC: DQFLUSH");
 		error = xfs_qm_dqflush(dqp, flush_flags);
 		xfs_dqunlock(dqp);
 		if (error && XFS_FORCED_SHUTDOWN(mp))
@@ -1350,7 +1350,8 @@ xfs_qm_reset_dqcounts(
 	xfs_disk_dquot_t	*ddq;
 	int			j;
 
-	xfs_buftrace("RESET DQUOTS", bp);
+	trace_xfs_reset_dqcounts(bp, _RET_IP_);
+
 	/*
 	 * Reset all counters and timers. They'll be
 	 * started afresh by xfs_qm_quotacheck.
@@ -1543,7 +1544,9 @@ xfs_qm_quotacheck_dqadjust(
 	xfs_qcnt_t		rtblks)
 {
 	ASSERT(XFS_DQ_IS_LOCKED(dqp));
-	xfs_dqtrace_entry(dqp, "QCHECK DQADJUST");
+
+	trace_xfs_dqadjust(dqp);
+
 	/*
 	 * Adjust the inode count and the block count to reflect this inode's
 	 * resource usage.
@@ -1994,7 +1997,9 @@ xfs_qm_shake_freelist(
 		 */
 		if (XFS_DQ_IS_DIRTY(dqp)) {
 			int	error;
-			xfs_dqtrace_entry(dqp, "DQSHAKE: DQDIRTY");
+
+			trace_xfs_dqshake_dirty(dqp);
+
 			/*
 			 * We flush it delayed write, so don't bother
 			 * releasing the mplock.
@@ -2038,7 +2043,9 @@ xfs_qm_shake_freelist(
 				return nreclaimed;
 			goto tryagain;
 		}
-		xfs_dqtrace_entry(dqp, "DQSHAKE: UNLINKING");
+
+		trace_xfs_dqshake_unlink(dqp);
+
 #ifdef QUOTADEBUG
 		cmn_err(CE_DEBUG, "Shake 0x%p, ID 0x%x\n",
 			dqp, be32_to_cpu(dqp->q_core.d_id));
@@ -2125,7 +2132,9 @@ xfs_qm_dqreclaim_one(void)
 		 */
 		if (dqp->dq_flags & XFS_DQ_WANT) {
 			ASSERT(! (dqp->dq_flags & XFS_DQ_INACTIVE));
-			xfs_dqtrace_entry(dqp, "DQRECLAIM: DQWANT");
+
+			trace_xfs_dqreclaim_want(dqp);
+
 			xfs_dqunlock(dqp);
 			xfs_qm_freelist_unlock(xfs_Gqm);
 			if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
@@ -2171,7 +2180,9 @@ xfs_qm_dqreclaim_one(void)
 		 */
 		if (XFS_DQ_IS_DIRTY(dqp)) {
 			int	error;
-			xfs_dqtrace_entry(dqp, "DQRECLAIM: DQDIRTY");
+
+			trace_xfs_dqreclaim_dirty(dqp);
+
 			/*
 			 * We flush it delayed write, so don't bother
 			 * releasing the freelist lock.
@@ -2194,8 +2205,9 @@ xfs_qm_dqreclaim_one(void)
 		if (!mutex_trylock(&dqp->q_hash->qh_lock))
 			goto mplistunlock;
 
+		trace_xfs_dqreclaim_unlink(dqp);
+
 		ASSERT(dqp->q_nrefs == 0);
-		xfs_dqtrace_entry(dqp, "DQRECLAIM: UNLINKING");
 		XQM_MPLIST_REMOVE(&(XFS_QI_MPL_LIST(dqp->q_mount)), dqp);
 		XQM_HASHLIST_REMOVE(dqp->q_hash, dqp);
 		XQM_FREELIST_REMOVE(dqp);
@@ -2430,7 +2442,7 @@ xfs_qm_vop_dqalloc(
 		}
 	}
 	if (uq)
-		xfs_dqtrace_entry_ino(uq, "DQALLOC", ip);
+		trace_xfs_dquot_dqalloc(ip);
 
 	xfs_iunlock(ip, lockflags);
 	if (O_udqpp)
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 5d1a3b98a6e6..873e07e29074 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -49,6 +49,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_utils.h"
 #include "xfs_qm.h"
+#include "xfs_trace.h"
 
 #ifdef DEBUG
 # define qdprintk(s, args...)	cmn_err(CE_DEBUG, s, ## args)
@@ -496,7 +497,6 @@ xfs_qm_scall_setqlim(
 		ASSERT(error != ENOENT);
 		return (error);
 	}
-	xfs_dqtrace_entry(dqp, "Q_SETQLIM: AFT DQGET");
 	xfs_trans_dqjoin(tp, dqp);
 	ddq = &dqp->q_core;
 
@@ -602,7 +602,6 @@ xfs_qm_scall_setqlim(
 	dqp->dq_flags |= XFS_DQ_DIRTY;
 	xfs_trans_log_dquot(tp, dqp);
 
-	xfs_dqtrace_entry(dqp, "Q_SETQLIM: COMMIT");
 	error = xfs_trans_commit(tp, 0);
 	xfs_qm_dqprint(dqp);
 	xfs_qm_dqrele(dqp);
@@ -630,7 +629,6 @@ xfs_qm_scall_getquota(
 		return (error);
 	}
 
-	xfs_dqtrace_entry(dqp, "Q_GETQUOTA SUCCESS");
 	/*
 	 * If everything's NULL, this dquot doesn't quite exist as far as
 	 * our utility programs are concerned.
@@ -893,7 +891,7 @@ xfs_qm_dqrele_all_inodes(
 	uint		 flags)
 {
 	ASSERT(mp->m_quotainfo);
-	xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG);
+	xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG, 0);
 }
 
 /*------------------------------------------------------------------------*/
diff --git a/fs/xfs/support/debug.h b/fs/xfs/support/debug.h
index 6f4fd37c67af..d2d20462fd4f 100644
--- a/fs/xfs/support/debug.h
+++ b/fs/xfs/support/debug.h
@@ -41,10 +41,6 @@ extern void assfail(char *expr, char *f, int l);
 # define STATIC static noinline
 #endif
 
-#ifndef STATIC_INLINE
-# define STATIC_INLINE static inline
-#endif
-
 #else /* DEBUG */
 
 #define ASSERT(expr)	\
@@ -54,19 +50,5 @@ extern void assfail(char *expr, char *f, int l);
 # define STATIC noinline
 #endif
 
-/*
- * We stop inlining of inline functions in debug mode.
- * Unfortunately, this means static inline in header files
- * get multiple definitions, so they need to remain static.
- * This then gives tonnes of warnings about unused but defined
- * functions, so we need to add the unused attribute to prevent
- * these spurious warnings.
- */
-#ifndef STATIC_INLINE
-# define STATIC_INLINE static __attribute__ ((unused)) noinline
-#endif
-
 #endif /* DEBUG */
-
-
 #endif  /* __XFS_SUPPORT_DEBUG_H__ */
diff --git a/fs/xfs/support/ktrace.c b/fs/xfs/support/ktrace.c
deleted file mode 100644
index 2d494c26717f..000000000000
--- a/fs/xfs/support/ktrace.c
+++ /dev/null
@@ -1,323 +0,0 @@
-/*
- * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-#include <xfs.h>
-
-static kmem_zone_t *ktrace_hdr_zone;
-static kmem_zone_t *ktrace_ent_zone;
-static int          ktrace_zentries;
-
-void __init
-ktrace_init(int zentries)
-{
-	ktrace_zentries = roundup_pow_of_two(zentries);
-
-	ktrace_hdr_zone = kmem_zone_init(sizeof(ktrace_t),
-					"ktrace_hdr");
-	ASSERT(ktrace_hdr_zone);
-
-	ktrace_ent_zone = kmem_zone_init(ktrace_zentries
-					* sizeof(ktrace_entry_t),
-					"ktrace_ent");
-	ASSERT(ktrace_ent_zone);
-}
-
-void __exit
-ktrace_uninit(void)
-{
-	kmem_zone_destroy(ktrace_hdr_zone);
-	kmem_zone_destroy(ktrace_ent_zone);
-}
-
-/*
- * ktrace_alloc()
- *
- * Allocate a ktrace header and enough buffering for the given
- * number of entries. Round the number of entries up to a
- * power of 2 so we can do fast masking to get the index from
- * the atomic index counter.
- */
-ktrace_t *
-ktrace_alloc(int nentries, unsigned int __nocast sleep)
-{
-	ktrace_t        *ktp;
-	ktrace_entry_t  *ktep;
-	int		entries;
-
-	ktp = (ktrace_t*)kmem_zone_alloc(ktrace_hdr_zone, sleep);
-
-	if (ktp == (ktrace_t*)NULL) {
-		/*
-		 * KM_SLEEP callers don't expect failure.
-		 */
-		if (sleep & KM_SLEEP)
-			panic("ktrace_alloc: NULL memory on KM_SLEEP request!");
-
-		return NULL;
-	}
-
-	/*
-	 * Special treatment for buffers with the ktrace_zentries entries
-	 */
-	entries = roundup_pow_of_two(nentries);
-	if (entries == ktrace_zentries) {
-		ktep = (ktrace_entry_t*)kmem_zone_zalloc(ktrace_ent_zone,
-							    sleep);
-	} else {
-		ktep = (ktrace_entry_t*)kmem_zalloc((entries * sizeof(*ktep)),
-							    sleep | KM_LARGE);
-	}
-
-	if (ktep == NULL) {
-		/*
-		 * KM_SLEEP callers don't expect failure.
-		 */
-		if (sleep & KM_SLEEP)
-			panic("ktrace_alloc: NULL memory on KM_SLEEP request!");
-
-		kmem_free(ktp);
-
-		return NULL;
-	}
-
-	ktp->kt_entries  = ktep;
-	ktp->kt_nentries = entries;
-	ASSERT(is_power_of_2(entries));
-	ktp->kt_index_mask = entries - 1;
-	atomic_set(&ktp->kt_index, 0);
-	ktp->kt_rollover = 0;
-	return ktp;
-}
-
-
-/*
- * ktrace_free()
- *
- * Free up the ktrace header and buffer.  It is up to the caller
- * to ensure that no-one is referencing it.
- */
-void
-ktrace_free(ktrace_t *ktp)
-{
-	if (ktp == (ktrace_t *)NULL)
-		return;
-
-	/*
-	 * Special treatment for the Vnode trace buffer.
-	 */
-	if (ktp->kt_nentries == ktrace_zentries)
-		kmem_zone_free(ktrace_ent_zone, ktp->kt_entries);
-	else
-		kmem_free(ktp->kt_entries);
-
-	kmem_zone_free(ktrace_hdr_zone, ktp);
-}
-
-
-/*
- * Enter the given values into the "next" entry in the trace buffer.
- * kt_index is always the index of the next entry to be filled.
- */
-void
-ktrace_enter(
-	ktrace_t        *ktp,
-	void            *val0,
-	void            *val1,
-	void            *val2,
-	void            *val3,
-	void            *val4,
-	void            *val5,
-	void            *val6,
-	void            *val7,
-	void            *val8,
-	void            *val9,
-	void            *val10,
-	void            *val11,
-	void            *val12,
-	void            *val13,
-	void            *val14,
-	void            *val15)
-{
-	int             index;
-	ktrace_entry_t  *ktep;
-
-	ASSERT(ktp != NULL);
-
-	/*
-	 * Grab an entry by pushing the index up to the next one.
-	 */
-	index = atomic_add_return(1, &ktp->kt_index);
-	index = (index - 1) & ktp->kt_index_mask;
-	if (!ktp->kt_rollover && index == ktp->kt_nentries - 1)
-		ktp->kt_rollover = 1;
-
-	ASSERT((index >= 0) && (index < ktp->kt_nentries));
-
-	ktep = &(ktp->kt_entries[index]);
-
-	ktep->val[0]  = val0;
-	ktep->val[1]  = val1;
-	ktep->val[2]  = val2;
-	ktep->val[3]  = val3;
-	ktep->val[4]  = val4;
-	ktep->val[5]  = val5;
-	ktep->val[6]  = val6;
-	ktep->val[7]  = val7;
-	ktep->val[8]  = val8;
-	ktep->val[9]  = val9;
-	ktep->val[10] = val10;
-	ktep->val[11] = val11;
-	ktep->val[12] = val12;
-	ktep->val[13] = val13;
-	ktep->val[14] = val14;
-	ktep->val[15] = val15;
-}
-
-/*
- * Return the number of entries in the trace buffer.
- */
-int
-ktrace_nentries(
-	ktrace_t        *ktp)
-{
-	int	index;
-	if (ktp == NULL)
-		return 0;
-
-	index = atomic_read(&ktp->kt_index) & ktp->kt_index_mask;
-	return (ktp->kt_rollover ? ktp->kt_nentries : index);
-}
-
-/*
- * ktrace_first()
- *
- * This is used to find the start of the trace buffer.
- * In conjunction with ktrace_next() it can be used to
- * iterate through the entire trace buffer.  This code does
- * not do any locking because it is assumed that it is called
- * from the debugger.
- *
- * The caller must pass in a pointer to a ktrace_snap
- * structure in which we will keep some state used to
- * iterate through the buffer.  This state must not touched
- * by any code outside of this module.
- */
-ktrace_entry_t *
-ktrace_first(ktrace_t   *ktp, ktrace_snap_t     *ktsp)
-{
-	ktrace_entry_t  *ktep;
-	int             index;
-	int             nentries;
-
-	if (ktp->kt_rollover)
-		index = atomic_read(&ktp->kt_index) & ktp->kt_index_mask;
-	else
-		index = 0;
-
-	ktsp->ks_start = index;
-	ktep = &(ktp->kt_entries[index]);
-
-	nentries = ktrace_nentries(ktp);
-	index++;
-	if (index < nentries) {
-		ktsp->ks_index = index;
-	} else {
-		ktsp->ks_index = 0;
-		if (index > nentries)
-			ktep = NULL;
-	}
-	return ktep;
-}
-
-/*
- * ktrace_next()
- *
- * This is used to iterate through the entries of the given
- * trace buffer.  The caller must pass in the ktrace_snap_t
- * structure initialized by ktrace_first().  The return value
- * will be either a pointer to the next ktrace_entry or NULL
- * if all of the entries have been traversed.
- */
-ktrace_entry_t *
-ktrace_next(
-	ktrace_t        *ktp,
-	ktrace_snap_t   *ktsp)
-{
-	int             index;
-	ktrace_entry_t  *ktep;
-
-	index = ktsp->ks_index;
-	if (index == ktsp->ks_start) {
-		ktep = NULL;
-	} else {
-		ktep = &ktp->kt_entries[index];
-	}
-
-	index++;
-	if (index == ktrace_nentries(ktp)) {
-		ktsp->ks_index = 0;
-	} else {
-		ktsp->ks_index = index;
-	}
-
-	return ktep;
-}
-
-/*
- * ktrace_skip()
- *
- * Skip the next "count" entries and return the entry after that.
- * Return NULL if this causes us to iterate past the beginning again.
- */
-ktrace_entry_t *
-ktrace_skip(
-	ktrace_t        *ktp,
-	int             count,
-	ktrace_snap_t   *ktsp)
-{
-	int             index;
-	int             new_index;
-	ktrace_entry_t  *ktep;
-	int             nentries = ktrace_nentries(ktp);
-
-	index = ktsp->ks_index;
-	new_index = index + count;
-	while (new_index >= nentries) {
-		new_index -= nentries;
-	}
-	if (index == ktsp->ks_start) {
-		/*
-		 * We've iterated around to the start, so we're done.
-		 */
-		ktep = NULL;
-	} else if ((new_index < index) && (index < ktsp->ks_index)) {
-		/*
-		 * We've skipped past the start again, so we're done.
-		 */
-		ktep = NULL;
-		ktsp->ks_index = ktsp->ks_start;
-	} else {
-		ktep = &(ktp->kt_entries[new_index]);
-		new_index++;
-		if (new_index == nentries) {
-			ktsp->ks_index = 0;
-		} else {
-			ktsp->ks_index = new_index;
-		}
-	}
-	return ktep;
-}
diff --git a/fs/xfs/support/ktrace.h b/fs/xfs/support/ktrace.h
deleted file mode 100644
index 741d6947ca60..000000000000
--- a/fs/xfs/support/ktrace.h
+++ /dev/null
@@ -1,85 +0,0 @@
-/*
- * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-#ifndef __XFS_SUPPORT_KTRACE_H__
-#define __XFS_SUPPORT_KTRACE_H__
-
-/*
- * Trace buffer entry structure.
- */
-typedef struct ktrace_entry {
-	void	*val[16];
-} ktrace_entry_t;
-
-/*
- * Trace buffer header structure.
- */
-typedef struct ktrace {
-	int		kt_nentries;	/* number of entries in trace buf */
-	atomic_t	kt_index;	/* current index in entries */
-	unsigned int	kt_index_mask;
-	int		kt_rollover;
-	ktrace_entry_t	*kt_entries;	/* buffer of entries */
-} ktrace_t;
-
-/*
- * Trace buffer snapshot structure.
- */
-typedef struct ktrace_snap {
-	int		ks_start;	/* kt_index at time of snap */
-	int		ks_index;	/* current index */
-} ktrace_snap_t;
-
-
-#ifdef CONFIG_XFS_TRACE
-
-extern void ktrace_init(int zentries);
-extern void ktrace_uninit(void);
-
-extern ktrace_t *ktrace_alloc(int, unsigned int __nocast);
-extern void ktrace_free(ktrace_t *);
-
-extern void ktrace_enter(
-	ktrace_t	*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*,
-	void		*);
-
-extern ktrace_entry_t   *ktrace_first(ktrace_t *, ktrace_snap_t *);
-extern int              ktrace_nentries(ktrace_t *);
-extern ktrace_entry_t   *ktrace_next(ktrace_t *, ktrace_snap_t *);
-extern ktrace_entry_t   *ktrace_skip(ktrace_t *, int, ktrace_snap_t *);
-
-#else
-#define ktrace_init(x)	do { } while (0)
-#define ktrace_uninit()	do { } while (0)
-#endif	/* CONFIG_XFS_TRACE */
-
-#endif	/* __XFS_SUPPORT_KTRACE_H__ */
diff --git a/fs/xfs/xfs.h b/fs/xfs/xfs.h
index 17254b529c54..5ad8ad3a1dcd 100644
--- a/fs/xfs/xfs.h
+++ b/fs/xfs/xfs.h
@@ -25,21 +25,5 @@
 /* #define QUOTADEBUG 1 */
 #endif
 
-#ifdef CONFIG_XFS_TRACE
-#define XFS_ALLOC_TRACE 1
-#define XFS_ATTR_TRACE 1
-#define XFS_BLI_TRACE 1
-#define XFS_BMAP_TRACE 1
-#define XFS_BTREE_TRACE 1
-#define XFS_DIR2_TRACE 1
-#define XFS_DQUOT_TRACE 1
-#define XFS_ILOCK_TRACE 1
-#define XFS_LOG_TRACE 1
-#define XFS_RW_TRACE 1
-#define XFS_BUF_TRACE 1
-#define XFS_INODE_TRACE 1
-#define XFS_FILESTREAMS_TRACE 1
-#endif
-
 #include <linux-2.6/xfs_linux.h>
 #endif	/* __XFS_H__ */
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 947b150df8ed..00fd357c3e46 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -49,7 +49,8 @@ extern int xfs_acl_chmod(struct inode *inode);
 extern int posix_acl_access_exists(struct inode *inode);
 extern int posix_acl_default_exists(struct inode *inode);
 
-extern struct xattr_handler xfs_xattr_system_handler;
+extern struct xattr_handler xfs_xattr_acl_access_handler;
+extern struct xattr_handler xfs_xattr_acl_default_handler;
 #else
 # define xfs_check_acl					NULL
 # define xfs_get_acl(inode, type)			NULL
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index a5d54bf4931b..6702bd865811 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h
@@ -86,6 +86,20 @@ typedef struct xfs_agf {
 #define	XFS_AGF_NUM_BITS	12
 #define	XFS_AGF_ALL_BITS	((1 << XFS_AGF_NUM_BITS) - 1)
 
+#define XFS_AGF_FLAGS \
+	{ XFS_AGF_MAGICNUM,	"MAGICNUM" }, \
+	{ XFS_AGF_VERSIONNUM,	"VERSIONNUM" }, \
+	{ XFS_AGF_SEQNO,	"SEQNO" }, \
+	{ XFS_AGF_LENGTH,	"LENGTH" }, \
+	{ XFS_AGF_ROOTS,	"ROOTS" }, \
+	{ XFS_AGF_LEVELS,	"LEVELS" }, \
+	{ XFS_AGF_FLFIRST,	"FLFIRST" }, \
+	{ XFS_AGF_FLLAST,	"FLLAST" }, \
+	{ XFS_AGF_FLCOUNT,	"FLCOUNT" }, \
+	{ XFS_AGF_FREEBLKS,	"FREEBLKS" }, \
+	{ XFS_AGF_LONGEST,	"LONGEST" }, \
+	{ XFS_AGF_BTREEBLKS,	"BTREEBLKS" }
+
 /* disk block (xfs_daddr_t) in the AG */
 #define XFS_AGF_DADDR(mp)	((xfs_daddr_t)(1 << (mp)->m_sectbb_log))
 #define	XFS_AGF_BLOCK(mp)	XFS_HDR_BLOCK(mp, XFS_AGF_DADDR(mp))
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index 2cf944eb796d..275b1f4f9430 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -38,6 +38,7 @@
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 
 #define XFS_ABSDIFF(a,b)	(((a) <= (b)) ? ((b) - (a)) : ((a) - (b)))
@@ -51,30 +52,6 @@ xfs_alloc_search_busy(xfs_trans_t *tp,
 		    xfs_agblock_t bno,
 		    xfs_extlen_t len);
 
-#if defined(XFS_ALLOC_TRACE)
-ktrace_t *xfs_alloc_trace_buf;
-
-#define	TRACE_ALLOC(s,a)	\
-	xfs_alloc_trace_alloc(__func__, s, a, __LINE__)
-#define	TRACE_FREE(s,a,b,x,f)	\
-	xfs_alloc_trace_free(__func__, s, mp, a, b, x, f, __LINE__)
-#define	TRACE_MODAGF(s,a,f)	\
-	xfs_alloc_trace_modagf(__func__, s, mp, a, f, __LINE__)
-#define	TRACE_BUSY(__func__,s,ag,agb,l,sl,tp)	\
-	xfs_alloc_trace_busy(__func__, s, mp, ag, agb, l, sl, tp, XFS_ALLOC_KTRACE_BUSY, __LINE__)
-#define	TRACE_UNBUSY(__func__,s,ag,sl,tp)	\
-	xfs_alloc_trace_busy(__func__, s, mp, ag, -1, -1, sl, tp, XFS_ALLOC_KTRACE_UNBUSY, __LINE__)
-#define	TRACE_BUSYSEARCH(__func__,s,ag,agb,l,tp)	\
-	xfs_alloc_trace_busy(__func__, s, mp, ag, agb, l, 0, tp, XFS_ALLOC_KTRACE_BUSYSEARCH, __LINE__)
-#else
-#define	TRACE_ALLOC(s,a)
-#define	TRACE_FREE(s,a,b,x,f)
-#define	TRACE_MODAGF(s,a,f)
-#define	TRACE_BUSY(s,a,ag,agb,l,sl,tp)
-#define	TRACE_UNBUSY(fname,s,ag,sl,tp)
-#define	TRACE_BUSYSEARCH(fname,s,ag,agb,l,tp)
-#endif	/* XFS_ALLOC_TRACE */
-
 /*
  * Prototypes for per-ag allocation routines
  */
@@ -498,124 +475,6 @@ xfs_alloc_read_agfl(
 	return 0;
 }
 
-#if defined(XFS_ALLOC_TRACE)
-/*
- * Add an allocation trace entry for an alloc call.
- */
-STATIC void
-xfs_alloc_trace_alloc(
-	const char	*name,		/* function tag string */
-	char		*str,		/* additional string */
-	xfs_alloc_arg_t	*args,		/* allocation argument structure */
-	int		line)		/* source line number */
-{
-	ktrace_enter(xfs_alloc_trace_buf,
-		(void *)(__psint_t)(XFS_ALLOC_KTRACE_ALLOC | (line << 16)),
-		(void *)name,
-		(void *)str,
-		(void *)args->mp,
-		(void *)(__psunsigned_t)args->agno,
-		(void *)(__psunsigned_t)args->agbno,
-		(void *)(__psunsigned_t)args->minlen,
-		(void *)(__psunsigned_t)args->maxlen,
-		(void *)(__psunsigned_t)args->mod,
-		(void *)(__psunsigned_t)args->prod,
-		(void *)(__psunsigned_t)args->minleft,
-		(void *)(__psunsigned_t)args->total,
-		(void *)(__psunsigned_t)args->alignment,
-		(void *)(__psunsigned_t)args->len,
-		(void *)((((__psint_t)args->type) << 16) |
-			 (__psint_t)args->otype),
-		(void *)(__psint_t)((args->wasdel << 3) |
-				    (args->wasfromfl << 2) |
-				    (args->isfl << 1) |
-				    (args->userdata << 0)));
-}
-
-/*
- * Add an allocation trace entry for a free call.
- */
-STATIC void
-xfs_alloc_trace_free(
-	const char	*name,		/* function tag string */
-	char		*str,		/* additional string */
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_agnumber_t	agno,		/* allocation group number */
-	xfs_agblock_t	agbno,		/* a.g. relative block number */
-	xfs_extlen_t	len,		/* length of extent */
-	int		isfl,		/* set if is freelist allocation/free */
-	int		line)		/* source line number */
-{
-	ktrace_enter(xfs_alloc_trace_buf,
-		(void *)(__psint_t)(XFS_ALLOC_KTRACE_FREE | (line << 16)),
-		(void *)name,
-		(void *)str,
-		(void *)mp,
-		(void *)(__psunsigned_t)agno,
-		(void *)(__psunsigned_t)agbno,
-		(void *)(__psunsigned_t)len,
-		(void *)(__psint_t)isfl,
-		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL);
-}
-
-/*
- * Add an allocation trace entry for modifying an agf.
- */
-STATIC void
-xfs_alloc_trace_modagf(
-	const char	*name,		/* function tag string */
-	char		*str,		/* additional string */
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_agf_t	*agf,		/* new agf value */
-	int		flags,		/* logging flags for agf */
-	int		line)		/* source line number */
-{
-	ktrace_enter(xfs_alloc_trace_buf,
-		(void *)(__psint_t)(XFS_ALLOC_KTRACE_MODAGF | (line << 16)),
-		(void *)name,
-		(void *)str,
-		(void *)mp,
-		(void *)(__psint_t)flags,
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_seqno),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_length),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_roots[XFS_BTNUM_BNO]),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_roots[XFS_BTNUM_CNT]),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_flfirst),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_fllast),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_flcount),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_freeblks),
-		(void *)(__psunsigned_t)be32_to_cpu(agf->agf_longest));
-}
-
-STATIC void
-xfs_alloc_trace_busy(
-	const char	*name,		/* function tag string */
-	char		*str,		/* additional string */
-	xfs_mount_t	*mp,		/* file system mount point */
-	xfs_agnumber_t	agno,		/* allocation group number */
-	xfs_agblock_t	agbno,		/* a.g. relative block number */
-	xfs_extlen_t	len,		/* length of extent */
-	int		slot,		/* perag Busy slot */
-	xfs_trans_t	*tp,
-	int		trtype,		/* type: add, delete, search */
-	int		line)		/* source line number */
-{
-	ktrace_enter(xfs_alloc_trace_buf,
-		(void *)(__psint_t)(trtype | (line << 16)),
-		(void *)name,
-		(void *)str,
-		(void *)mp,
-		(void *)(__psunsigned_t)agno,
-		(void *)(__psunsigned_t)agbno,
-		(void *)(__psunsigned_t)len,
-		(void *)(__psint_t)slot,
-		(void *)tp,
-		NULL, NULL, NULL, NULL, NULL, NULL, NULL);
-}
-#endif	/* XFS_ALLOC_TRACE */
-
 /*
  * Allocation group level functions.
  */
@@ -665,9 +524,6 @@ xfs_alloc_ag_vextent(
 	 */
 	if (args->agbno != NULLAGBLOCK) {
 		xfs_agf_t	*agf;	/* allocation group freelist header */
-#ifdef XFS_ALLOC_TRACE
-		xfs_mount_t	*mp = args->mp;
-#endif
 		long		slen = (long)args->len;
 
 		ASSERT(args->len >= args->minlen && args->len <= args->maxlen);
@@ -682,7 +538,6 @@ xfs_alloc_ag_vextent(
 			args->pag->pagf_freeblks -= args->len;
 			ASSERT(be32_to_cpu(agf->agf_freeblks) <=
 				be32_to_cpu(agf->agf_length));
-			TRACE_MODAGF(NULL, agf, XFS_AGF_FREEBLKS);
 			xfs_alloc_log_agf(args->tp, args->agbp,
 						XFS_AGF_FREEBLKS);
 			/* search the busylist for these blocks */
@@ -792,13 +647,14 @@ xfs_alloc_ag_vextent_exact(
 	}
 	xfs_btree_del_cursor(bno_cur, XFS_BTREE_NOERROR);
 	xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-	TRACE_ALLOC("normal", args);
+
+	trace_xfs_alloc_exact_done(args);
 	args->wasfromfl = 0;
 	return 0;
 
 error0:
 	xfs_btree_del_cursor(bno_cur, XFS_BTREE_ERROR);
-	TRACE_ALLOC("error", args);
+	trace_xfs_alloc_exact_error(args);
 	return error;
 }
 
@@ -958,7 +814,7 @@ xfs_alloc_ag_vextent_near(
 		args->len = blen;
 		if (!xfs_alloc_fix_minleft(args)) {
 			xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-			TRACE_ALLOC("nominleft", args);
+			trace_xfs_alloc_near_nominleft(args);
 			return 0;
 		}
 		blen = args->len;
@@ -981,7 +837,8 @@ xfs_alloc_ag_vextent_near(
 			goto error0;
 		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
 		xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_NOERROR);
-		TRACE_ALLOC("first", args);
+
+		trace_xfs_alloc_near_first(args);
 		return 0;
 	}
 	/*
@@ -1272,7 +1129,7 @@ xfs_alloc_ag_vextent_near(
 	 * If we couldn't get anything, give up.
 	 */
 	if (bno_cur_lt == NULL && bno_cur_gt == NULL) {
-		TRACE_ALLOC("neither", args);
+		trace_xfs_alloc_size_neither(args);
 		args->agbno = NULLAGBLOCK;
 		return 0;
 	}
@@ -1299,7 +1156,7 @@ xfs_alloc_ag_vextent_near(
 	args->len = XFS_EXTLEN_MIN(ltlena, args->maxlen);
 	xfs_alloc_fix_len(args);
 	if (!xfs_alloc_fix_minleft(args)) {
-		TRACE_ALLOC("nominleft", args);
+		trace_xfs_alloc_near_nominleft(args);
 		xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_NOERROR);
 		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
 		return 0;
@@ -1314,13 +1171,18 @@ xfs_alloc_ag_vextent_near(
 	if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno, ltlen,
 			ltnew, rlen, XFSA_FIXUP_BNO_OK)))
 		goto error0;
-	TRACE_ALLOC(j ? "gt" : "lt", args);
+
+	if (j)
+		trace_xfs_alloc_near_greater(args);
+	else
+		trace_xfs_alloc_near_lesser(args);
+
 	xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
 	xfs_btree_del_cursor(bno_cur_lt, XFS_BTREE_NOERROR);
 	return 0;
 
  error0:
-	TRACE_ALLOC("error", args);
+	trace_xfs_alloc_near_error(args);
 	if (cnt_cur != NULL)
 		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_ERROR);
 	if (bno_cur_lt != NULL)
@@ -1371,7 +1233,7 @@ xfs_alloc_ag_vextent_size(
 			goto error0;
 		if (i == 0 || flen == 0) {
 			xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-			TRACE_ALLOC("noentry", args);
+			trace_xfs_alloc_size_noentry(args);
 			return 0;
 		}
 		ASSERT(i == 1);
@@ -1448,7 +1310,7 @@ xfs_alloc_ag_vextent_size(
 	xfs_alloc_fix_len(args);
 	if (rlen < args->minlen || !xfs_alloc_fix_minleft(args)) {
 		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR);
-		TRACE_ALLOC("nominleft", args);
+		trace_xfs_alloc_size_nominleft(args);
 		args->agbno = NULLAGBLOCK;
 		return 0;
 	}
@@ -1471,11 +1333,11 @@ xfs_alloc_ag_vextent_size(
 		args->agbno + args->len <=
 			be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length),
 		error0);
-	TRACE_ALLOC("normal", args);
+	trace_xfs_alloc_size_done(args);
 	return 0;
 
 error0:
-	TRACE_ALLOC("error", args);
+	trace_xfs_alloc_size_error(args);
 	if (cnt_cur)
 		xfs_btree_del_cursor(cnt_cur, XFS_BTREE_ERROR);
 	if (bno_cur)
@@ -1534,7 +1396,7 @@ xfs_alloc_ag_vextent_small(
 				be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length),
 				error0);
 			args->wasfromfl = 1;
-			TRACE_ALLOC("freelist", args);
+			trace_xfs_alloc_small_freelist(args);
 			*stat = 0;
 			return 0;
 		}
@@ -1556,17 +1418,17 @@ xfs_alloc_ag_vextent_small(
 	 */
 	if (flen < args->minlen) {
 		args->agbno = NULLAGBLOCK;
-		TRACE_ALLOC("notenough", args);
+		trace_xfs_alloc_small_notenough(args);
 		flen = 0;
 	}
 	*fbnop = fbno;
 	*flenp = flen;
 	*stat = 1;
-	TRACE_ALLOC("normal", args);
+	trace_xfs_alloc_small_done(args);
 	return 0;
 
 error0:
-	TRACE_ALLOC("error", args);
+	trace_xfs_alloc_small_error(args);
 	return error;
 }
 
@@ -1809,17 +1671,14 @@ xfs_free_ag_extent(
 			be32_to_cpu(agf->agf_freeblks) <=
 			be32_to_cpu(agf->agf_length),
 			error0);
-		TRACE_MODAGF(NULL, agf, XFS_AGF_FREEBLKS);
 		xfs_alloc_log_agf(tp, agbp, XFS_AGF_FREEBLKS);
 		if (!isfl)
 			xfs_trans_mod_sb(tp, XFS_TRANS_SB_FDBLOCKS, (long)len);
 		XFS_STATS_INC(xs_freex);
 		XFS_STATS_ADD(xs_freeb, len);
 	}
-	TRACE_FREE(haveleft ?
-			(haveright ? "both" : "left") :
-			(haveright ? "right" : "none"),
-		agno, bno, len, isfl);
+
+	trace_xfs_free_extent(mp, agno, bno, len, isfl, haveleft, haveright);
 
 	/*
 	 * Since blocks move to the free list without the coordination
@@ -1836,7 +1695,7 @@ xfs_free_ag_extent(
 	return 0;
 
  error0:
-	TRACE_FREE("error", agno, bno, len, isfl);
+	trace_xfs_free_extent(mp, agno, bno, len, isfl, -1, -1);
 	if (bno_cur)
 		xfs_btree_del_cursor(bno_cur, XFS_BTREE_ERROR);
 	if (cnt_cur)
@@ -2122,7 +1981,6 @@ xfs_alloc_get_freelist(
 		logflags |= XFS_AGF_BTREEBLKS;
 	}
 
-	TRACE_MODAGF(NULL, agf, logflags);
 	xfs_alloc_log_agf(tp, agbp, logflags);
 	*bnop = bno;
 
@@ -2165,6 +2023,8 @@ xfs_alloc_log_agf(
 		sizeof(xfs_agf_t)
 	};
 
+	trace_xfs_agf(tp->t_mountp, XFS_BUF_TO_AGF(bp), fields, _RET_IP_);
+
 	xfs_btree_offsets(fields, offsets, XFS_AGF_NUM_BITS, &first, &last);
 	xfs_trans_log_buf(tp, bp, (uint)first, (uint)last);
 }
@@ -2230,13 +2090,11 @@ xfs_alloc_put_freelist(
 		logflags |= XFS_AGF_BTREEBLKS;
 	}
 
-	TRACE_MODAGF(NULL, agf, logflags);
 	xfs_alloc_log_agf(tp, agbp, logflags);
 
 	ASSERT(be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp));
 	blockp = &agfl->agfl_bno[be32_to_cpu(agf->agf_fllast)];
 	*blockp = cpu_to_be32(bno);
-	TRACE_MODAGF(NULL, agf, logflags);
 	xfs_alloc_log_agf(tp, agbp, logflags);
 	xfs_trans_log_buf(tp, agflbp,
 		(int)((xfs_caddr_t)blockp - (xfs_caddr_t)agfl),
@@ -2399,7 +2257,7 @@ xfs_alloc_vextent(
 	    args->minlen > args->maxlen || args->minlen > agsize ||
 	    args->mod >= args->prod) {
 		args->fsbno = NULLFSBLOCK;
-		TRACE_ALLOC("badargs", args);
+		trace_xfs_alloc_vextent_badargs(args);
 		return 0;
 	}
 	minleft = args->minleft;
@@ -2418,12 +2276,12 @@ xfs_alloc_vextent(
 		error = xfs_alloc_fix_freelist(args, 0);
 		args->minleft = minleft;
 		if (error) {
-			TRACE_ALLOC("nofix", args);
+			trace_xfs_alloc_vextent_nofix(args);
 			goto error0;
 		}
 		if (!args->agbp) {
 			up_read(&mp->m_peraglock);
-			TRACE_ALLOC("noagbp", args);
+			trace_xfs_alloc_vextent_noagbp(args);
 			break;
 		}
 		args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno);
@@ -2488,7 +2346,7 @@ xfs_alloc_vextent(
 			error = xfs_alloc_fix_freelist(args, flags);
 			args->minleft = minleft;
 			if (error) {
-				TRACE_ALLOC("nofix", args);
+				trace_xfs_alloc_vextent_nofix(args);
 				goto error0;
 			}
 			/*
@@ -2499,7 +2357,9 @@ xfs_alloc_vextent(
 					goto error0;
 				break;
 			}
-			TRACE_ALLOC("loopfailed", args);
+
+			trace_xfs_alloc_vextent_loopfailed(args);
+
 			/*
 			 * Didn't work, figure out the next iteration.
 			 */
@@ -2526,7 +2386,7 @@ xfs_alloc_vextent(
 			if (args->agno == sagno) {
 				if (no_min == 1) {
 					args->agbno = NULLAGBLOCK;
-					TRACE_ALLOC("allfailed", args);
+					trace_xfs_alloc_vextent_allfailed(args);
 					break;
 				}
 				if (flags == 0) {
@@ -2642,16 +2502,16 @@ xfs_alloc_mark_busy(xfs_trans_t *tp,
 		}
 	}
 
+	trace_xfs_alloc_busy(mp, agno, bno, len, n);
+
 	if (n < XFS_PAGB_NUM_SLOTS) {
 		bsy = &mp->m_perag[agno].pagb_list[n];
 		mp->m_perag[agno].pagb_count++;
-		TRACE_BUSY("xfs_alloc_mark_busy", "got", agno, bno, len, n, tp);
 		bsy->busy_start = bno;
 		bsy->busy_length = len;
 		bsy->busy_tp = tp;
 		xfs_trans_add_busy(tp, agno, n);
 	} else {
-		TRACE_BUSY("xfs_alloc_mark_busy", "FULL", agno, bno, len, -1, tp);
 		/*
 		 * The busy list is full!  Since it is now not possible to
 		 * track the free block, make this a synchronous transaction
@@ -2678,12 +2538,12 @@ xfs_alloc_clear_busy(xfs_trans_t *tp,
 	list = mp->m_perag[agno].pagb_list;
 
 	ASSERT(idx < XFS_PAGB_NUM_SLOTS);
+
+	trace_xfs_alloc_unbusy(mp, agno, idx, list[idx].busy_tp == tp);
+
 	if (list[idx].busy_tp == tp) {
-		TRACE_UNBUSY("xfs_alloc_clear_busy", "found", agno, idx, tp);
 		list[idx].busy_tp = NULL;
 		mp->m_perag[agno].pagb_count--;
-	} else {
-		TRACE_UNBUSY("xfs_alloc_clear_busy", "missing", agno, idx, tp);
 	}
 
 	spin_unlock(&mp->m_perag[agno].pagb_lock);
@@ -2703,45 +2563,41 @@ xfs_alloc_search_busy(xfs_trans_t *tp,
 	xfs_mount_t		*mp;
 	xfs_perag_busy_t	*bsy;
 	xfs_agblock_t		uend, bend;
-	xfs_lsn_t		lsn;
+	xfs_lsn_t		lsn = 0;
 	int			cnt;
 
 	mp = tp->t_mountp;
 
 	spin_lock(&mp->m_perag[agno].pagb_lock);
-	cnt = mp->m_perag[agno].pagb_count;
 
 	uend = bno + len - 1;
 
-	/* search pagb_list for this slot, skipping open slots */
-	for (bsy = mp->m_perag[agno].pagb_list; cnt; bsy++) {
+	/*
+	 * search pagb_list for this slot, skipping open slots. We have to
+	 * search the entire array as there may be multiple overlaps and
+	 * we have to get the most recent LSN for the log force to push out
+	 * all the transactions that span the range.
+	 */
+	for (cnt = 0; cnt < mp->m_perag[agno].pagb_count; cnt++) {
+		bsy = &mp->m_perag[agno].pagb_list[cnt];
+		if (!bsy->busy_tp)
+			continue;
 
-		/*
-		 * (start1,length1) within (start2, length2)
-		 */
-		if (bsy->busy_tp != NULL) {
-			bend = bsy->busy_start + bsy->busy_length - 1;
-			if ((bno > bend) || (uend < bsy->busy_start)) {
-				cnt--;
-			} else {
-				TRACE_BUSYSEARCH("xfs_alloc_search_busy",
-					 "found1", agno, bno, len, tp);
-				break;
-			}
-		}
+		bend = bsy->busy_start + bsy->busy_length - 1;
+		if (bno > bend || uend < bsy->busy_start)
+			continue;
+
+		/* (start1,length1) within (start2, length2) */
+		if (XFS_LSN_CMP(bsy->busy_tp->t_commit_lsn, lsn) > 0)
+			lsn = bsy->busy_tp->t_commit_lsn;
 	}
+	spin_unlock(&mp->m_perag[agno].pagb_lock);
+	trace_xfs_alloc_busysearch(tp->t_mountp, agno, bno, len, lsn);
 
 	/*
 	 * If a block was found, force the log through the LSN of the
 	 * transaction that freed the block
 	 */
-	if (cnt) {
-		TRACE_BUSYSEARCH("xfs_alloc_search_busy", "found", agno, bno, len, tp);
-		lsn = bsy->busy_tp->t_commit_lsn;
-		spin_unlock(&mp->m_perag[agno].pagb_lock);
+	if (lsn)
 		xfs_log_force(mp, lsn, XFS_LOG_FORCE|XFS_LOG_SYNC);
-	} else {
-		TRACE_BUSYSEARCH("xfs_alloc_search_busy", "not-found", agno, bno, len, tp);
-		spin_unlock(&mp->m_perag[agno].pagb_lock);
-	}
 }
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h
index e704caee10df..599bffa39784 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/xfs_alloc.h
@@ -37,6 +37,15 @@ typedef enum xfs_alloctype
 	XFS_ALLOCTYPE_THIS_BNO		/* at exactly this block */
 } xfs_alloctype_t;
 
+#define XFS_ALLOC_TYPES \
+	{ XFS_ALLOCTYPE_ANY_AG,		"ANY_AG" }, \
+	{ XFS_ALLOCTYPE_FIRST_AG,	"FIRST_AG" }, \
+	{ XFS_ALLOCTYPE_START_AG,	"START_AG" }, \
+	{ XFS_ALLOCTYPE_THIS_AG,	"THIS_AG" }, \
+	{ XFS_ALLOCTYPE_START_BNO,	"START_BNO" }, \
+	{ XFS_ALLOCTYPE_NEAR_BNO,	"NEAR_BNO" }, \
+	{ XFS_ALLOCTYPE_THIS_BNO,	"THIS_BNO" }
+
 /*
  * Flags for xfs_alloc_fix_freelist.
  */
@@ -109,24 +118,6 @@ xfs_alloc_longest_free_extent(struct xfs_mount *mp,
 
 #ifdef __KERNEL__
 
-#if defined(XFS_ALLOC_TRACE)
-/*
- * Allocation tracing buffer size.
- */
-#define	XFS_ALLOC_TRACE_SIZE	4096
-extern ktrace_t *xfs_alloc_trace_buf;
-
-/*
- * Types for alloc tracing.
- */
-#define	XFS_ALLOC_KTRACE_ALLOC	1
-#define	XFS_ALLOC_KTRACE_FREE	2
-#define	XFS_ALLOC_KTRACE_MODAGF	3
-#define	XFS_ALLOC_KTRACE_BUSY	4
-#define	XFS_ALLOC_KTRACE_UNBUSY	5
-#define	XFS_ALLOC_KTRACE_BUSYSEARCH	6
-#endif
-
 void
 xfs_alloc_mark_busy(xfs_trans_t *tp,
 		xfs_agnumber_t agno,
diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c
index c10c3a292d30..adbd9141aea1 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/xfs_alloc_btree.c
@@ -39,6 +39,7 @@
 #include "xfs_ialloc.h"
 #include "xfs_alloc.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 
 STATIC struct xfs_btree_cur *
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index 4ece1906bd41..e953b6cfb2a8 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -47,6 +47,7 @@
 #include "xfs_trans_space.h"
 #include "xfs_rw.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 /*
  * xfs_attr.c
@@ -89,10 +90,6 @@ STATIC int xfs_attr_rmtval_remove(xfs_da_args_t *args);
 
 #define ATTR_RMTVALUE_MAPSIZE	1	/* # of map entries at once */
 
-#if defined(XFS_ATTR_TRACE)
-ktrace_t *xfs_attr_trace_buf;
-#endif
-
 STATIC int
 xfs_attr_name_to_xname(
 	struct xfs_name	*xname,
@@ -123,9 +120,13 @@ xfs_inode_hasattr(
  * Overall external interface routines.
  *========================================================================*/
 
-int
-xfs_attr_fetch(xfs_inode_t *ip, struct xfs_name *name,
-		char *value, int *valuelenp, int flags)
+STATIC int
+xfs_attr_get_int(
+	struct xfs_inode	*ip,
+	struct xfs_name		*name,
+	char			*value,
+	int			*valuelenp,
+	int			flags)
 {
 	xfs_da_args_t   args;
 	int             error;
@@ -188,7 +189,7 @@ xfs_attr_get(
 		return error;
 
 	xfs_ilock(ip, XFS_ILOCK_SHARED);
-	error = xfs_attr_fetch(ip, &xname, value, valuelenp, flags);
+	error = xfs_attr_get_int(ip, &xname, value, valuelenp, flags);
 	xfs_iunlock(ip, XFS_ILOCK_SHARED);
 	return(error);
 }
@@ -636,7 +637,6 @@ xfs_attr_list_int(xfs_attr_list_context_t *context)
 		return EIO;
 
 	xfs_ilock(dp, XFS_ILOCK_SHARED);
-	xfs_attr_trace_l_c("syscall start", context);
 
 	/*
 	 * Decide on what work routines to call based on the inode size.
@@ -652,7 +652,6 @@ xfs_attr_list_int(xfs_attr_list_context_t *context)
 	}
 
 	xfs_iunlock(dp, XFS_ILOCK_SHARED);
-	xfs_attr_trace_l_c("syscall end", context);
 
 	return error;
 }
@@ -698,7 +697,7 @@ xfs_attr_put_listent(xfs_attr_list_context_t *context, int flags,
 			context->count * sizeof(alist->al_offset[0]);
 	context->firstu -= ATTR_ENTSIZE(namelen);
 	if (context->firstu < arraytop) {
-		xfs_attr_trace_l_c("buffer full", context);
+		trace_xfs_attr_list_full(context);
 		alist->al_more = 1;
 		context->seen_enough = 1;
 		return 1;
@@ -710,7 +709,7 @@ xfs_attr_put_listent(xfs_attr_list_context_t *context, int flags,
 	aep->a_name[namelen] = 0;
 	alist->al_offset[context->count++] = context->firstu;
 	alist->al_count = context->count;
-	xfs_attr_trace_l_c("add", context);
+	trace_xfs_attr_list_add(context);
 	return 0;
 }
 
@@ -1849,7 +1848,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 			node = bp->data;
 			switch (be16_to_cpu(node->hdr.info.magic)) {
 			case XFS_DA_NODE_MAGIC:
-				xfs_attr_trace_l_cn("wrong blk", context, node);
+				trace_xfs_attr_list_wrong_blk(context);
 				xfs_da_brelse(NULL, bp);
 				bp = NULL;
 				break;
@@ -1857,20 +1856,18 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 				leaf = bp->data;
 				if (cursor->hashval > be32_to_cpu(leaf->entries[
 				    be16_to_cpu(leaf->hdr.count)-1].hashval)) {
-					xfs_attr_trace_l_cl("wrong blk",
-							   context, leaf);
+					trace_xfs_attr_list_wrong_blk(context);
 					xfs_da_brelse(NULL, bp);
 					bp = NULL;
 				} else if (cursor->hashval <=
 					     be32_to_cpu(leaf->entries[0].hashval)) {
-					xfs_attr_trace_l_cl("maybe wrong blk",
-							   context, leaf);
+					trace_xfs_attr_list_wrong_blk(context);
 					xfs_da_brelse(NULL, bp);
 					bp = NULL;
 				}
 				break;
 			default:
-				xfs_attr_trace_l_c("wrong blk - ??", context);
+				trace_xfs_attr_list_wrong_blk(context);
 				xfs_da_brelse(NULL, bp);
 				bp = NULL;
 			}
@@ -1915,8 +1912,8 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
 				if (cursor->hashval
 						<= be32_to_cpu(btree->hashval)) {
 					cursor->blkno = be32_to_cpu(btree->before);
-					xfs_attr_trace_l_cb("descending",
-							    context, btree);
+					trace_xfs_attr_list_node_descend(context,
+									 btree);
 					break;
 				}
 			}
@@ -2143,8 +2140,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
 		dblkno = XFS_FSB_TO_DADDR(mp, map.br_startblock),
 		blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
 
-		bp = xfs_buf_get_flags(mp->m_ddev_targp, dblkno, blkcnt,
-				       XFS_BUF_LOCK | XBF_DONT_BLOCK);
+		bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt,
+				 XFS_BUF_LOCK | XBF_DONT_BLOCK);
 		ASSERT(bp);
 		ASSERT(!XFS_BUF_GETERROR(bp));
 
@@ -2266,85 +2263,3 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
 	}
 	return(0);
 }
-
-#if defined(XFS_ATTR_TRACE)
-/*
- * Add a trace buffer entry for an attr_list context structure.
- */
-void
-xfs_attr_trace_l_c(char *where, struct xfs_attr_list_context *context)
-{
-	xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_C, where, context,
-		(__psunsigned_t)NULL,
-		(__psunsigned_t)NULL,
-		(__psunsigned_t)NULL);
-}
-
-/*
- * Add a trace buffer entry for a context structure and a Btree node.
- */
-void
-xfs_attr_trace_l_cn(char *where, struct xfs_attr_list_context *context,
-			 struct xfs_da_intnode *node)
-{
-	xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CN, where, context,
-		(__psunsigned_t)be16_to_cpu(node->hdr.count),
-		(__psunsigned_t)be32_to_cpu(node->btree[0].hashval),
-		(__psunsigned_t)be32_to_cpu(node->btree[
-				    be16_to_cpu(node->hdr.count)-1].hashval));
-}
-
-/*
- * Add a trace buffer entry for a context structure and a Btree element.
- */
-void
-xfs_attr_trace_l_cb(char *where, struct xfs_attr_list_context *context,
-			  struct xfs_da_node_entry *btree)
-{
-	xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CB, where, context,
-		(__psunsigned_t)be32_to_cpu(btree->hashval),
-		(__psunsigned_t)be32_to_cpu(btree->before),
-		(__psunsigned_t)NULL);
-}
-
-/*
- * Add a trace buffer entry for a context structure and a leaf block.
- */
-void
-xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context,
-			      struct xfs_attr_leafblock *leaf)
-{
-	xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CL, where, context,
-		(__psunsigned_t)be16_to_cpu(leaf->hdr.count),
-		(__psunsigned_t)be32_to_cpu(leaf->entries[0].hashval),
-		(__psunsigned_t)be32_to_cpu(leaf->entries[
-				be16_to_cpu(leaf->hdr.count)-1].hashval));
-}
-
-/*
- * Add a trace buffer entry for the arguments given to the routine,
- * generic form.
- */
-void
-xfs_attr_trace_enter(int type, char *where,
-			 struct xfs_attr_list_context *context,
-			 __psunsigned_t a13, __psunsigned_t a14,
-			 __psunsigned_t a15)
-{
-	ASSERT(xfs_attr_trace_buf);
-	ktrace_enter(xfs_attr_trace_buf, (void *)((__psunsigned_t)type),
-		(void *)((__psunsigned_t)where),
-		(void *)((__psunsigned_t)context->dp),
-		(void *)((__psunsigned_t)context->cursor->hashval),
-		(void *)((__psunsigned_t)context->cursor->blkno),
-		(void *)((__psunsigned_t)context->cursor->offset),
-		(void *)((__psunsigned_t)context->alist),
-		(void *)((__psunsigned_t)context->bufsize),
-		(void *)((__psunsigned_t)context->count),
-		(void *)((__psunsigned_t)context->firstu),
-		NULL,
-		(void *)((__psunsigned_t)context->dupcnt),
-		(void *)((__psunsigned_t)context->flags),
-		(void *)a13, (void *)a14, (void *)a15);
-}
-#endif	/* XFS_ATTR_TRACE */
diff --git a/fs/xfs/xfs_attr.h b/fs/xfs/xfs_attr.h
index fb3b2a68b9b9..59b410ce69a1 100644
--- a/fs/xfs/xfs_attr.h
+++ b/fs/xfs/xfs_attr.h
@@ -48,6 +48,16 @@ struct xfs_attr_list_context;
 #define ATTR_KERNOTIME	0x1000	/* [kernel] don't update inode timestamps */
 #define ATTR_KERNOVAL	0x2000	/* [kernel] get attr size only, not value */
 
+#define XFS_ATTR_FLAGS \
+	{ ATTR_DONTFOLLOW, 	"DONTFOLLOW" }, \
+	{ ATTR_ROOT,		"ROOT" }, \
+	{ ATTR_TRUST,		"TRUST" }, \
+	{ ATTR_SECURE,		"SECURE" }, \
+	{ ATTR_CREATE,		"CREATE" }, \
+	{ ATTR_REPLACE,		"REPLACE" }, \
+	{ ATTR_KERNOTIME,	"KERNOTIME" }, \
+	{ ATTR_KERNOVAL,	"KERNOVAL" }
+
 /*
  * The maximum size (into the kernel or returned from the kernel) of an
  * attribute value or the buffer used for an attr_list() call.  Larger
@@ -131,7 +141,6 @@ typedef struct xfs_attr_list_context {
  */
 int xfs_attr_calc_size(struct xfs_inode *, int, int, int *);
 int xfs_attr_inactive(struct xfs_inode *dp);
-int xfs_attr_fetch(struct xfs_inode *, struct xfs_name *, char *, int *, int);
 int xfs_attr_rmtval_get(struct xfs_da_args *args);
 int xfs_attr_list_int(struct xfs_attr_list_context *);
 
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c
index afdc8911637d..baf41b5af756 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/xfs_attr_leaf.c
@@ -42,6 +42,7 @@
 #include "xfs_attr.h"
 #include "xfs_attr_leaf.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 /*
  * xfs_attr_leaf.c
@@ -98,7 +99,7 @@ STATIC int xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index);
  * If namespace bits don't match return 0.
  * If all match then return 1.
  */
-STATIC_INLINE int
+STATIC int
 xfs_attr_namesp_match(int arg_flags, int ondisk_flags)
 {
 	return XFS_ATTR_NSP_ONDISK(ondisk_flags) == XFS_ATTR_NSP_ARGS_TO_ONDISK(arg_flags);
@@ -594,7 +595,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	cursor = context->cursor;
 	ASSERT(cursor != NULL);
 
-	xfs_attr_trace_l_c("sf start", context);
+	trace_xfs_attr_list_sf(context);
 
 	/*
 	 * If the buffer is large enough and the cursor is at the start,
@@ -627,7 +628,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 				return error;
 			sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
 		}
-		xfs_attr_trace_l_c("sf big-gulp", context);
+		trace_xfs_attr_list_sf_all(context);
 		return(0);
 	}
 
@@ -653,7 +654,6 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 			XFS_CORRUPTION_ERROR("xfs_attr_shortform_list",
 					     XFS_ERRLEVEL_LOW,
 					     context->dp->i_mount, sfe);
-			xfs_attr_trace_l_c("sf corrupted", context);
 			kmem_free(sbuf);
 			return XFS_ERROR(EFSCORRUPTED);
 		}
@@ -693,7 +693,6 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	}
 	if (i == nsbuf) {
 		kmem_free(sbuf);
-		xfs_attr_trace_l_c("blk end", context);
 		return(0);
 	}
 
@@ -719,7 +718,6 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
 	}
 
 	kmem_free(sbuf);
-	xfs_attr_trace_l_c("sf E-O-F", context);
 	return(0);
 }
 
@@ -2323,7 +2321,7 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
 	cursor = context->cursor;
 	cursor->initted = 1;
 
-	xfs_attr_trace_l_cl("blk start", context, leaf);
+	trace_xfs_attr_list_leaf(context);
 
 	/*
 	 * Re-find our place in the leaf block if this is a new syscall.
@@ -2344,7 +2342,7 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
 			}
 		}
 		if (i == be16_to_cpu(leaf->hdr.count)) {
-			xfs_attr_trace_l_c("not found", context);
+			trace_xfs_attr_list_notfound(context);
 			return(0);
 		}
 	} else {
@@ -2419,7 +2417,7 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
 			break;
 		cursor->offset++;
 	}
-	xfs_attr_trace_l_cl("blk end", context, leaf);
+	trace_xfs_attr_list_leaf_end(context);
 	return(retval);
 }
 
diff --git a/fs/xfs/xfs_attr_sf.h b/fs/xfs/xfs_attr_sf.h
index ea22839caed2..76ab7b0cbb3a 100644
--- a/fs/xfs/xfs_attr_sf.h
+++ b/fs/xfs/xfs_attr_sf.h
@@ -25,8 +25,6 @@
  * to fit into the literal area of the inode.
  */
 
-struct xfs_inode;
-
 /*
  * Entries are packed toward the top as tight as possible.
  */
@@ -69,42 +67,4 @@ typedef struct xfs_attr_sf_sort {
 	(be16_to_cpu(((xfs_attr_shortform_t *)	\
 		((dp)->i_afp->if_u1.if_data))->hdr.totsize))
 
-#if defined(XFS_ATTR_TRACE)
-/*
- * Kernel tracing support for attribute lists
- */
-struct xfs_attr_list_context;
-struct xfs_da_intnode;
-struct xfs_da_node_entry;
-struct xfs_attr_leafblock;
-
-#define	XFS_ATTR_TRACE_SIZE	4096	/* size of global trace buffer */
-extern ktrace_t	*xfs_attr_trace_buf;
-
-/*
- * Trace record types.
- */
-#define	XFS_ATTR_KTRACE_L_C	1	/* context */
-#define	XFS_ATTR_KTRACE_L_CN	2	/* context, node */
-#define	XFS_ATTR_KTRACE_L_CB	3	/* context, btree */
-#define	XFS_ATTR_KTRACE_L_CL	4	/* context, leaf */
-
-void xfs_attr_trace_l_c(char *where, struct xfs_attr_list_context *context);
-void xfs_attr_trace_l_cn(char *where, struct xfs_attr_list_context *context,
-			      struct xfs_da_intnode *node);
-void xfs_attr_trace_l_cb(char *where, struct xfs_attr_list_context *context,
-			      struct xfs_da_node_entry *btree);
-void xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context,
-			      struct xfs_attr_leafblock *leaf);
-void xfs_attr_trace_enter(int type, char *where,
-			     struct xfs_attr_list_context *context,
-			     __psunsigned_t a13, __psunsigned_t a14,
-			     __psunsigned_t a15);
-#else
-#define	xfs_attr_trace_l_c(w,c)
-#define	xfs_attr_trace_l_cn(w,c,n)
-#define	xfs_attr_trace_l_cb(w,c,b)
-#define	xfs_attr_trace_l_cl(w,c,l)
-#endif /* XFS_ATTR_TRACE */
-
 #endif	/* __XFS_ATTR_SF_H__ */
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 8971fb09d387..98251cdc52aa 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -54,6 +54,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_filestream.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 
 #ifdef DEBUG
@@ -272,71 +273,6 @@ xfs_bmap_isaeof(
 	int             whichfork,	/* data or attribute fork */
 	char		*aeof);		/* return value */
 
-#ifdef XFS_BMAP_TRACE
-/*
- * Add bmap trace entry prior to a call to xfs_iext_remove.
- */
-STATIC void
-xfs_bmap_trace_delete(
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry(entries) deleted */
-	xfs_extnum_t	cnt,		/* count of entries deleted, 1 or 2 */
-	int		whichfork);	/* data or attr fork */
-
-/*
- * Add bmap trace entry prior to a call to xfs_iext_insert, or
- * reading in the extents list from the disk (in the btree).
- */
-STATIC void
-xfs_bmap_trace_insert(
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry(entries) inserted */
-	xfs_extnum_t	cnt,		/* count of entries inserted, 1 or 2 */
-	xfs_bmbt_irec_t	*r1,		/* inserted record 1 */
-	xfs_bmbt_irec_t	*r2,		/* inserted record 2 or null */
-	int		whichfork);	/* data or attr fork */
-
-/*
- * Add bmap trace entry after updating an extent record in place.
- */
-STATIC void
-xfs_bmap_trace_post_update(
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry updated */
-	int		whichfork);	/* data or attr fork */
-
-/*
- * Add bmap trace entry prior to updating an extent record in place.
- */
-STATIC void
-xfs_bmap_trace_pre_update(
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry to be updated */
-	int		whichfork);	/* data or attr fork */
-
-#define	XFS_BMAP_TRACE_DELETE(d,ip,i,c,w)	\
-	xfs_bmap_trace_delete(__func__,d,ip,i,c,w)
-#define	XFS_BMAP_TRACE_INSERT(d,ip,i,c,r1,r2,w)	\
-	xfs_bmap_trace_insert(__func__,d,ip,i,c,r1,r2,w)
-#define	XFS_BMAP_TRACE_POST_UPDATE(d,ip,i,w)	\
-	xfs_bmap_trace_post_update(__func__,d,ip,i,w)
-#define	XFS_BMAP_TRACE_PRE_UPDATE(d,ip,i,w)	\
-	xfs_bmap_trace_pre_update(__func__,d,ip,i,w)
-#else
-#define	XFS_BMAP_TRACE_DELETE(d,ip,i,c,w)
-#define	XFS_BMAP_TRACE_INSERT(d,ip,i,c,r1,r2,w)
-#define	XFS_BMAP_TRACE_POST_UPDATE(d,ip,i,w)
-#define	XFS_BMAP_TRACE_PRE_UPDATE(d,ip,i,w)
-#endif	/* XFS_BMAP_TRACE */
-
 /*
  * Compute the worst-case number of indirect blocks that will be used
  * for ip's delayed extent of length "len".
@@ -363,18 +299,6 @@ xfs_bmap_validate_ret(
 #define	xfs_bmap_validate_ret(bno,len,flags,mval,onmap,nmap)
 #endif /* DEBUG */
 
-#if defined(XFS_RW_TRACE)
-STATIC void
-xfs_bunmap_trace(
-	xfs_inode_t		*ip,
-	xfs_fileoff_t		bno,
-	xfs_filblks_t		len,
-	int			flags,
-	inst_t			*ra);
-#else
-#define	xfs_bunmap_trace(ip, bno, len, flags, ra)
-#endif	/* XFS_RW_TRACE */
-
 STATIC int
 xfs_bmap_count_tree(
 	xfs_mount_t     *mp,
@@ -590,9 +514,9 @@ xfs_bmap_add_extent(
 	 * already extents in the list.
 	 */
 	if (nextents == 0) {
-		XFS_BMAP_TRACE_INSERT("insert empty", ip, 0, 1, new, NULL,
-			whichfork);
-		xfs_iext_insert(ifp, 0, 1, new);
+		xfs_iext_insert(ip, 0, 1, new,
+				whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0);
+
 		ASSERT(cur == NULL);
 		ifp->if_lastex = 0;
 		if (!isnullstartblock(new->br_startblock)) {
@@ -759,26 +683,10 @@ xfs_bmap_add_extent_delay_real(
 	xfs_filblks_t		temp=0;	/* value for dnew calculations */
 	xfs_filblks_t		temp2=0;/* value for dnew calculations */
 	int			tmp_rval;	/* partial logging flags */
-	enum {				/* bit number definitions for state */
-		LEFT_CONTIG,	RIGHT_CONTIG,
-		LEFT_FILLING,	RIGHT_FILLING,
-		LEFT_DELAY,	RIGHT_DELAY,
-		LEFT_VALID,	RIGHT_VALID
-	};
 
 #define	LEFT		r[0]
 #define	RIGHT		r[1]
 #define	PREV		r[2]
-#define	MASK(b)		(1 << (b))
-#define	MASK2(a,b)	(MASK(a) | MASK(b))
-#define	MASK3(a,b,c)	(MASK2(a,b) | MASK(c))
-#define	MASK4(a,b,c,d)	(MASK3(a,b,c) | MASK(d))
-#define	STATE_SET(b,v)	((v) ? (state |= MASK(b)) : (state &= ~MASK(b)))
-#define	STATE_TEST(b)	(state & MASK(b))
-#define	STATE_SET_TEST(b,v)	((v) ? ((state |= MASK(b)), 1) : \
-				       ((state &= ~MASK(b)), 0))
-#define	SWITCH_STATE		\
-	(state & MASK4(LEFT_FILLING, RIGHT_FILLING, LEFT_CONTIG, RIGHT_CONTIG))
 
 	/*
 	 * Set up a bunch of variables to make the tests simpler.
@@ -790,69 +698,80 @@ xfs_bmap_add_extent_delay_real(
 	new_endoff = new->br_startoff + new->br_blockcount;
 	ASSERT(PREV.br_startoff <= new->br_startoff);
 	ASSERT(PREV.br_startoff + PREV.br_blockcount >= new_endoff);
+
 	/*
 	 * Set flags determining what part of the previous delayed allocation
 	 * extent is being replaced by a real allocation.
 	 */
-	STATE_SET(LEFT_FILLING, PREV.br_startoff == new->br_startoff);
-	STATE_SET(RIGHT_FILLING,
-		PREV.br_startoff + PREV.br_blockcount == new_endoff);
+	if (PREV.br_startoff == new->br_startoff)
+		state |= BMAP_LEFT_FILLING;
+	if (PREV.br_startoff + PREV.br_blockcount == new_endoff)
+		state |= BMAP_RIGHT_FILLING;
+
 	/*
 	 * Check and set flags if this segment has a left neighbor.
 	 * Don't set contiguous if the combined extent would be too large.
 	 */
-	if (STATE_SET_TEST(LEFT_VALID, idx > 0)) {
+	if (idx > 0) {
+		state |= BMAP_LEFT_VALID;
 		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &LEFT);
-		STATE_SET(LEFT_DELAY, isnullstartblock(LEFT.br_startblock));
+
+		if (isnullstartblock(LEFT.br_startblock))
+			state |= BMAP_LEFT_DELAY;
 	}
-	STATE_SET(LEFT_CONTIG,
-		STATE_TEST(LEFT_VALID) && !STATE_TEST(LEFT_DELAY) &&
-		LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
-		LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
-		LEFT.br_state == new->br_state &&
-		LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN);
+
+	if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
+	    LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
+	    LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
+	    LEFT.br_state == new->br_state &&
+	    LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+		state |= BMAP_LEFT_CONTIG;
+
 	/*
 	 * Check and set flags if this segment has a right neighbor.
 	 * Don't set contiguous if the combined extent would be too large.
 	 * Also check for all-three-contiguous being too large.
 	 */
-	if (STATE_SET_TEST(RIGHT_VALID,
-			idx <
-			ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1)) {
+	if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
+		state |= BMAP_RIGHT_VALID;
 		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx + 1), &RIGHT);
-		STATE_SET(RIGHT_DELAY, isnullstartblock(RIGHT.br_startblock));
+
+		if (isnullstartblock(RIGHT.br_startblock))
+			state |= BMAP_RIGHT_DELAY;
 	}
-	STATE_SET(RIGHT_CONTIG,
-		STATE_TEST(RIGHT_VALID) && !STATE_TEST(RIGHT_DELAY) &&
-		new_endoff == RIGHT.br_startoff &&
-		new->br_startblock + new->br_blockcount ==
-		    RIGHT.br_startblock &&
-		new->br_state == RIGHT.br_state &&
-		new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
-		((state & MASK3(LEFT_CONTIG, LEFT_FILLING, RIGHT_FILLING)) !=
-		  MASK3(LEFT_CONTIG, LEFT_FILLING, RIGHT_FILLING) ||
-		 LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
-		     <= MAXEXTLEN));
+
+	if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
+	    new_endoff == RIGHT.br_startoff &&
+	    new->br_startblock + new->br_blockcount == RIGHT.br_startblock &&
+	    new->br_state == RIGHT.br_state &&
+	    new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
+	    ((state & (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
+		       BMAP_RIGHT_FILLING)) !=
+		      (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
+		       BMAP_RIGHT_FILLING) ||
+	     LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
+			<= MAXEXTLEN))
+		state |= BMAP_RIGHT_CONTIG;
+
 	error = 0;
 	/*
 	 * Switch out based on the FILLING and CONTIG state bits.
 	 */
-	switch (SWITCH_STATE) {
-
-	case MASK4(LEFT_FILLING, RIGHT_FILLING, LEFT_CONTIG, RIGHT_CONTIG):
+	switch (state & (BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
+			 BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG)) {
+	case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
+	     BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
 		/*
 		 * Filling in all of a previously delayed allocation extent.
 		 * The left and right neighbors are both contiguous with new.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|RF|LC|RC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
 			LEFT.br_blockcount + PREV.br_blockcount +
 			RIGHT.br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|RF|LC|RC", ip, idx - 1,
-			XFS_DATA_FORK);
-		XFS_BMAP_TRACE_DELETE("LF|RF|LC|RC", ip, idx, 2, XFS_DATA_FORK);
-		xfs_iext_remove(ifp, idx, 2);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
+		xfs_iext_remove(ip, idx, 2, state);
 		ip->i_df.if_lastex = idx - 1;
 		ip->i_d.di_nextents--;
 		if (cur == NULL)
@@ -885,20 +804,18 @@ xfs_bmap_add_extent_delay_real(
 			RIGHT.br_blockcount;
 		break;
 
-	case MASK3(LEFT_FILLING, RIGHT_FILLING, LEFT_CONTIG):
+	case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
 		/*
 		 * Filling in all of a previously delayed allocation extent.
 		 * The left neighbor is contiguous, the right is not.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|RF|LC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
 			LEFT.br_blockcount + PREV.br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|RF|LC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx - 1;
-		XFS_BMAP_TRACE_DELETE("LF|RF|LC", ip, idx, 1, XFS_DATA_FORK);
-		xfs_iext_remove(ifp, idx, 1);
+		xfs_iext_remove(ip, idx, 1, state);
 		if (cur == NULL)
 			rval = XFS_ILOG_DEXT;
 		else {
@@ -921,19 +838,19 @@ xfs_bmap_add_extent_delay_real(
 			PREV.br_blockcount;
 		break;
 
-	case MASK3(LEFT_FILLING, RIGHT_FILLING, RIGHT_CONTIG):
+	case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
 		/*
 		 * Filling in all of a previously delayed allocation extent.
 		 * The right neighbor is contiguous, the left is not.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|RF|RC", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_startblock(ep, new->br_startblock);
 		xfs_bmbt_set_blockcount(ep,
 			PREV.br_blockcount + RIGHT.br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|RF|RC", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx;
-		XFS_BMAP_TRACE_DELETE("LF|RF|RC", ip, idx + 1, 1, XFS_DATA_FORK);
-		xfs_iext_remove(ifp, idx + 1, 1);
+		xfs_iext_remove(ip, idx + 1, 1, state);
 		if (cur == NULL)
 			rval = XFS_ILOG_DEXT;
 		else {
@@ -956,15 +873,16 @@ xfs_bmap_add_extent_delay_real(
 			RIGHT.br_blockcount;
 		break;
 
-	case MASK2(LEFT_FILLING, RIGHT_FILLING):
+	case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
 		/*
 		 * Filling in all of a previously delayed allocation extent.
 		 * Neither the left nor right neighbors are contiguous with
 		 * the new one.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|RF", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_startblock(ep, new->br_startblock);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|RF", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx;
 		ip->i_d.di_nextents++;
 		if (cur == NULL)
@@ -987,19 +905,20 @@ xfs_bmap_add_extent_delay_real(
 		temp2 = new->br_blockcount;
 		break;
 
-	case MASK2(LEFT_FILLING, LEFT_CONTIG):
+	case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
 		/*
 		 * Filling in the first part of a previous delayed allocation.
 		 * The left neighbor is contiguous.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|LC", ip, idx - 1, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
 			LEFT.br_blockcount + new->br_blockcount);
 		xfs_bmbt_set_startoff(ep,
 			PREV.br_startoff + new->br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|LC", ip, idx - 1, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
 		temp = PREV.br_blockcount - new->br_blockcount;
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|LC", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep, temp);
 		ip->i_df.if_lastex = idx - 1;
 		if (cur == NULL)
@@ -1021,7 +940,7 @@ xfs_bmap_add_extent_delay_real(
 		temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
 			startblockval(PREV.br_startblock));
 		xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
-		XFS_BMAP_TRACE_POST_UPDATE("LF|LC", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 		*dnew = temp;
 		/* DELTA: The boundary between two in-core extents moved. */
 		temp = LEFT.br_startoff;
@@ -1029,18 +948,16 @@ xfs_bmap_add_extent_delay_real(
 			PREV.br_blockcount;
 		break;
 
-	case MASK(LEFT_FILLING):
+	case BMAP_LEFT_FILLING:
 		/*
 		 * Filling in the first part of a previous delayed allocation.
 		 * The left neighbor is not contiguous.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_startoff(ep, new_endoff);
 		temp = PREV.br_blockcount - new->br_blockcount;
 		xfs_bmbt_set_blockcount(ep, temp);
-		XFS_BMAP_TRACE_INSERT("LF", ip, idx, 1, new, NULL,
-			XFS_DATA_FORK);
-		xfs_iext_insert(ifp, idx, 1, new);
+		xfs_iext_insert(ip, idx, 1, new, state);
 		ip->i_df.if_lastex = idx;
 		ip->i_d.di_nextents++;
 		if (cur == NULL)
@@ -1071,27 +988,27 @@ xfs_bmap_add_extent_delay_real(
 			(cur ? cur->bc_private.b.allocated : 0));
 		ep = xfs_iext_get_ext(ifp, idx + 1);
 		xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
-		XFS_BMAP_TRACE_POST_UPDATE("LF", ip, idx + 1, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_);
 		*dnew = temp;
 		/* DELTA: One in-core extent is split in two. */
 		temp = PREV.br_startoff;
 		temp2 = PREV.br_blockcount;
 		break;
 
-	case MASK2(RIGHT_FILLING, RIGHT_CONTIG):
+	case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
 		/*
 		 * Filling in the last part of a previous delayed allocation.
 		 * The right neighbor is contiguous with the new allocation.
 		 */
 		temp = PREV.br_blockcount - new->br_blockcount;
-		XFS_BMAP_TRACE_PRE_UPDATE("RF|RC", ip, idx, XFS_DATA_FORK);
-		XFS_BMAP_TRACE_PRE_UPDATE("RF|RC", ip, idx + 1, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+		trace_xfs_bmap_pre_update(ip, idx + 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep, temp);
 		xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, idx + 1),
 			new->br_startoff, new->br_startblock,
 			new->br_blockcount + RIGHT.br_blockcount,
 			RIGHT.br_state);
-		XFS_BMAP_TRACE_POST_UPDATE("RF|RC", ip, idx + 1, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_);
 		ip->i_df.if_lastex = idx + 1;
 		if (cur == NULL)
 			rval = XFS_ILOG_DEXT;
@@ -1112,7 +1029,7 @@ xfs_bmap_add_extent_delay_real(
 		temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
 			startblockval(PREV.br_startblock));
 		xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
-		XFS_BMAP_TRACE_POST_UPDATE("RF|RC", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 		*dnew = temp;
 		/* DELTA: The boundary between two in-core extents moved. */
 		temp = PREV.br_startoff;
@@ -1120,17 +1037,15 @@ xfs_bmap_add_extent_delay_real(
 			RIGHT.br_blockcount;
 		break;
 
-	case MASK(RIGHT_FILLING):
+	case BMAP_RIGHT_FILLING:
 		/*
 		 * Filling in the last part of a previous delayed allocation.
 		 * The right neighbor is not contiguous.
 		 */
 		temp = PREV.br_blockcount - new->br_blockcount;
-		XFS_BMAP_TRACE_PRE_UPDATE("RF", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep, temp);
-		XFS_BMAP_TRACE_INSERT("RF", ip, idx + 1, 1, new, NULL,
-			XFS_DATA_FORK);
-		xfs_iext_insert(ifp, idx + 1, 1, new);
+		xfs_iext_insert(ip, idx + 1, 1, new, state);
 		ip->i_df.if_lastex = idx + 1;
 		ip->i_d.di_nextents++;
 		if (cur == NULL)
@@ -1161,7 +1076,7 @@ xfs_bmap_add_extent_delay_real(
 			(cur ? cur->bc_private.b.allocated : 0));
 		ep = xfs_iext_get_ext(ifp, idx);
 		xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
-		XFS_BMAP_TRACE_POST_UPDATE("RF", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 		*dnew = temp;
 		/* DELTA: One in-core extent is split in two. */
 		temp = PREV.br_startoff;
@@ -1175,7 +1090,7 @@ xfs_bmap_add_extent_delay_real(
 		 * This case is avoided almost all the time.
 		 */
 		temp = new->br_startoff - PREV.br_startoff;
-		XFS_BMAP_TRACE_PRE_UPDATE("0", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, 0, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep, temp);
 		r[0] = *new;
 		r[1].br_state = PREV.br_state;
@@ -1183,9 +1098,7 @@ xfs_bmap_add_extent_delay_real(
 		r[1].br_startoff = new_endoff;
 		temp2 = PREV.br_startoff + PREV.br_blockcount - new_endoff;
 		r[1].br_blockcount = temp2;
-		XFS_BMAP_TRACE_INSERT("0", ip, idx + 1, 2, &r[0], &r[1],
-			XFS_DATA_FORK);
-		xfs_iext_insert(ifp, idx + 1, 2, &r[0]);
+		xfs_iext_insert(ip, idx + 1, 2, &r[0], state);
 		ip->i_df.if_lastex = idx + 1;
 		ip->i_d.di_nextents++;
 		if (cur == NULL)
@@ -1242,24 +1155,24 @@ xfs_bmap_add_extent_delay_real(
 		}
 		ep = xfs_iext_get_ext(ifp, idx);
 		xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
-		XFS_BMAP_TRACE_POST_UPDATE("0", ip, idx, XFS_DATA_FORK);
-		XFS_BMAP_TRACE_PRE_UPDATE("0", ip, idx + 2, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+		trace_xfs_bmap_pre_update(ip, idx + 2, state, _THIS_IP_);
 		xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx + 2),
 			nullstartblock((int)temp2));
-		XFS_BMAP_TRACE_POST_UPDATE("0", ip, idx + 2, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx + 2, state, _THIS_IP_);
 		*dnew = temp + temp2;
 		/* DELTA: One in-core extent is split in three. */
 		temp = PREV.br_startoff;
 		temp2 = PREV.br_blockcount;
 		break;
 
-	case MASK3(LEFT_FILLING, LEFT_CONTIG, RIGHT_CONTIG):
-	case MASK3(RIGHT_FILLING, LEFT_CONTIG, RIGHT_CONTIG):
-	case MASK2(LEFT_FILLING, RIGHT_CONTIG):
-	case MASK2(RIGHT_FILLING, LEFT_CONTIG):
-	case MASK2(LEFT_CONTIG, RIGHT_CONTIG):
-	case MASK(LEFT_CONTIG):
-	case MASK(RIGHT_CONTIG):
+	case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+	case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+	case BMAP_LEFT_FILLING | BMAP_RIGHT_CONTIG:
+	case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
+	case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+	case BMAP_LEFT_CONTIG:
+	case BMAP_RIGHT_CONTIG:
 		/*
 		 * These cases are all impossible.
 		 */
@@ -1279,14 +1192,6 @@ done:
 #undef	LEFT
 #undef	RIGHT
 #undef	PREV
-#undef	MASK
-#undef	MASK2
-#undef	MASK3
-#undef	MASK4
-#undef	STATE_SET
-#undef	STATE_TEST
-#undef	STATE_SET_TEST
-#undef	SWITCH_STATE
 }
 
 /*
@@ -1316,27 +1221,10 @@ xfs_bmap_add_extent_unwritten_real(
 	int			state = 0;/* state bits, accessed thru macros */
 	xfs_filblks_t		temp=0;
 	xfs_filblks_t		temp2=0;
-	enum {				/* bit number definitions for state */
-		LEFT_CONTIG,	RIGHT_CONTIG,
-		LEFT_FILLING,	RIGHT_FILLING,
-		LEFT_DELAY,	RIGHT_DELAY,
-		LEFT_VALID,	RIGHT_VALID
-	};
 
 #define	LEFT		r[0]
 #define	RIGHT		r[1]
 #define	PREV		r[2]
-#define	MASK(b)		(1 << (b))
-#define	MASK2(a,b)	(MASK(a) | MASK(b))
-#define	MASK3(a,b,c)	(MASK2(a,b) | MASK(c))
-#define	MASK4(a,b,c,d)	(MASK3(a,b,c) | MASK(d))
-#define	STATE_SET(b,v)	((v) ? (state |= MASK(b)) : (state &= ~MASK(b)))
-#define	STATE_TEST(b)	(state & MASK(b))
-#define	STATE_SET_TEST(b,v)	((v) ? ((state |= MASK(b)), 1) : \
-				       ((state &= ~MASK(b)), 0))
-#define	SWITCH_STATE		\
-	(state & MASK4(LEFT_FILLING, RIGHT_FILLING, LEFT_CONTIG, RIGHT_CONTIG))
-
 	/*
 	 * Set up a bunch of variables to make the tests simpler.
 	 */
@@ -1352,68 +1240,78 @@ xfs_bmap_add_extent_unwritten_real(
 	new_endoff = new->br_startoff + new->br_blockcount;
 	ASSERT(PREV.br_startoff <= new->br_startoff);
 	ASSERT(PREV.br_startoff + PREV.br_blockcount >= new_endoff);
+
 	/*
 	 * Set flags determining what part of the previous oldext allocation
 	 * extent is being replaced by a newext allocation.
 	 */
-	STATE_SET(LEFT_FILLING, PREV.br_startoff == new->br_startoff);
-	STATE_SET(RIGHT_FILLING,
-		PREV.br_startoff + PREV.br_blockcount == new_endoff);
+	if (PREV.br_startoff == new->br_startoff)
+		state |= BMAP_LEFT_FILLING;
+	if (PREV.br_startoff + PREV.br_blockcount == new_endoff)
+		state |= BMAP_RIGHT_FILLING;
+
 	/*
 	 * Check and set flags if this segment has a left neighbor.
 	 * Don't set contiguous if the combined extent would be too large.
 	 */
-	if (STATE_SET_TEST(LEFT_VALID, idx > 0)) {
+	if (idx > 0) {
+		state |= BMAP_LEFT_VALID;
 		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &LEFT);
-		STATE_SET(LEFT_DELAY, isnullstartblock(LEFT.br_startblock));
+
+		if (isnullstartblock(LEFT.br_startblock))
+			state |= BMAP_LEFT_DELAY;
 	}
-	STATE_SET(LEFT_CONTIG,
-		STATE_TEST(LEFT_VALID) && !STATE_TEST(LEFT_DELAY) &&
-		LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
-		LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
-		LEFT.br_state == newext &&
-		LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN);
+
+	if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
+	    LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
+	    LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
+	    LEFT.br_state == newext &&
+	    LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+		state |= BMAP_LEFT_CONTIG;
+
 	/*
 	 * Check and set flags if this segment has a right neighbor.
 	 * Don't set contiguous if the combined extent would be too large.
 	 * Also check for all-three-contiguous being too large.
 	 */
-	if (STATE_SET_TEST(RIGHT_VALID,
-			idx <
-			ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1)) {
+	if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1) {
+		state |= BMAP_RIGHT_VALID;
 		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx + 1), &RIGHT);
-		STATE_SET(RIGHT_DELAY, isnullstartblock(RIGHT.br_startblock));
+		if (isnullstartblock(RIGHT.br_startblock))
+			state |= BMAP_RIGHT_DELAY;
 	}
-	STATE_SET(RIGHT_CONTIG,
-		STATE_TEST(RIGHT_VALID) && !STATE_TEST(RIGHT_DELAY) &&
-		new_endoff == RIGHT.br_startoff &&
-		new->br_startblock + new->br_blockcount ==
-		    RIGHT.br_startblock &&
-		newext == RIGHT.br_state &&
-		new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
-		((state & MASK3(LEFT_CONTIG, LEFT_FILLING, RIGHT_FILLING)) !=
-		  MASK3(LEFT_CONTIG, LEFT_FILLING, RIGHT_FILLING) ||
-		 LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
-		     <= MAXEXTLEN));
+
+	if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
+	    new_endoff == RIGHT.br_startoff &&
+	    new->br_startblock + new->br_blockcount == RIGHT.br_startblock &&
+	    newext == RIGHT.br_state &&
+	    new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
+	    ((state & (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
+		       BMAP_RIGHT_FILLING)) !=
+		      (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
+		       BMAP_RIGHT_FILLING) ||
+	     LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
+			<= MAXEXTLEN))
+		state |= BMAP_RIGHT_CONTIG;
+
 	/*
 	 * Switch out based on the FILLING and CONTIG state bits.
 	 */
-	switch (SWITCH_STATE) {
-
-	case MASK4(LEFT_FILLING, RIGHT_FILLING, LEFT_CONTIG, RIGHT_CONTIG):
+	switch (state & (BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
+			 BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG)) {
+	case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG |
+	     BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
 		/*
 		 * Setting all of a previous oldext extent to newext.
 		 * The left and right neighbors are both contiguous with new.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|RF|LC|RC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
 			LEFT.br_blockcount + PREV.br_blockcount +
 			RIGHT.br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|RF|LC|RC", ip, idx - 1,
-			XFS_DATA_FORK);
-		XFS_BMAP_TRACE_DELETE("LF|RF|LC|RC", ip, idx, 2, XFS_DATA_FORK);
-		xfs_iext_remove(ifp, idx, 2);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
+		xfs_iext_remove(ip, idx, 2, state);
 		ip->i_df.if_lastex = idx - 1;
 		ip->i_d.di_nextents -= 2;
 		if (cur == NULL)
@@ -1450,20 +1348,18 @@ xfs_bmap_add_extent_unwritten_real(
 			RIGHT.br_blockcount;
 		break;
 
-	case MASK3(LEFT_FILLING, RIGHT_FILLING, LEFT_CONTIG):
+	case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
 		/*
 		 * Setting all of a previous oldext extent to newext.
 		 * The left neighbor is contiguous, the right is not.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|RF|LC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
 			LEFT.br_blockcount + PREV.br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|RF|LC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx - 1;
-		XFS_BMAP_TRACE_DELETE("LF|RF|LC", ip, idx, 1, XFS_DATA_FORK);
-		xfs_iext_remove(ifp, idx, 1);
+		xfs_iext_remove(ip, idx, 1, state);
 		ip->i_d.di_nextents--;
 		if (cur == NULL)
 			rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1492,21 +1388,18 @@ xfs_bmap_add_extent_unwritten_real(
 			PREV.br_blockcount;
 		break;
 
-	case MASK3(LEFT_FILLING, RIGHT_FILLING, RIGHT_CONTIG):
+	case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
 		/*
 		 * Setting all of a previous oldext extent to newext.
 		 * The right neighbor is contiguous, the left is not.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|RF|RC", ip, idx,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep,
 			PREV.br_blockcount + RIGHT.br_blockcount);
 		xfs_bmbt_set_state(ep, newext);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|RF|RC", ip, idx,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 		ip->i_df.if_lastex = idx;
-		XFS_BMAP_TRACE_DELETE("LF|RF|RC", ip, idx + 1, 1, XFS_DATA_FORK);
-		xfs_iext_remove(ifp, idx + 1, 1);
+		xfs_iext_remove(ip, idx + 1, 1, state);
 		ip->i_d.di_nextents--;
 		if (cur == NULL)
 			rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1535,17 +1428,16 @@ xfs_bmap_add_extent_unwritten_real(
 			RIGHT.br_blockcount;
 		break;
 
-	case MASK2(LEFT_FILLING, RIGHT_FILLING):
+	case BMAP_LEFT_FILLING | BMAP_RIGHT_FILLING:
 		/*
 		 * Setting all of a previous oldext extent to newext.
 		 * Neither the left nor right neighbors are contiguous with
 		 * the new one.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|RF", ip, idx,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_state(ep, newext);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|RF", ip, idx,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx;
 		if (cur == NULL)
 			rval = XFS_ILOG_DEXT;
@@ -1566,27 +1458,25 @@ xfs_bmap_add_extent_unwritten_real(
 		temp2 = new->br_blockcount;
 		break;
 
-	case MASK2(LEFT_FILLING, LEFT_CONTIG):
+	case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG:
 		/*
 		 * Setting the first part of a previous oldext extent to newext.
 		 * The left neighbor is contiguous.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|LC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
 			LEFT.br_blockcount + new->br_blockcount);
 		xfs_bmbt_set_startoff(ep,
 			PREV.br_startoff + new->br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|LC", ip, idx - 1,
-			XFS_DATA_FORK);
-		XFS_BMAP_TRACE_PRE_UPDATE("LF|LC", ip, idx,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_startblock(ep,
 			new->br_startblock + new->br_blockcount);
 		xfs_bmbt_set_blockcount(ep,
 			PREV.br_blockcount - new->br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF|LC", ip, idx,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx - 1;
 		if (cur == NULL)
 			rval = XFS_ILOG_DEXT;
@@ -1617,22 +1507,21 @@ xfs_bmap_add_extent_unwritten_real(
 			PREV.br_blockcount;
 		break;
 
-	case MASK(LEFT_FILLING):
+	case BMAP_LEFT_FILLING:
 		/*
 		 * Setting the first part of a previous oldext extent to newext.
 		 * The left neighbor is not contiguous.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LF", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		ASSERT(ep && xfs_bmbt_get_state(ep) == oldext);
 		xfs_bmbt_set_startoff(ep, new_endoff);
 		xfs_bmbt_set_blockcount(ep,
 			PREV.br_blockcount - new->br_blockcount);
 		xfs_bmbt_set_startblock(ep,
 			new->br_startblock + new->br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LF", ip, idx, XFS_DATA_FORK);
-		XFS_BMAP_TRACE_INSERT("LF", ip, idx, 1, new, NULL,
-			XFS_DATA_FORK);
-		xfs_iext_insert(ifp, idx, 1, new);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
+		xfs_iext_insert(ip, idx, 1, new, state);
 		ip->i_df.if_lastex = idx;
 		ip->i_d.di_nextents++;
 		if (cur == NULL)
@@ -1660,24 +1549,21 @@ xfs_bmap_add_extent_unwritten_real(
 		temp2 = PREV.br_blockcount;
 		break;
 
-	case MASK2(RIGHT_FILLING, RIGHT_CONTIG):
+	case BMAP_RIGHT_FILLING | BMAP_RIGHT_CONTIG:
 		/*
 		 * Setting the last part of a previous oldext extent to newext.
 		 * The right neighbor is contiguous with the new allocation.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("RF|RC", ip, idx,
-			XFS_DATA_FORK);
-		XFS_BMAP_TRACE_PRE_UPDATE("RF|RC", ip, idx + 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
+		trace_xfs_bmap_pre_update(ip, idx + 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep,
 			PREV.br_blockcount - new->br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("RF|RC", ip, idx,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, idx + 1),
 			new->br_startoff, new->br_startblock,
 			new->br_blockcount + RIGHT.br_blockcount, newext);
-		XFS_BMAP_TRACE_POST_UPDATE("RF|RC", ip, idx + 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx + 1, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx + 1;
 		if (cur == NULL)
 			rval = XFS_ILOG_DEXT;
@@ -1707,18 +1593,17 @@ xfs_bmap_add_extent_unwritten_real(
 			RIGHT.br_blockcount;
 		break;
 
-	case MASK(RIGHT_FILLING):
+	case BMAP_RIGHT_FILLING:
 		/*
 		 * Setting the last part of a previous oldext extent to newext.
 		 * The right neighbor is not contiguous.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("RF", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep,
 			PREV.br_blockcount - new->br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("RF", ip, idx, XFS_DATA_FORK);
-		XFS_BMAP_TRACE_INSERT("RF", ip, idx + 1, 1, new, NULL,
-			XFS_DATA_FORK);
-		xfs_iext_insert(ifp, idx + 1, 1, new);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
+		xfs_iext_insert(ip, idx + 1, 1, new, state);
 		ip->i_df.if_lastex = idx + 1;
 		ip->i_d.di_nextents++;
 		if (cur == NULL)
@@ -1756,19 +1641,18 @@ xfs_bmap_add_extent_unwritten_real(
 		 * newext.  Contiguity is impossible here.
 		 * One extent becomes three extents.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("0", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep,
 			new->br_startoff - PREV.br_startoff);
-		XFS_BMAP_TRACE_POST_UPDATE("0", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
 		r[0] = *new;
 		r[1].br_startoff = new_endoff;
 		r[1].br_blockcount =
 			PREV.br_startoff + PREV.br_blockcount - new_endoff;
 		r[1].br_startblock = new->br_startblock + new->br_blockcount;
 		r[1].br_state = oldext;
-		XFS_BMAP_TRACE_INSERT("0", ip, idx + 1, 2, &r[0], &r[1],
-			XFS_DATA_FORK);
-		xfs_iext_insert(ifp, idx + 1, 2, &r[0]);
+		xfs_iext_insert(ip, idx + 1, 2, &r[0], state);
 		ip->i_df.if_lastex = idx + 1;
 		ip->i_d.di_nextents += 2;
 		if (cur == NULL)
@@ -1813,13 +1697,13 @@ xfs_bmap_add_extent_unwritten_real(
 		temp2 = PREV.br_blockcount;
 		break;
 
-	case MASK3(LEFT_FILLING, LEFT_CONTIG, RIGHT_CONTIG):
-	case MASK3(RIGHT_FILLING, LEFT_CONTIG, RIGHT_CONTIG):
-	case MASK2(LEFT_FILLING, RIGHT_CONTIG):
-	case MASK2(RIGHT_FILLING, LEFT_CONTIG):
-	case MASK2(LEFT_CONTIG, RIGHT_CONTIG):
-	case MASK(LEFT_CONTIG):
-	case MASK(RIGHT_CONTIG):
+	case BMAP_LEFT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+	case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+	case BMAP_LEFT_FILLING | BMAP_RIGHT_CONTIG:
+	case BMAP_RIGHT_FILLING | BMAP_LEFT_CONTIG:
+	case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
+	case BMAP_LEFT_CONTIG:
+	case BMAP_RIGHT_CONTIG:
 		/*
 		 * These cases are all impossible.
 		 */
@@ -1839,14 +1723,6 @@ done:
 #undef	LEFT
 #undef	RIGHT
 #undef	PREV
-#undef	MASK
-#undef	MASK2
-#undef	MASK3
-#undef	MASK4
-#undef	STATE_SET
-#undef	STATE_TEST
-#undef	STATE_SET_TEST
-#undef	SWITCH_STATE
 }
 
 /*
@@ -1872,62 +1748,57 @@ xfs_bmap_add_extent_hole_delay(
 	int			state;  /* state bits, accessed thru macros */
 	xfs_filblks_t		temp=0;	/* temp for indirect calculations */
 	xfs_filblks_t		temp2=0;
-	enum {				/* bit number definitions for state */
-		LEFT_CONTIG,	RIGHT_CONTIG,
-		LEFT_DELAY,	RIGHT_DELAY,
-		LEFT_VALID,	RIGHT_VALID
-	};
-
-#define	MASK(b)			(1 << (b))
-#define	MASK2(a,b)		(MASK(a) | MASK(b))
-#define	STATE_SET(b,v)		((v) ? (state |= MASK(b)) : (state &= ~MASK(b)))
-#define	STATE_TEST(b)		(state & MASK(b))
-#define	STATE_SET_TEST(b,v)	((v) ? ((state |= MASK(b)), 1) : \
-				       ((state &= ~MASK(b)), 0))
-#define	SWITCH_STATE		(state & MASK2(LEFT_CONTIG, RIGHT_CONTIG))
 
 	ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
 	ep = xfs_iext_get_ext(ifp, idx);
 	state = 0;
 	ASSERT(isnullstartblock(new->br_startblock));
+
 	/*
 	 * Check and set flags if this segment has a left neighbor
 	 */
-	if (STATE_SET_TEST(LEFT_VALID, idx > 0)) {
+	if (idx > 0) {
+		state |= BMAP_LEFT_VALID;
 		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &left);
-		STATE_SET(LEFT_DELAY, isnullstartblock(left.br_startblock));
+
+		if (isnullstartblock(left.br_startblock))
+			state |= BMAP_LEFT_DELAY;
 	}
+
 	/*
 	 * Check and set flags if the current (right) segment exists.
 	 * If it doesn't exist, we're converting the hole at end-of-file.
 	 */
-	if (STATE_SET_TEST(RIGHT_VALID,
-			   idx <
-			   ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t))) {
+	if (idx < ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
+		state |= BMAP_RIGHT_VALID;
 		xfs_bmbt_get_all(ep, &right);
-		STATE_SET(RIGHT_DELAY, isnullstartblock(right.br_startblock));
+
+		if (isnullstartblock(right.br_startblock))
+			state |= BMAP_RIGHT_DELAY;
 	}
+
 	/*
 	 * Set contiguity flags on the left and right neighbors.
 	 * Don't let extents get too large, even if the pieces are contiguous.
 	 */
-	STATE_SET(LEFT_CONTIG,
-		STATE_TEST(LEFT_VALID) && STATE_TEST(LEFT_DELAY) &&
-		left.br_startoff + left.br_blockcount == new->br_startoff &&
-		left.br_blockcount + new->br_blockcount <= MAXEXTLEN);
-	STATE_SET(RIGHT_CONTIG,
-		STATE_TEST(RIGHT_VALID) && STATE_TEST(RIGHT_DELAY) &&
-		new->br_startoff + new->br_blockcount == right.br_startoff &&
-		new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
-		(!STATE_TEST(LEFT_CONTIG) ||
-		 (left.br_blockcount + new->br_blockcount +
-		     right.br_blockcount <= MAXEXTLEN)));
+	if ((state & BMAP_LEFT_VALID) && (state & BMAP_LEFT_DELAY) &&
+	    left.br_startoff + left.br_blockcount == new->br_startoff &&
+	    left.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+		state |= BMAP_LEFT_CONTIG;
+
+	if ((state & BMAP_RIGHT_VALID) && (state & BMAP_RIGHT_DELAY) &&
+	    new->br_startoff + new->br_blockcount == right.br_startoff &&
+	    new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
+	    (!(state & BMAP_LEFT_CONTIG) ||
+	     (left.br_blockcount + new->br_blockcount +
+	      right.br_blockcount <= MAXEXTLEN)))
+		state |= BMAP_RIGHT_CONTIG;
+
 	/*
 	 * Switch out based on the contiguity flags.
 	 */
-	switch (SWITCH_STATE) {
-
-	case MASK2(LEFT_CONTIG, RIGHT_CONTIG):
+	switch (state & (BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG)) {
+	case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
 		/*
 		 * New allocation is contiguous with delayed allocations
 		 * on the left and on the right.
@@ -1935,8 +1806,8 @@ xfs_bmap_add_extent_hole_delay(
 		 */
 		temp = left.br_blockcount + new->br_blockcount +
 			right.br_blockcount;
-		XFS_BMAP_TRACE_PRE_UPDATE("LC|RC", ip, idx - 1,
-			XFS_DATA_FORK);
+
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), temp);
 		oldlen = startblockval(left.br_startblock) +
 			startblockval(new->br_startblock) +
@@ -1944,53 +1815,52 @@ xfs_bmap_add_extent_hole_delay(
 		newlen = xfs_bmap_worst_indlen(ip, temp);
 		xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx - 1),
 			nullstartblock((int)newlen));
-		XFS_BMAP_TRACE_POST_UPDATE("LC|RC", ip, idx - 1,
-			XFS_DATA_FORK);
-		XFS_BMAP_TRACE_DELETE("LC|RC", ip, idx, 1, XFS_DATA_FORK);
-		xfs_iext_remove(ifp, idx, 1);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
+		xfs_iext_remove(ip, idx, 1, state);
 		ip->i_df.if_lastex = idx - 1;
 		/* DELTA: Two in-core extents were replaced by one. */
 		temp2 = temp;
 		temp = left.br_startoff;
 		break;
 
-	case MASK(LEFT_CONTIG):
+	case BMAP_LEFT_CONTIG:
 		/*
 		 * New allocation is contiguous with a delayed allocation
 		 * on the left.
 		 * Merge the new allocation with the left neighbor.
 		 */
 		temp = left.br_blockcount + new->br_blockcount;
-		XFS_BMAP_TRACE_PRE_UPDATE("LC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), temp);
 		oldlen = startblockval(left.br_startblock) +
 			startblockval(new->br_startblock);
 		newlen = xfs_bmap_worst_indlen(ip, temp);
 		xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx - 1),
 			nullstartblock((int)newlen));
-		XFS_BMAP_TRACE_POST_UPDATE("LC", ip, idx - 1,
-			XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx - 1;
 		/* DELTA: One in-core extent grew into a hole. */
 		temp2 = temp;
 		temp = left.br_startoff;
 		break;
 
-	case MASK(RIGHT_CONTIG):
+	case BMAP_RIGHT_CONTIG:
 		/*
 		 * New allocation is contiguous with a delayed allocation
 		 * on the right.
 		 * Merge the new allocation with the right neighbor.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("RC", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		temp = new->br_blockcount + right.br_blockcount;
 		oldlen = startblockval(new->br_startblock) +
 			startblockval(right.br_startblock);
 		newlen = xfs_bmap_worst_indlen(ip, temp);
 		xfs_bmbt_set_allf(ep, new->br_startoff,
 			nullstartblock((int)newlen), temp, right.br_state);
-		XFS_BMAP_TRACE_POST_UPDATE("RC", ip, idx, XFS_DATA_FORK);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
 		ip->i_df.if_lastex = idx;
 		/* DELTA: One in-core extent grew into a hole. */
 		temp2 = temp;
@@ -2004,9 +1874,7 @@ xfs_bmap_add_extent_hole_delay(
 		 * Insert a new entry.
 		 */
 		oldlen = newlen = 0;
-		XFS_BMAP_TRACE_INSERT("0", ip, idx, 1, new, NULL,
-			XFS_DATA_FORK);
-		xfs_iext_insert(ifp, idx, 1, new);
+		xfs_iext_insert(ip, idx, 1, new, state);
 		ip->i_df.if_lastex = idx;
 		/* DELTA: A new in-core extent was added in a hole. */
 		temp2 = new->br_blockcount;
@@ -2030,12 +1898,6 @@ xfs_bmap_add_extent_hole_delay(
 	}
 	*logflagsp = 0;
 	return 0;
-#undef	MASK
-#undef	MASK2
-#undef	STATE_SET
-#undef	STATE_TEST
-#undef	STATE_SET_TEST
-#undef	SWITCH_STATE
 }
 
 /*
@@ -2062,83 +1924,75 @@ xfs_bmap_add_extent_hole_real(
 	int			state;	/* state bits, accessed thru macros */
 	xfs_filblks_t		temp=0;
 	xfs_filblks_t		temp2=0;
-	enum {				/* bit number definitions for state */
-		LEFT_CONTIG,	RIGHT_CONTIG,
-		LEFT_DELAY,	RIGHT_DELAY,
-		LEFT_VALID,	RIGHT_VALID
-	};
-
-#define	MASK(b)			(1 << (b))
-#define	MASK2(a,b)		(MASK(a) | MASK(b))
-#define	STATE_SET(b,v)		((v) ? (state |= MASK(b)) : (state &= ~MASK(b)))
-#define	STATE_TEST(b)		(state & MASK(b))
-#define	STATE_SET_TEST(b,v)	((v) ? ((state |= MASK(b)), 1) : \
-				       ((state &= ~MASK(b)), 0))
-#define	SWITCH_STATE		(state & MASK2(LEFT_CONTIG, RIGHT_CONTIG))
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	ASSERT(idx <= ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t));
 	ep = xfs_iext_get_ext(ifp, idx);
 	state = 0;
+
+	if (whichfork == XFS_ATTR_FORK)
+		state |= BMAP_ATTRFORK;
+
 	/*
 	 * Check and set flags if this segment has a left neighbor.
 	 */
-	if (STATE_SET_TEST(LEFT_VALID, idx > 0)) {
+	if (idx > 0) {
+		state |= BMAP_LEFT_VALID;
 		xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &left);
-		STATE_SET(LEFT_DELAY, isnullstartblock(left.br_startblock));
+		if (isnullstartblock(left.br_startblock))
+			state |= BMAP_LEFT_DELAY;
 	}
+
 	/*
 	 * Check and set flags if this segment has a current value.
 	 * Not true if we're inserting into the "hole" at eof.
 	 */
-	if (STATE_SET_TEST(RIGHT_VALID,
-			   idx <
-			   ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t))) {
+	if (idx < ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)) {
+		state |= BMAP_RIGHT_VALID;
 		xfs_bmbt_get_all(ep, &right);
-		STATE_SET(RIGHT_DELAY, isnullstartblock(right.br_startblock));
+		if (isnullstartblock(right.br_startblock))
+			state |= BMAP_RIGHT_DELAY;
 	}
+
 	/*
 	 * We're inserting a real allocation between "left" and "right".
 	 * Set the contiguity flags.  Don't let extents get too large.
 	 */
-	STATE_SET(LEFT_CONTIG,
-		STATE_TEST(LEFT_VALID) && !STATE_TEST(LEFT_DELAY) &&
-		left.br_startoff + left.br_blockcount == new->br_startoff &&
-		left.br_startblock + left.br_blockcount == new->br_startblock &&
-		left.br_state == new->br_state &&
-		left.br_blockcount + new->br_blockcount <= MAXEXTLEN);
-	STATE_SET(RIGHT_CONTIG,
-		STATE_TEST(RIGHT_VALID) && !STATE_TEST(RIGHT_DELAY) &&
-		new->br_startoff + new->br_blockcount == right.br_startoff &&
-		new->br_startblock + new->br_blockcount ==
-		    right.br_startblock &&
-		new->br_state == right.br_state &&
-		new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
-		(!STATE_TEST(LEFT_CONTIG) ||
-		 left.br_blockcount + new->br_blockcount +
-		     right.br_blockcount <= MAXEXTLEN));
+	if ((state & BMAP_LEFT_VALID) && !(state & BMAP_LEFT_DELAY) &&
+	    left.br_startoff + left.br_blockcount == new->br_startoff &&
+	    left.br_startblock + left.br_blockcount == new->br_startblock &&
+	    left.br_state == new->br_state &&
+	    left.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+		state |= BMAP_LEFT_CONTIG;
+
+	if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
+	    new->br_startoff + new->br_blockcount == right.br_startoff &&
+	    new->br_startblock + new->br_blockcount == right.br_startblock &&
+	    new->br_state == right.br_state &&
+	    new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
+	    (!(state & BMAP_LEFT_CONTIG) ||
+	     left.br_blockcount + new->br_blockcount +
+	     right.br_blockcount <= MAXEXTLEN))
+		state |= BMAP_RIGHT_CONTIG;
 
 	error = 0;
 	/*
 	 * Select which case we're in here, and implement it.
 	 */
-	switch (SWITCH_STATE) {
-
-	case MASK2(LEFT_CONTIG, RIGHT_CONTIG):
+	switch (state & (BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG)) {
+	case BMAP_LEFT_CONTIG | BMAP_RIGHT_CONTIG:
 		/*
 		 * New allocation is contiguous with real allocations on the
 		 * left and on the right.
 		 * Merge all three into a single extent record.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LC|RC", ip, idx - 1,
-			whichfork);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
 			left.br_blockcount + new->br_blockcount +
 			right.br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LC|RC", ip, idx - 1,
-			whichfork);
-		XFS_BMAP_TRACE_DELETE("LC|RC", ip, idx, 1, whichfork);
-		xfs_iext_remove(ifp, idx, 1);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
+		xfs_iext_remove(ip, idx, 1, state);
 		ifp->if_lastex = idx - 1;
 		XFS_IFORK_NEXT_SET(ip, whichfork,
 			XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
@@ -2173,16 +2027,17 @@ xfs_bmap_add_extent_hole_real(
 			right.br_blockcount;
 		break;
 
-	case MASK(LEFT_CONTIG):
+	case BMAP_LEFT_CONTIG:
 		/*
 		 * New allocation is contiguous with a real allocation
 		 * on the left.
 		 * Merge the new allocation with the left neighbor.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("LC", ip, idx - 1, whichfork);
+		trace_xfs_bmap_pre_update(ip, idx - 1, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
 			left.br_blockcount + new->br_blockcount);
-		XFS_BMAP_TRACE_POST_UPDATE("LC", ip, idx - 1, whichfork);
+		trace_xfs_bmap_post_update(ip, idx - 1, state, _THIS_IP_);
+
 		ifp->if_lastex = idx - 1;
 		if (cur == NULL) {
 			rval = xfs_ilog_fext(whichfork);
@@ -2207,17 +2062,18 @@ xfs_bmap_add_extent_hole_real(
 			new->br_blockcount;
 		break;
 
-	case MASK(RIGHT_CONTIG):
+	case BMAP_RIGHT_CONTIG:
 		/*
 		 * New allocation is contiguous with a real allocation
 		 * on the right.
 		 * Merge the new allocation with the right neighbor.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("RC", ip, idx, whichfork);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_allf(ep, new->br_startoff, new->br_startblock,
 			new->br_blockcount + right.br_blockcount,
 			right.br_state);
-		XFS_BMAP_TRACE_POST_UPDATE("RC", ip, idx, whichfork);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+
 		ifp->if_lastex = idx;
 		if (cur == NULL) {
 			rval = xfs_ilog_fext(whichfork);
@@ -2248,8 +2104,7 @@ xfs_bmap_add_extent_hole_real(
 		 * real allocation.
 		 * Insert a new entry.
 		 */
-		XFS_BMAP_TRACE_INSERT("0", ip, idx, 1, new, NULL, whichfork);
-		xfs_iext_insert(ifp, idx, 1, new);
+		xfs_iext_insert(ip, idx, 1, new, state);
 		ifp->if_lastex = idx;
 		XFS_IFORK_NEXT_SET(ip, whichfork,
 			XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
@@ -2283,12 +2138,6 @@ xfs_bmap_add_extent_hole_real(
 done:
 	*logflagsp = rval;
 	return error;
-#undef	MASK
-#undef	MASK2
-#undef	STATE_SET
-#undef	STATE_TEST
-#undef	STATE_SET_TEST
-#undef	SWITCH_STATE
 }
 
 /*
@@ -3115,8 +2964,13 @@ xfs_bmap_del_extent(
 	uint			qfield;	/* quota field to update */
 	xfs_filblks_t		temp;	/* for indirect length calculations */
 	xfs_filblks_t		temp2;	/* for indirect length calculations */
+	int			state = 0;
 
 	XFS_STATS_INC(xs_del_exlist);
+
+	if (whichfork == XFS_ATTR_FORK)
+		state |= BMAP_ATTRFORK;
+
 	mp = ip->i_mount;
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	ASSERT((idx >= 0) && (idx < ifp->if_bytes /
@@ -3196,8 +3050,8 @@ xfs_bmap_del_extent(
 		/*
 		 * Matches the whole extent.  Delete the entry.
 		 */
-		XFS_BMAP_TRACE_DELETE("3", ip, idx, 1, whichfork);
-		xfs_iext_remove(ifp, idx, 1);
+		xfs_iext_remove(ip, idx, 1,
+				whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0);
 		ifp->if_lastex = idx;
 		if (delay)
 			break;
@@ -3217,7 +3071,7 @@ xfs_bmap_del_extent(
 		/*
 		 * Deleting the first part of the extent.
 		 */
-		XFS_BMAP_TRACE_PRE_UPDATE("2", ip, idx, whichfork);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_startoff(ep, del_endoff);
 		temp = got.br_blockcount - del->br_blockcount;
 		xfs_bmbt_set_blockcount(ep, temp);
@@ -3226,13 +3080,12 @@ xfs_bmap_del_extent(
 			temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
 				da_old);
 			xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
-			XFS_BMAP_TRACE_POST_UPDATE("2", ip, idx,
-				whichfork);
+			trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 			da_new = temp;
 			break;
 		}
 		xfs_bmbt_set_startblock(ep, del_endblock);
-		XFS_BMAP_TRACE_POST_UPDATE("2", ip, idx, whichfork);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 		if (!cur) {
 			flags |= xfs_ilog_fext(whichfork);
 			break;
@@ -3248,19 +3101,18 @@ xfs_bmap_del_extent(
 		 * Deleting the last part of the extent.
 		 */
 		temp = got.br_blockcount - del->br_blockcount;
-		XFS_BMAP_TRACE_PRE_UPDATE("1", ip, idx, whichfork);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep, temp);
 		ifp->if_lastex = idx;
 		if (delay) {
 			temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
 				da_old);
 			xfs_bmbt_set_startblock(ep, nullstartblock((int)temp));
-			XFS_BMAP_TRACE_POST_UPDATE("1", ip, idx,
-				whichfork);
+			trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 			da_new = temp;
 			break;
 		}
-		XFS_BMAP_TRACE_POST_UPDATE("1", ip, idx, whichfork);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
 		if (!cur) {
 			flags |= xfs_ilog_fext(whichfork);
 			break;
@@ -3277,7 +3129,7 @@ xfs_bmap_del_extent(
 		 * Deleting the middle of the extent.
 		 */
 		temp = del->br_startoff - got.br_startoff;
-		XFS_BMAP_TRACE_PRE_UPDATE("0", ip, idx, whichfork);
+		trace_xfs_bmap_pre_update(ip, idx, state, _THIS_IP_);
 		xfs_bmbt_set_blockcount(ep, temp);
 		new.br_startoff = del_endoff;
 		temp2 = got_endoff - del_endoff;
@@ -3364,10 +3216,8 @@ xfs_bmap_del_extent(
 				}
 			}
 		}
-		XFS_BMAP_TRACE_POST_UPDATE("0", ip, idx, whichfork);
-		XFS_BMAP_TRACE_INSERT("0", ip, idx + 1, 1, &new, NULL,
-			whichfork);
-		xfs_iext_insert(ifp, idx + 1, 1, &new);
+		trace_xfs_bmap_post_update(ip, idx, state, _THIS_IP_);
+		xfs_iext_insert(ip, idx + 1, 1, &new, state);
 		ifp->if_lastex = idx + 1;
 		break;
 	}
@@ -3687,7 +3537,9 @@ xfs_bmap_local_to_extents(
 		xfs_iext_add(ifp, 0, 1);
 		ep = xfs_iext_get_ext(ifp, 0);
 		xfs_bmbt_set_allf(ep, 0, args.fsbno, 1, XFS_EXT_NORM);
-		XFS_BMAP_TRACE_POST_UPDATE("new", ip, 0, whichfork);
+		trace_xfs_bmap_post_update(ip, 0,
+				whichfork == XFS_ATTR_FORK ? BMAP_ATTRFORK : 0,
+				_THIS_IP_);
 		XFS_IFORK_NEXT_SET(ip, whichfork, 1);
 		ip->i_d.di_nblocks = 1;
 		xfs_trans_mod_dquot_byino(tp, ip,
@@ -3800,158 +3652,6 @@ xfs_bmap_search_extents(
 	return ep;
 }
 
-
-#ifdef XFS_BMAP_TRACE
-ktrace_t	*xfs_bmap_trace_buf;
-
-/*
- * Add a bmap trace buffer entry.  Base routine for the others.
- */
-STATIC void
-xfs_bmap_trace_addentry(
-	int		opcode,		/* operation */
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry(ies) */
-	xfs_extnum_t	cnt,		/* count of entries, 1 or 2 */
-	xfs_bmbt_rec_host_t *r1,	/* first record */
-	xfs_bmbt_rec_host_t *r2,	/* second record or null */
-	int		whichfork)	/* data or attr fork */
-{
-	xfs_bmbt_rec_host_t tr2;
-
-	ASSERT(cnt == 1 || cnt == 2);
-	ASSERT(r1 != NULL);
-	if (cnt == 1) {
-		ASSERT(r2 == NULL);
-		r2 = &tr2;
-		memset(&tr2, 0, sizeof(tr2));
-	} else
-		ASSERT(r2 != NULL);
-	ktrace_enter(xfs_bmap_trace_buf,
-		(void *)(__psint_t)(opcode | (whichfork << 16)),
-		(void *)fname, (void *)desc, (void *)ip,
-		(void *)(__psint_t)idx,
-		(void *)(__psint_t)cnt,
-		(void *)(__psunsigned_t)(ip->i_ino >> 32),
-		(void *)(__psunsigned_t)(unsigned)ip->i_ino,
-		(void *)(__psunsigned_t)(r1->l0 >> 32),
-		(void *)(__psunsigned_t)(unsigned)(r1->l0),
-		(void *)(__psunsigned_t)(r1->l1 >> 32),
-		(void *)(__psunsigned_t)(unsigned)(r1->l1),
-		(void *)(__psunsigned_t)(r2->l0 >> 32),
-		(void *)(__psunsigned_t)(unsigned)(r2->l0),
-		(void *)(__psunsigned_t)(r2->l1 >> 32),
-		(void *)(__psunsigned_t)(unsigned)(r2->l1)
-		);
-	ASSERT(ip->i_xtrace);
-	ktrace_enter(ip->i_xtrace,
-		(void *)(__psint_t)(opcode | (whichfork << 16)),
-		(void *)fname, (void *)desc, (void *)ip,
-		(void *)(__psint_t)idx,
-		(void *)(__psint_t)cnt,
-		(void *)(__psunsigned_t)(ip->i_ino >> 32),
-		(void *)(__psunsigned_t)(unsigned)ip->i_ino,
-		(void *)(__psunsigned_t)(r1->l0 >> 32),
-		(void *)(__psunsigned_t)(unsigned)(r1->l0),
-		(void *)(__psunsigned_t)(r1->l1 >> 32),
-		(void *)(__psunsigned_t)(unsigned)(r1->l1),
-		(void *)(__psunsigned_t)(r2->l0 >> 32),
-		(void *)(__psunsigned_t)(unsigned)(r2->l0),
-		(void *)(__psunsigned_t)(r2->l1 >> 32),
-		(void *)(__psunsigned_t)(unsigned)(r2->l1)
-		);
-}
-
-/*
- * Add bmap trace entry prior to a call to xfs_iext_remove.
- */
-STATIC void
-xfs_bmap_trace_delete(
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry(entries) deleted */
-	xfs_extnum_t	cnt,		/* count of entries deleted, 1 or 2 */
-	int		whichfork)	/* data or attr fork */
-{
-	xfs_ifork_t	*ifp;		/* inode fork pointer */
-
-	ifp = XFS_IFORK_PTR(ip, whichfork);
-	xfs_bmap_trace_addentry(XFS_BMAP_KTRACE_DELETE, fname, desc, ip, idx,
-		cnt, xfs_iext_get_ext(ifp, idx),
-		cnt == 2 ? xfs_iext_get_ext(ifp, idx + 1) : NULL,
-		whichfork);
-}
-
-/*
- * Add bmap trace entry prior to a call to xfs_iext_insert, or
- * reading in the extents list from the disk (in the btree).
- */
-STATIC void
-xfs_bmap_trace_insert(
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry(entries) inserted */
-	xfs_extnum_t	cnt,		/* count of entries inserted, 1 or 2 */
-	xfs_bmbt_irec_t	*r1,		/* inserted record 1 */
-	xfs_bmbt_irec_t	*r2,		/* inserted record 2 or null */
-	int		whichfork)	/* data or attr fork */
-{
-	xfs_bmbt_rec_host_t tr1;	/* compressed record 1 */
-	xfs_bmbt_rec_host_t tr2;	/* compressed record 2 if needed */
-
-	xfs_bmbt_set_all(&tr1, r1);
-	if (cnt == 2) {
-		ASSERT(r2 != NULL);
-		xfs_bmbt_set_all(&tr2, r2);
-	} else {
-		ASSERT(cnt == 1);
-		ASSERT(r2 == NULL);
-	}
-	xfs_bmap_trace_addentry(XFS_BMAP_KTRACE_INSERT, fname, desc, ip, idx,
-		cnt, &tr1, cnt == 2 ? &tr2 : NULL, whichfork);
-}
-
-/*
- * Add bmap trace entry after updating an extent record in place.
- */
-STATIC void
-xfs_bmap_trace_post_update(
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry updated */
-	int		whichfork)	/* data or attr fork */
-{
-	xfs_ifork_t	*ifp;		/* inode fork pointer */
-
-	ifp = XFS_IFORK_PTR(ip, whichfork);
-	xfs_bmap_trace_addentry(XFS_BMAP_KTRACE_POST_UP, fname, desc, ip, idx,
-		1, xfs_iext_get_ext(ifp, idx), NULL, whichfork);
-}
-
-/*
- * Add bmap trace entry prior to updating an extent record in place.
- */
-STATIC void
-xfs_bmap_trace_pre_update(
-	const char	*fname,		/* function name */
-	char		*desc,		/* operation description */
-	xfs_inode_t	*ip,		/* incore inode pointer */
-	xfs_extnum_t	idx,		/* index of entry to be updated */
-	int		whichfork)	/* data or attr fork */
-{
-	xfs_ifork_t	*ifp;		/* inode fork pointer */
-
-	ifp = XFS_IFORK_PTR(ip, whichfork);
-	xfs_bmap_trace_addentry(XFS_BMAP_KTRACE_PRE_UP, fname, desc, ip, idx, 1,
-		xfs_iext_get_ext(ifp, idx), NULL, whichfork);
-}
-#endif	/* XFS_BMAP_TRACE */
-
 /*
  * Compute the worst-case number of indirect blocks that will be used
  * for ip's delayed extent of length "len".
@@ -3983,37 +3683,6 @@ xfs_bmap_worst_indlen(
 	return rval;
 }
 
-#if defined(XFS_RW_TRACE)
-STATIC void
-xfs_bunmap_trace(
-	xfs_inode_t		*ip,
-	xfs_fileoff_t		bno,
-	xfs_filblks_t		len,
-	int			flags,
-	inst_t			*ra)
-{
-	if (ip->i_rwtrace == NULL)
-		return;
-	ktrace_enter(ip->i_rwtrace,
-		(void *)(__psint_t)XFS_BUNMAP,
-		(void *)ip,
-		(void *)(__psint_t)((ip->i_d.di_size >> 32) & 0xffffffff),
-		(void *)(__psint_t)(ip->i_d.di_size & 0xffffffff),
-		(void *)(__psint_t)(((xfs_dfiloff_t)bno >> 32) & 0xffffffff),
-		(void *)(__psint_t)((xfs_dfiloff_t)bno & 0xffffffff),
-		(void *)(__psint_t)len,
-		(void *)(__psint_t)flags,
-		(void *)(unsigned long)current_cpu(),
-		(void *)ra,
-		(void *)0,
-		(void *)0,
-		(void *)0,
-		(void *)0,
-		(void *)0,
-		(void *)0);
-}
-#endif
-
 /*
  * Convert inode from non-attributed to attributed.
  * Must not be in a transaction, ip must not be locked.
@@ -4702,34 +4371,30 @@ error0:
 	return XFS_ERROR(EFSCORRUPTED);
 }
 
-#ifdef XFS_BMAP_TRACE
+#ifdef DEBUG
 /*
  * Add bmap trace insert entries for all the contents of the extent records.
  */
 void
 xfs_bmap_trace_exlist(
-	const char	*fname,		/* function name */
 	xfs_inode_t	*ip,		/* incore inode pointer */
 	xfs_extnum_t	cnt,		/* count of entries in the list */
-	int		whichfork)	/* data or attr fork */
+	int		whichfork,	/* data or attr fork */
+	unsigned long	caller_ip)
 {
-	xfs_bmbt_rec_host_t *ep;	/* current extent record */
 	xfs_extnum_t	idx;		/* extent record index */
 	xfs_ifork_t	*ifp;		/* inode fork pointer */
-	xfs_bmbt_irec_t	s;		/* file extent record */
+	int		state = 0;
+
+	if (whichfork == XFS_ATTR_FORK)
+		state |= BMAP_ATTRFORK;
 
 	ifp = XFS_IFORK_PTR(ip, whichfork);
 	ASSERT(cnt == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)));
-	for (idx = 0; idx < cnt; idx++) {
-		ep = xfs_iext_get_ext(ifp, idx);
-		xfs_bmbt_get_all(ep, &s);
-		XFS_BMAP_TRACE_INSERT("exlist", ip, idx, 1, &s, NULL,
-			whichfork);
-	}
+	for (idx = 0; idx < cnt; idx++)
+		trace_xfs_extlist(ip, idx, whichfork, caller_ip);
 }
-#endif
 
-#ifdef DEBUG
 /*
  * Validate that the bmbt_irecs being returned from bmapi are valid
  * given the callers original parameters.  Specifically check the
@@ -5478,7 +5143,8 @@ xfs_bunmapi(
 	int			rsvd;		/* OK to allocate reserved blocks */
 	xfs_fsblock_t		sum;
 
-	xfs_bunmap_trace(ip, bno, len, flags, (inst_t *)__return_address);
+	trace_xfs_bunmap(ip, bno, len, flags, _RET_IP_);
+
 	whichfork = (flags & XFS_BMAPI_ATTRFORK) ?
 		XFS_ATTR_FORK : XFS_DATA_FORK;
 	ifp = XFS_IFORK_PTR(ip, whichfork);
diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h
index 56f62d2edc35..419dafb9d87d 100644
--- a/fs/xfs/xfs_bmap.h
+++ b/fs/xfs/xfs_bmap.h
@@ -95,6 +95,21 @@ typedef	struct xfs_bmap_free
 					/* need write cache flushing and no */
 					/* additional allocation alignments */
 
+#define XFS_BMAPI_FLAGS \
+	{ XFS_BMAPI_WRITE,	"WRITE" }, \
+	{ XFS_BMAPI_DELAY,	"DELAY" }, \
+	{ XFS_BMAPI_ENTIRE,	"ENTIRE" }, \
+	{ XFS_BMAPI_METADATA,	"METADATA" }, \
+	{ XFS_BMAPI_EXACT,	"EXACT" }, \
+	{ XFS_BMAPI_ATTRFORK,	"ATTRFORK" }, \
+	{ XFS_BMAPI_ASYNC,	"ASYNC" }, \
+	{ XFS_BMAPI_RSVBLOCKS,	"RSVBLOCKS" }, \
+	{ XFS_BMAPI_PREALLOC,	"PREALLOC" }, \
+	{ XFS_BMAPI_IGSTATE,	"IGSTATE" }, \
+	{ XFS_BMAPI_CONTIG,	"CONTIG" }, \
+	{ XFS_BMAPI_CONVERT,	"CONVERT" }
+
+
 static inline int xfs_bmapi_aflag(int w)
 {
 	return (w == XFS_ATTR_FORK ? XFS_BMAPI_ATTRFORK : 0);
@@ -135,36 +150,43 @@ typedef struct xfs_bmalloca {
 	char			conv;	/* overwriting unwritten extents */
 } xfs_bmalloca_t;
 
-#if defined(__KERNEL__) && defined(XFS_BMAP_TRACE)
 /*
- * Trace operations for bmap extent tracing
+ * Flags for xfs_bmap_add_extent*.
  */
-#define	XFS_BMAP_KTRACE_DELETE	1
-#define	XFS_BMAP_KTRACE_INSERT	2
-#define	XFS_BMAP_KTRACE_PRE_UP	3
-#define	XFS_BMAP_KTRACE_POST_UP	4
-
-#define	XFS_BMAP_TRACE_SIZE	4096	/* size of global trace buffer */
-#define	XFS_BMAP_KTRACE_SIZE	32	/* size of per-inode trace buffer */
-extern ktrace_t	*xfs_bmap_trace_buf;
+#define BMAP_LEFT_CONTIG	(1 << 0)
+#define BMAP_RIGHT_CONTIG	(1 << 1)
+#define BMAP_LEFT_FILLING	(1 << 2)
+#define BMAP_RIGHT_FILLING	(1 << 3)
+#define BMAP_LEFT_DELAY		(1 << 4)
+#define BMAP_RIGHT_DELAY	(1 << 5)
+#define BMAP_LEFT_VALID		(1 << 6)
+#define BMAP_RIGHT_VALID	(1 << 7)
+#define BMAP_ATTRFORK		(1 << 8)
+
+#define XFS_BMAP_EXT_FLAGS \
+	{ BMAP_LEFT_CONTIG,	"LC" }, \
+	{ BMAP_RIGHT_CONTIG,	"RC" }, \
+	{ BMAP_LEFT_FILLING,	"LF" }, \
+	{ BMAP_RIGHT_FILLING,	"RF" }, \
+	{ BMAP_ATTRFORK,	"ATTR" }
 
 /*
  * Add bmap trace insert entries for all the contents of the extent list.
+ *
+ * Quite excessive tracing.  Only do this for debug builds.
  */
+#if defined(__KERNEL) && defined(DEBUG)
 void
 xfs_bmap_trace_exlist(
-	const char		*fname,		/* function name */
 	struct xfs_inode	*ip,		/* incore inode pointer */
 	xfs_extnum_t		cnt,		/* count of entries in list */
-	int			whichfork);	/* data or attr fork */
+	int			whichfork,
+	unsigned long		caller_ip);	/* data or attr fork */
 #define	XFS_BMAP_TRACE_EXLIST(ip,c,w)	\
-	xfs_bmap_trace_exlist(__func__,ip,c,w)
-
-#else	/* __KERNEL__ && XFS_BMAP_TRACE */
-
+	xfs_bmap_trace_exlist(ip,c,w, _THIS_IP_)
+#else
 #define	XFS_BMAP_TRACE_EXLIST(ip,c,w)
-
-#endif	/* __KERNEL__ && XFS_BMAP_TRACE */
+#endif
 
 /*
  * Convert inode from non-attributed to attributed.
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index eb7b702d0690..38751d5fac6f 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -98,8 +98,7 @@ xfs_bmdr_to_bmbt(
  * This code must be in sync with the routines xfs_bmbt_get_startoff,
  * xfs_bmbt_get_startblock, xfs_bmbt_get_blockcount and xfs_bmbt_get_state.
  */
-
-STATIC_INLINE void
+STATIC void
 __xfs_bmbt_get_all(
 		__uint64_t l0,
 		__uint64_t l1,
@@ -769,12 +768,6 @@ xfs_bmbt_trace_enter(
 		(void *)a0, (void *)a1, (void *)a2, (void *)a3,
 		(void *)a4, (void *)a5, (void *)a6, (void *)a7,
 		(void *)a8, (void *)a9, (void *)a10);
-	ktrace_enter(ip->i_btrace,
-		(void *)((__psint_t)type | (whichfork << 8) | (line << 16)),
-		(void *)func, (void *)s, (void *)ip, (void *)cur,
-		(void *)a0, (void *)a1, (void *)a2, (void *)a3,
-		(void *)a4, (void *)a5, (void *)a6, (void *)a7,
-		(void *)a8, (void *)a9, (void *)a10);
 }
 
 STATIC void
diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h
index 5549d495947f..cf07ca7c22e7 100644
--- a/fs/xfs/xfs_bmap_btree.h
+++ b/fs/xfs/xfs_bmap_btree.h
@@ -46,20 +46,12 @@ typedef struct xfs_bmdr_block {
 #define BMBT_STARTBLOCK_BITLEN	52
 #define BMBT_BLOCKCOUNT_BITLEN	21
 
-
-#define BMBT_USE_64	1
-
-typedef struct xfs_bmbt_rec_32
-{
-	__uint32_t		l0, l1, l2, l3;
-} xfs_bmbt_rec_32_t;
-typedef struct xfs_bmbt_rec_64
-{
+typedef struct xfs_bmbt_rec {
 	__be64			l0, l1;
-} xfs_bmbt_rec_64_t;
+} xfs_bmbt_rec_t;
 
 typedef __uint64_t	xfs_bmbt_rec_base_t;	/* use this for casts */
-typedef xfs_bmbt_rec_64_t xfs_bmbt_rec_t, xfs_bmdr_rec_t;
+typedef xfs_bmbt_rec_t xfs_bmdr_rec_t;
 
 typedef struct xfs_bmbt_rec_host {
 	__uint64_t		l0, l1;
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c
index 52b5f14d0c32..36a0992dd669 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/xfs_btree.c
@@ -39,6 +39,7 @@
 #include "xfs_btree_trace.h"
 #include "xfs_ialloc.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 /*
  * Cursor allocation zone.
@@ -81,7 +82,7 @@ xfs_btree_check_lblock(
 			XFS_ERRTAG_BTREE_CHECK_LBLOCK,
 			XFS_RANDOM_BTREE_CHECK_LBLOCK))) {
 		if (bp)
-			xfs_buftrace("LBTREE ERROR", bp);
+			trace_xfs_btree_corrupt(bp, _RET_IP_);
 		XFS_ERROR_REPORT("xfs_btree_check_lblock", XFS_ERRLEVEL_LOW,
 				 mp);
 		return XFS_ERROR(EFSCORRUPTED);
@@ -119,7 +120,7 @@ xfs_btree_check_sblock(
 			XFS_ERRTAG_BTREE_CHECK_SBLOCK,
 			XFS_RANDOM_BTREE_CHECK_SBLOCK))) {
 		if (bp)
-			xfs_buftrace("SBTREE ERROR", bp);
+			trace_xfs_btree_corrupt(bp, _RET_IP_);
 		XFS_CORRUPTION_ERROR("xfs_btree_check_sblock",
 			XFS_ERRLEVEL_LOW, cur->bc_mp, block);
 		return XFS_ERROR(EFSCORRUPTED);
diff --git a/fs/xfs/xfs_btree_trace.h b/fs/xfs/xfs_btree_trace.h
index b3f5eb3c3c6c..2d8a309873ea 100644
--- a/fs/xfs/xfs_btree_trace.h
+++ b/fs/xfs/xfs_btree_trace.h
@@ -58,8 +58,6 @@ void xfs_btree_trace_argbi(const char *, struct xfs_btree_cur *,
 		struct xfs_buf *, int, int);
 void xfs_btree_trace_argbii(const char *, struct xfs_btree_cur *,
 		struct xfs_buf *, int, int, int);
-void xfs_btree_trace_argfffi(const char *, struct xfs_btree_cur *,
-		xfs_dfiloff_t, xfs_dfsbno_t, xfs_dfilblks_t, int, int);
 void xfs_btree_trace_argi(const char *, struct xfs_btree_cur *, int, int);
 void xfs_btree_trace_argipk(const char *, struct xfs_btree_cur *, int,
 		union xfs_btree_ptr, union xfs_btree_key *, int);
@@ -71,24 +69,10 @@ void xfs_btree_trace_argr(const char *, struct xfs_btree_cur *,
 		union xfs_btree_rec *, int);
 void xfs_btree_trace_cursor(const char *, struct xfs_btree_cur *, int, int);
 
-
-#define XFS_ALLOCBT_TRACE_SIZE	4096	/* size of global trace buffer */
-extern ktrace_t	*xfs_allocbt_trace_buf;
-
-#define XFS_INOBT_TRACE_SIZE	4096	/* size of global trace buffer */
-extern ktrace_t	*xfs_inobt_trace_buf;
-
-#define XFS_BMBT_TRACE_SIZE	4096	/* size of global trace buffer */
-#define XFS_BMBT_KTRACE_SIZE	32	/* size of per-inode trace buffer */
-extern ktrace_t	*xfs_bmbt_trace_buf;
-
-
 #define	XFS_BTREE_TRACE_ARGBI(c, b, i)	\
 	xfs_btree_trace_argbi(__func__, c, b, i, __LINE__)
 #define	XFS_BTREE_TRACE_ARGBII(c, b, i, j)	\
 	xfs_btree_trace_argbii(__func__, c, b, i, j, __LINE__)
-#define	XFS_BTREE_TRACE_ARGFFFI(c, o, b, i, j)	\
-	xfs_btree_trace_argfffi(__func__, c, o, b, i, j, __LINE__)
 #define	XFS_BTREE_TRACE_ARGI(c, i)	\
 	xfs_btree_trace_argi(__func__, c, i, __LINE__)
 #define	XFS_BTREE_TRACE_ARGIPK(c, i, p, k)	\
@@ -104,7 +88,6 @@ extern ktrace_t	*xfs_bmbt_trace_buf;
 #else
 #define	XFS_BTREE_TRACE_ARGBI(c, b, i)
 #define	XFS_BTREE_TRACE_ARGBII(c, b, i, j)
-#define	XFS_BTREE_TRACE_ARGFFFI(c, o, b, i, j)
 #define	XFS_BTREE_TRACE_ARGI(c, i)
 #define	XFS_BTREE_TRACE_ARGIPK(c, i, p, s)
 #define	XFS_BTREE_TRACE_ARGIPR(c, i, p, r)
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 92af4098c7e8..a30f7e9eb2b9 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -29,6 +29,7 @@
 #include "xfs_buf_item.h"
 #include "xfs_trans_priv.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 
 kmem_zone_t	*xfs_buf_item_zone;
@@ -164,7 +165,7 @@ xfs_buf_item_size(
 		 * is the buf log format structure with the
 		 * cancel flag in it.
 		 */
-		xfs_buf_item_trace("SIZE STALE", bip);
+		trace_xfs_buf_item_size_stale(bip);
 		ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
 		return 1;
 	}
@@ -206,7 +207,7 @@ xfs_buf_item_size(
 		}
 	}
 
-	xfs_buf_item_trace("SIZE NORM", bip);
+	trace_xfs_buf_item_size(bip);
 	return nvecs;
 }
 
@@ -259,7 +260,7 @@ xfs_buf_item_format(
 		 * is the buf log format structure with the
 		 * cancel flag in it.
 		 */
-		xfs_buf_item_trace("FORMAT STALE", bip);
+		trace_xfs_buf_item_format_stale(bip);
 		ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
 		bip->bli_format.blf_size = nvecs;
 		return;
@@ -335,7 +336,7 @@ xfs_buf_item_format(
 	/*
 	 * Check to make sure everything is consistent.
 	 */
-	xfs_buf_item_trace("FORMAT NORM", bip);
+	trace_xfs_buf_item_format(bip);
 	xfs_buf_item_log_check(bip);
 }
 
@@ -355,8 +356,7 @@ xfs_buf_item_pin(
 	ASSERT(atomic_read(&bip->bli_refcount) > 0);
 	ASSERT((bip->bli_flags & XFS_BLI_LOGGED) ||
 	       (bip->bli_flags & XFS_BLI_STALE));
-	xfs_buf_item_trace("PIN", bip);
-	xfs_buftrace("XFS_PIN", bp);
+	trace_xfs_buf_item_pin(bip);
 	xfs_bpin(bp);
 }
 
@@ -383,8 +383,7 @@ xfs_buf_item_unpin(
 	ASSERT(bp != NULL);
 	ASSERT(XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *) == bip);
 	ASSERT(atomic_read(&bip->bli_refcount) > 0);
-	xfs_buf_item_trace("UNPIN", bip);
-	xfs_buftrace("XFS_UNPIN", bp);
+	trace_xfs_buf_item_unpin(bip);
 
 	freed = atomic_dec_and_test(&bip->bli_refcount);
 	ailp = bip->bli_item.li_ailp;
@@ -395,8 +394,8 @@ xfs_buf_item_unpin(
 		ASSERT(!(XFS_BUF_ISDELAYWRITE(bp)));
 		ASSERT(XFS_BUF_ISSTALE(bp));
 		ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
-		xfs_buf_item_trace("UNPIN STALE", bip);
-		xfs_buftrace("XFS_UNPIN STALE", bp);
+		trace_xfs_buf_item_unpin_stale(bip);
+
 		/*
 		 * If we get called here because of an IO error, we may
 		 * or may not have the item on the AIL. xfs_trans_ail_delete()
@@ -440,8 +439,8 @@ xfs_buf_item_unpin_remove(
 	if ((atomic_read(&bip->bli_refcount) == 1) &&
 	    (bip->bli_flags & XFS_BLI_STALE)) {
 		ASSERT(XFS_BUF_VALUSEMA(bip->bli_buf) <= 0);
-		xfs_buf_item_trace("UNPIN REMOVE", bip);
-		xfs_buftrace("XFS_UNPIN_REMOVE", bp);
+		trace_xfs_buf_item_unpin_stale(bip);
+
 		/*
 		 * yes -- clear the xaction descriptor in-use flag
 		 * and free the chunk if required.  We can safely
@@ -495,7 +494,7 @@ xfs_buf_item_trylock(
 	XFS_BUF_HOLD(bp);
 
 	ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
-	xfs_buf_item_trace("TRYLOCK SUCCESS", bip);
+	trace_xfs_buf_item_trylock(bip);
 	return XFS_ITEM_SUCCESS;
 }
 
@@ -524,7 +523,6 @@ xfs_buf_item_unlock(
 	uint		hold;
 
 	bp = bip->bli_buf;
-	xfs_buftrace("XFS_UNLOCK", bp);
 
 	/*
 	 * Clear the buffer's association with this transaction.
@@ -547,7 +545,7 @@ xfs_buf_item_unlock(
 	 */
 	if (bip->bli_flags & XFS_BLI_STALE) {
 		bip->bli_flags &= ~XFS_BLI_LOGGED;
-		xfs_buf_item_trace("UNLOCK STALE", bip);
+		trace_xfs_buf_item_unlock_stale(bip);
 		ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
 		if (!aborted)
 			return;
@@ -574,7 +572,7 @@ xfs_buf_item_unlock(
 	 * release the buffer at the end of this routine.
 	 */
 	hold = bip->bli_flags & XFS_BLI_HOLD;
-	xfs_buf_item_trace("UNLOCK", bip);
+	trace_xfs_buf_item_unlock(bip);
 
 	/*
 	 * If the buf item isn't tracking any data, free it.
@@ -618,7 +616,8 @@ xfs_buf_item_committed(
 	xfs_buf_log_item_t	*bip,
 	xfs_lsn_t		lsn)
 {
-	xfs_buf_item_trace("COMMITTED", bip);
+	trace_xfs_buf_item_committed(bip);
+
 	if ((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
 	    (bip->bli_item.li_lsn != 0)) {
 		return bip->bli_item.li_lsn;
@@ -640,7 +639,7 @@ xfs_buf_item_push(
 	xfs_buf_t	*bp;
 
 	ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
-	xfs_buf_item_trace("PUSH", bip);
+	trace_xfs_buf_item_push(bip);
 
 	bp = bip->bli_buf;
 
@@ -738,9 +737,6 @@ xfs_buf_item_init(
 	bip->bli_format.blf_blkno = (__int64_t)XFS_BUF_ADDR(bp);
 	bip->bli_format.blf_len = (ushort)BTOBB(XFS_BUF_COUNT(bp));
 	bip->bli_format.blf_map_size = map_size;
-#ifdef XFS_BLI_TRACE
-	bip->bli_trace = ktrace_alloc(XFS_BLI_TRACE_SIZE, KM_NOFS);
-#endif
 
 #ifdef XFS_TRANS_DEBUG
 	/*
@@ -878,9 +874,6 @@ xfs_buf_item_free(
 	kmem_free(bip->bli_logged);
 #endif /* XFS_TRANS_DEBUG */
 
-#ifdef XFS_BLI_TRACE
-	ktrace_free(bip->bli_trace);
-#endif
 	kmem_zone_free(xfs_buf_item_zone, bip);
 }
 
@@ -897,7 +890,8 @@ xfs_buf_item_relse(
 {
 	xfs_buf_log_item_t	*bip;
 
-	xfs_buftrace("XFS_RELSE", bp);
+	trace_xfs_buf_item_relse(bp, _RET_IP_);
+
 	bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*);
 	XFS_BUF_SET_FSPRIVATE(bp, bip->bli_item.li_bio_list);
 	if ((XFS_BUF_FSPRIVATE(bp, void *) == NULL) &&
@@ -994,7 +988,7 @@ xfs_buf_iodone_callbacks(
 		if (XFS_FORCED_SHUTDOWN(mp)) {
 			ASSERT(XFS_BUF_TARGET(bp) == mp->m_ddev_targp);
 			XFS_BUF_SUPER_STALE(bp);
-			xfs_buftrace("BUF_IODONE_CB", bp);
+			trace_xfs_buf_item_iodone(bp, _RET_IP_);
 			xfs_buf_do_callbacks(bp, lip);
 			XFS_BUF_SET_FSPRIVATE(bp, NULL);
 			XFS_BUF_CLR_IODONE_FUNC(bp);
@@ -1030,7 +1024,7 @@ xfs_buf_iodone_callbacks(
 				XFS_BUF_SET_START(bp);
 			}
 			ASSERT(XFS_BUF_IODONE_FUNC(bp));
-			xfs_buftrace("BUF_IODONE ASYNC", bp);
+			trace_xfs_buf_item_iodone_async(bp, _RET_IP_);
 			xfs_buf_relse(bp);
 		} else {
 			/*
@@ -1053,9 +1047,7 @@ xfs_buf_iodone_callbacks(
 		}
 		return;
 	}
-#ifdef XFSERRORDEBUG
-	xfs_buftrace("XFS BUFCB NOERR", bp);
-#endif
+
 	xfs_buf_do_callbacks(bp, lip);
 	XFS_BUF_SET_FSPRIVATE(bp, NULL);
 	XFS_BUF_CLR_IODONE_FUNC(bp);
@@ -1081,7 +1073,9 @@ xfs_buf_error_relse(
 	XFS_BUF_DONE(bp);
 	XFS_BUF_UNDELAYWRITE(bp);
 	XFS_BUF_ERROR(bp,0);
-	xfs_buftrace("BUF_ERROR_RELSE", bp);
+
+	trace_xfs_buf_error_relse(bp, _RET_IP_);
+
 	if (! XFS_FORCED_SHUTDOWN(mp))
 		xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
 	/*
@@ -1128,34 +1122,3 @@ xfs_buf_iodone(
 	xfs_trans_ail_delete(ailp, (xfs_log_item_t *)bip);
 	xfs_buf_item_free(bip);
 }
-
-#if defined(XFS_BLI_TRACE)
-void
-xfs_buf_item_trace(
-	char			*id,
-	xfs_buf_log_item_t	*bip)
-{
-	xfs_buf_t		*bp;
-	ASSERT(bip->bli_trace != NULL);
-
-	bp = bip->bli_buf;
-	ktrace_enter(bip->bli_trace,
-		     (void *)id,
-		     (void *)bip->bli_buf,
-		     (void *)((unsigned long)bip->bli_flags),
-		     (void *)((unsigned long)bip->bli_recur),
-		     (void *)((unsigned long)atomic_read(&bip->bli_refcount)),
-		     (void *)((unsigned long)
-				(0xFFFFFFFF & XFS_BUF_ADDR(bp) >> 32)),
-		     (void *)((unsigned long)(0xFFFFFFFF & XFS_BUF_ADDR(bp))),
-		     (void *)((unsigned long)XFS_BUF_COUNT(bp)),
-		     (void *)((unsigned long)XFS_BUF_BFLAGS(bp)),
-		     XFS_BUF_FSPRIVATE(bp, void *),
-		     XFS_BUF_FSPRIVATE2(bp, void *),
-		     (void *)(unsigned long)XFS_BUF_ISPINNED(bp),
-		     (void *)XFS_BUF_IODONE_FUNC(bp),
-		     (void *)((unsigned long)(XFS_BUF_VALUSEMA(bp))),
-		     (void *)bip->bli_item.li_desc,
-		     (void *)((unsigned long)bip->bli_item.li_flags));
-}
-#endif /* XFS_BLI_TRACE */
diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h
index 5a41c348bb1c..217f34af00cb 100644
--- a/fs/xfs/xfs_buf_item.h
+++ b/fs/xfs/xfs_buf_item.h
@@ -70,22 +70,21 @@ typedef struct xfs_buf_log_format_t {
 #define	XFS_BLI_INODE_ALLOC_BUF	0x10
 #define XFS_BLI_STALE_INODE	0x20
 
+#define XFS_BLI_FLAGS \
+	{ XFS_BLI_HOLD,		"HOLD" }, \
+	{ XFS_BLI_DIRTY,	"DIRTY" }, \
+	{ XFS_BLI_STALE,	"STALE" }, \
+	{ XFS_BLI_LOGGED,	"LOGGED" }, \
+	{ XFS_BLI_INODE_ALLOC_BUF, "INODE_ALLOC" }, \
+	{ XFS_BLI_STALE_INODE,	"STALE_INODE" }
+
 
 #ifdef __KERNEL__
 
 struct xfs_buf;
-struct ktrace;
 struct xfs_mount;
 struct xfs_buf_log_item;
 
-#if defined(XFS_BLI_TRACE)
-#define	XFS_BLI_TRACE_SIZE	32
-
-void	xfs_buf_item_trace(char *, struct xfs_buf_log_item *);
-#else
-#define	xfs_buf_item_trace(id, bip)
-#endif
-
 /*
  * This is the in core log item structure used to track information
  * needed to log buffers.  It tracks how many times the lock has been
@@ -97,9 +96,6 @@ typedef struct xfs_buf_log_item {
 	unsigned int		bli_flags;	/* misc flags */
 	unsigned int		bli_recur;	/* lock recursion count */
 	atomic_t		bli_refcount;	/* cnt of tp refs */
-#ifdef XFS_BLI_TRACE
-	struct ktrace		*bli_trace;	/* event trace buf */
-#endif
 #ifdef XFS_TRANS_DEBUG
 	char			*bli_orig;	/* original buffer copy */
 	char			*bli_logged;	/* bytes logged (bitmap) */
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c
index 2847bbc1c534..c0c8869115b1 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/xfs_da_btree.c
@@ -46,6 +46,7 @@
 #include "xfs_dir2_block.h"
 #include "xfs_dir2_node.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 /*
  * xfs_da_btree.c
@@ -2107,7 +2108,7 @@ xfs_da_do_buf(
 				   (be32_to_cpu(free->hdr.magic) != XFS_DIR2_FREE_MAGIC),
 				mp, XFS_ERRTAG_DA_READ_BUF,
 				XFS_RANDOM_DA_READ_BUF))) {
-			xfs_buftrace("DA READ ERROR", rbp->bps[0]);
+			trace_xfs_da_btree_corrupt(rbp->bps[0], _RET_IP_);
 			XFS_CORRUPTION_ERROR("xfs_da_do_buf(2)",
 					     XFS_ERRLEVEL_LOW, mp, info);
 			error = XFS_ERROR(EFSCORRUPTED);
diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h
index 8c536167bf75..30cd08f56a3a 100644
--- a/fs/xfs/xfs_da_btree.h
+++ b/fs/xfs/xfs_da_btree.h
@@ -125,6 +125,13 @@ typedef struct xfs_da_args {
 #define XFS_DA_OP_OKNOENT	0x0008	/* lookup/add op, ENOENT ok, else die */
 #define XFS_DA_OP_CILOOKUP	0x0010	/* lookup to return CI name if found */
 
+#define XFS_DA_OP_FLAGS \
+	{ XFS_DA_OP_JUSTCHECK,	"JUSTCHECK" }, \
+	{ XFS_DA_OP_RENAME,	"RENAME" }, \
+	{ XFS_DA_OP_ADDNAME,	"ADDNAME" }, \
+	{ XFS_DA_OP_OKNOENT,	"OKNOENT" }, \
+	{ XFS_DA_OP_CILOOKUP,	"CILOOKUP" }
+
 /*
  * Structure to describe buffer(s) for a block.
  * This is needed in the directory version 2 format case, when
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index ab89a7e94a0f..84ca1cf16a1e 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -43,6 +43,7 @@
 #include "xfs_error.h"
 #include "xfs_rw.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 /*
  * Syssgi interface for swapext
@@ -113,10 +114,82 @@ xfs_swapext(
 	return error;
 }
 
+/*
+ * We need to check that the format of the data fork in the temporary inode is
+ * valid for the target inode before doing the swap. This is not a problem with
+ * attr1 because of the fixed fork offset, but attr2 has a dynamically sized
+ * data fork depending on the space the attribute fork is taking so we can get
+ * invalid formats on the target inode.
+ *
+ * E.g. target has space for 7 extents in extent format, temp inode only has
+ * space for 6.  If we defragment down to 7 extents, then the tmp format is a
+ * btree, but when swapped it needs to be in extent format. Hence we can't just
+ * blindly swap data forks on attr2 filesystems.
+ *
+ * Note that we check the swap in both directions so that we don't end up with
+ * a corrupt temporary inode, either.
+ *
+ * Note that fixing the way xfs_fsr sets up the attribute fork in the source
+ * inode will prevent this situation from occurring, so all we do here is
+ * reject and log the attempt. basically we are putting the responsibility on
+ * userspace to get this right.
+ */
+static int
+xfs_swap_extents_check_format(
+	xfs_inode_t	*ip,	/* target inode */
+	xfs_inode_t	*tip)	/* tmp inode */
+{
+
+	/* Should never get a local format */
+	if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
+	    tip->i_d.di_format == XFS_DINODE_FMT_LOCAL)
+		return EINVAL;
+
+	/*
+	 * if the target inode has less extents that then temporary inode then
+	 * why did userspace call us?
+	 */
+	if (ip->i_d.di_nextents < tip->i_d.di_nextents)
+		return EINVAL;
+
+	/*
+	 * if the target inode is in extent form and the temp inode is in btree
+	 * form then we will end up with the target inode in the wrong format
+	 * as we already know there are less extents in the temp inode.
+	 */
+	if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+	    tip->i_d.di_format == XFS_DINODE_FMT_BTREE)
+		return EINVAL;
+
+	/* Check temp in extent form to max in target */
+	if (tip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+	    XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) > ip->i_df.if_ext_max)
+		return EINVAL;
+
+	/* Check target in extent form to max in temp */
+	if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+	    XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > tip->i_df.if_ext_max)
+		return EINVAL;
+
+	/* Check root block of temp in btree form to max in target */
+	if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
+	    XFS_IFORK_BOFF(ip) &&
+	    tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip))
+		return EINVAL;
+
+	/* Check root block of target in btree form to max in temp */
+	if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
+	    XFS_IFORK_BOFF(tip) &&
+	    ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip))
+		return EINVAL;
+
+	return 0;
+}
+
 int
 xfs_swap_extents(
-	xfs_inode_t	*ip,
-	xfs_inode_t	*tip,
+	xfs_inode_t	*ip,	/* target inode */
+	xfs_inode_t	*tip,	/* tmp inode */
 	xfs_swapext_t	*sxp)
 {
 	xfs_mount_t	*mp;
@@ -160,15 +233,7 @@ xfs_swap_extents(
 		goto out_unlock;
 	}
 
-	/* Should never get a local format */
-	if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
-	    tip->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
-		error = XFS_ERROR(EINVAL);
-		goto out_unlock;
-	}
-
 	if (VN_CACHED(VFS_I(tip)) != 0) {
-		xfs_inval_cached_trace(tip, 0, -1, 0, -1);
 		error = xfs_flushinval_pages(tip, 0, -1,
 				FI_REMAPF_LOCKED);
 		if (error)
@@ -189,13 +254,12 @@ xfs_swap_extents(
 		goto out_unlock;
 	}
 
-	/*
-	 * If the target has extended attributes, the tmp file
-	 * must also in order to ensure the correct data fork
-	 * format.
-	 */
-	if ( XFS_IFORK_Q(ip) != XFS_IFORK_Q(tip) ) {
-		error = XFS_ERROR(EINVAL);
+	/* check inode formats now that data is flushed */
+	error = xfs_swap_extents_check_format(ip, tip);
+	if (error) {
+		xfs_fs_cmn_err(CE_NOTE, mp,
+		    "%s: inode 0x%llx format is incompatible for exchanging.",
+				__FILE__, ip->i_ino);
 		goto out_unlock;
 	}
 
@@ -276,6 +340,16 @@ xfs_swap_extents(
 	*tifp = *tempifp;	/* struct copy */
 
 	/*
+	 * Fix the in-memory data fork values that are dependent on the fork
+	 * offset in the inode. We can't assume they remain the same as attr2
+	 * has dynamic fork offsets.
+	 */
+	ifp->if_ext_max = XFS_IFORK_SIZE(ip, XFS_DATA_FORK) /
+					(uint)sizeof(xfs_bmbt_rec_t);
+	tifp->if_ext_max = XFS_IFORK_SIZE(tip, XFS_DATA_FORK) /
+					(uint)sizeof(xfs_bmbt_rec_t);
+
+	/*
 	 * Fix the on-disk inode values
 	 */
 	tmp = (__uint64_t)ip->i_d.di_nblocks;
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c
index bb1d58eb3982..93634a7e90e9 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/xfs_dir2.c
@@ -40,9 +40,9 @@
 #include "xfs_dir2_leaf.h"
 #include "xfs_dir2_block.h"
 #include "xfs_dir2_node.h"
-#include "xfs_dir2_trace.h"
 #include "xfs_error.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 struct xfs_name xfs_name_dotdot = {"..", 2};
 
@@ -525,7 +525,8 @@ xfs_dir2_grow_inode(
 	xfs_trans_t	*tp;
 	xfs_drfsbno_t	nblks;
 
-	xfs_dir2_trace_args_s("grow_inode", args, space);
+	trace_xfs_dir2_grow_inode(args, space);
+
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
@@ -703,7 +704,8 @@ xfs_dir2_shrink_inode(
 	xfs_mount_t	*mp;
 	xfs_trans_t	*tp;
 
-	xfs_dir2_trace_args_db("shrink_inode", args, db, bp);
+	trace_xfs_dir2_shrink_inode(args, db);
+
 	dp = args->dp;
 	mp = dp->i_mount;
 	tp = args->trans;
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c
index ab52e9e1c1ee..ddc4ecc7807f 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/xfs_dir2_block.c
@@ -36,8 +36,8 @@
 #include "xfs_dir2_data.h"
 #include "xfs_dir2_leaf.h"
 #include "xfs_dir2_block.h"
-#include "xfs_dir2_trace.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 /*
  * Local function prototypes.
@@ -94,7 +94,8 @@ xfs_dir2_block_addname(
 	__be16			*tagp;		/* pointer to tag value */
 	xfs_trans_t		*tp;		/* transaction structure */
 
-	xfs_dir2_trace_args("block_addname", args);
+	trace_xfs_dir2_block_addname(args);
+
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
@@ -590,7 +591,8 @@ xfs_dir2_block_lookup(
 	int			error;		/* error return value */
 	xfs_mount_t		*mp;		/* filesystem mount point */
 
-	xfs_dir2_trace_args("block_lookup", args);
+	trace_xfs_dir2_block_lookup(args);
+
 	/*
 	 * Get the buffer, look up the entry.
 	 * If not found (ENOENT) then return, have no buffer.
@@ -747,7 +749,8 @@ xfs_dir2_block_removename(
 	int			size;		/* shortform size */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args("block_removename", args);
+	trace_xfs_dir2_block_removename(args);
+
 	/*
 	 * Look up the entry in the block.  Gets the buffer and entry index.
 	 * It will always be there, the vnodeops level does a lookup first.
@@ -823,7 +826,8 @@ xfs_dir2_block_replace(
 	int			error;		/* error return value */
 	xfs_mount_t		*mp;		/* filesystem mount point */
 
-	xfs_dir2_trace_args("block_replace", args);
+	trace_xfs_dir2_block_replace(args);
+
 	/*
 	 * Lookup the entry in the directory.  Get buffer and entry index.
 	 * This will always succeed since the caller has already done a lookup.
@@ -897,7 +901,8 @@ xfs_dir2_leaf_to_block(
 	int			to;		/* block/leaf to index */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args_bb("leaf_to_block", args, lbp, dbp);
+	trace_xfs_dir2_leaf_to_block(args);
+
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
@@ -1044,7 +1049,8 @@ xfs_dir2_sf_to_block(
 	xfs_trans_t		*tp;		/* transaction pointer */
 	struct xfs_name		name;
 
-	xfs_dir2_trace_args("sf_to_block", args);
+	trace_xfs_dir2_sf_to_block(args);
+
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c
index 41ad537c49e9..29f484c11b3a 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/xfs_dir2_leaf.c
@@ -38,8 +38,8 @@
 #include "xfs_dir2_leaf.h"
 #include "xfs_dir2_block.h"
 #include "xfs_dir2_node.h"
-#include "xfs_dir2_trace.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 /*
  * Local function declarations.
@@ -80,7 +80,8 @@ xfs_dir2_block_to_leaf(
 	int			needscan;	/* need to rescan bestfree */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args_b("block_to_leaf", args, dbp);
+	trace_xfs_dir2_block_to_leaf(args);
+
 	dp = args->dp;
 	mp = dp->i_mount;
 	tp = args->trans;
@@ -188,7 +189,8 @@ xfs_dir2_leaf_addname(
 	xfs_trans_t		*tp;		/* transaction pointer */
 	xfs_dir2_db_t		use_block;	/* data block number */
 
-	xfs_dir2_trace_args("leaf_addname", args);
+	trace_xfs_dir2_leaf_addname(args);
+
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
@@ -1266,7 +1268,8 @@ xfs_dir2_leaf_lookup(
 	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args("leaf_lookup", args);
+	trace_xfs_dir2_leaf_lookup(args);
+
 	/*
 	 * Look up name in the leaf block, returning both buffers and index.
 	 */
@@ -1454,7 +1457,8 @@ xfs_dir2_leaf_removename(
 	xfs_dir2_data_off_t	oldbest;	/* old value of best free */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args("leaf_removename", args);
+	trace_xfs_dir2_leaf_removename(args);
+
 	/*
 	 * Lookup the leaf entry, get the leaf and data blocks read in.
 	 */
@@ -1586,7 +1590,8 @@ xfs_dir2_leaf_replace(
 	xfs_dir2_leaf_entry_t	*lep;		/* leaf entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args("leaf_replace", args);
+	trace_xfs_dir2_leaf_replace(args);
+
 	/*
 	 * Look up the entry.
 	 */
@@ -1766,7 +1771,9 @@ xfs_dir2_node_to_leaf(
 	if (state->path.active > 1)
 		return 0;
 	args = state->args;
-	xfs_dir2_trace_args("node_to_leaf", args);
+
+	trace_xfs_dir2_node_to_leaf(args);
+
 	mp = state->mp;
 	dp = args->dp;
 	tp = args->trans;
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c
index 5a81ccd1045b..ce6e355199b5 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/xfs_dir2_node.c
@@ -37,8 +37,8 @@
 #include "xfs_dir2_leaf.h"
 #include "xfs_dir2_block.h"
 #include "xfs_dir2_node.h"
-#include "xfs_dir2_trace.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 /*
  * Function declarations.
@@ -123,7 +123,8 @@ xfs_dir2_leaf_to_node(
 	__be16			*to;		/* pointer to freespace entry */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args_b("leaf_to_node", args, lbp);
+	trace_xfs_dir2_leaf_to_node(args);
+
 	dp = args->dp;
 	mp = dp->i_mount;
 	tp = args->trans;
@@ -196,7 +197,8 @@ xfs_dir2_leafn_add(
 	xfs_mount_t		*mp;		/* filesystem mount point */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args_sb("leafn_add", args, index, bp);
+	trace_xfs_dir2_leafn_add(args, index);
+
 	dp = args->dp;
 	mp = dp->i_mount;
 	tp = args->trans;
@@ -711,8 +713,8 @@ xfs_dir2_leafn_moveents(
 	int		stale;			/* count stale leaves copied */
 	xfs_trans_t	*tp;			/* transaction pointer */
 
-	xfs_dir2_trace_args_bibii("leafn_moveents", args, bp_s, start_s, bp_d,
-		start_d, count);
+	trace_xfs_dir2_leafn_moveents(args, start_s, start_d, count);
+
 	/*
 	 * Silently return if nothing to do.
 	 */
@@ -933,7 +935,8 @@ xfs_dir2_leafn_remove(
 	int			needscan;	/* need to rescan data frees */
 	xfs_trans_t		*tp;		/* transaction pointer */
 
-	xfs_dir2_trace_args_sb("leafn_remove", args, index, bp);
+	trace_xfs_dir2_leafn_remove(args, index);
+
 	dp = args->dp;
 	tp = args->trans;
 	mp = dp->i_mount;
@@ -1363,7 +1366,8 @@ xfs_dir2_node_addname(
 	int			rval;		/* sub-return value */
 	xfs_da_state_t		*state;		/* btree cursor */
 
-	xfs_dir2_trace_args("node_addname", args);
+	trace_xfs_dir2_node_addname(args);
+
 	/*
 	 * Allocate and initialize the state (btree cursor).
 	 */
@@ -1822,7 +1826,8 @@ xfs_dir2_node_lookup(
 	int		rval;			/* operation return value */
 	xfs_da_state_t	*state;			/* btree cursor */
 
-	xfs_dir2_trace_args("node_lookup", args);
+	trace_xfs_dir2_node_lookup(args);
+
 	/*
 	 * Allocate and initialize the btree cursor.
 	 */
@@ -1875,7 +1880,8 @@ xfs_dir2_node_removename(
 	int			rval;		/* operation return value */
 	xfs_da_state_t		*state;		/* btree cursor */
 
-	xfs_dir2_trace_args("node_removename", args);
+	trace_xfs_dir2_node_removename(args);
+
 	/*
 	 * Allocate and initialize the btree cursor.
 	 */
@@ -1944,7 +1950,8 @@ xfs_dir2_node_replace(
 	int			rval;		/* internal return value */
 	xfs_da_state_t		*state;		/* btree cursor */
 
-	xfs_dir2_trace_args("node_replace", args);
+	trace_xfs_dir2_node_replace(args);
+
 	/*
 	 * Allocate and initialize the btree cursor.
 	 */
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c
index e89734e84646..9d4f17a69676 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/xfs_dir2_sf.c
@@ -37,7 +37,7 @@
 #include "xfs_dir2_data.h"
 #include "xfs_dir2_leaf.h"
 #include "xfs_dir2_block.h"
-#include "xfs_dir2_trace.h"
+#include "xfs_trace.h"
 
 /*
  * Prototypes for internal functions.
@@ -169,7 +169,8 @@ xfs_dir2_block_to_sf(
 	xfs_dir2_sf_t		*sfp;		/* shortform structure */
 	xfs_ino_t               temp;
 
-	xfs_dir2_trace_args_sb("block_to_sf", args, size, bp);
+	trace_xfs_dir2_block_to_sf(args);
+
 	dp = args->dp;
 	mp = dp->i_mount;
 
@@ -281,7 +282,8 @@ xfs_dir2_sf_addname(
 	xfs_dir2_sf_t		*sfp;		/* shortform structure */
 	xfs_dir2_sf_entry_t	*sfep = NULL;	/* shortform entry */
 
-	xfs_dir2_trace_args("sf_addname", args);
+	trace_xfs_dir2_sf_addname(args);
+
 	ASSERT(xfs_dir2_sf_lookup(args) == ENOENT);
 	dp = args->dp;
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
@@ -654,7 +656,8 @@ xfs_dir2_sf_create(
 	xfs_dir2_sf_t	*sfp;		/* shortform structure */
 	int		size;		/* directory size */
 
-	xfs_dir2_trace_args_i("sf_create", args, pino);
+	trace_xfs_dir2_sf_create(args);
+
 	dp = args->dp;
 
 	ASSERT(dp != NULL);
@@ -808,7 +811,8 @@ xfs_dir2_sf_lookup(
 	enum xfs_dacmp		cmp;		/* comparison result */
 	xfs_dir2_sf_entry_t	*ci_sfep;	/* case-insens. entry */
 
-	xfs_dir2_trace_args("sf_lookup", args);
+	trace_xfs_dir2_sf_lookup(args);
+
 	xfs_dir2_sf_check(args);
 	dp = args->dp;
 
@@ -891,7 +895,8 @@ xfs_dir2_sf_removename(
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
 	xfs_dir2_sf_t		*sfp;		/* shortform structure */
 
-	xfs_dir2_trace_args("sf_removename", args);
+	trace_xfs_dir2_sf_removename(args);
+
 	dp = args->dp;
 
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
@@ -982,7 +987,8 @@ xfs_dir2_sf_replace(
 	xfs_dir2_sf_entry_t	*sfep;		/* shortform directory entry */
 	xfs_dir2_sf_t		*sfp;		/* shortform structure */
 
-	xfs_dir2_trace_args("sf_replace", args);
+	trace_xfs_dir2_sf_replace(args);
+
 	dp = args->dp;
 
 	ASSERT(dp->i_df.if_flags & XFS_IFINLINE);
@@ -1125,7 +1131,8 @@ xfs_dir2_sf_toino4(
 	xfs_dir2_sf_entry_t	*sfep;		/* new sf entry */
 	xfs_dir2_sf_t		*sfp;		/* new sf directory */
 
-	xfs_dir2_trace_args("sf_toino4", args);
+	trace_xfs_dir2_sf_toino4(args);
+
 	dp = args->dp;
 
 	/*
@@ -1202,7 +1209,8 @@ xfs_dir2_sf_toino8(
 	xfs_dir2_sf_entry_t	*sfep;		/* new sf entry */
 	xfs_dir2_sf_t		*sfp;		/* new sf directory */
 
-	xfs_dir2_trace_args("sf_toino8", args);
+	trace_xfs_dir2_sf_toino8(args);
+
 	dp = args->dp;
 
 	/*
diff --git a/fs/xfs/xfs_dir2_trace.c b/fs/xfs/xfs_dir2_trace.c
deleted file mode 100644
index 6cc7c0c681ac..000000000000
--- a/fs/xfs/xfs_dir2_trace.c
+++ /dev/null
@@ -1,216 +0,0 @@
-/*
- * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-#include "xfs.h"
-#include "xfs_fs.h"
-#include "xfs_types.h"
-#include "xfs_inum.h"
-#include "xfs_dir2.h"
-#include "xfs_da_btree.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
-#include "xfs_dir2_trace.h"
-
-#ifdef XFS_DIR2_TRACE
-ktrace_t	*xfs_dir2_trace_buf;
-
-/*
- * Enter something in the trace buffers.
- */
-static void
-xfs_dir2_trace_enter(
-	xfs_inode_t	*dp,
-	int		type,
-	char		*where,
-	char		*name,
-	int		namelen,
-	void		*a0,
-	void		*a1,
-	void		*a2,
-	void		*a3,
-	void		*a4,
-	void		*a5,
-	void		*a6,
-	void		*a7)
-{
-	void		*n[5];
-
-	ASSERT(xfs_dir2_trace_buf);
-	ASSERT(dp->i_dir_trace);
-	if (name)
-		memcpy(n, name, min((int)sizeof(n), namelen));
-	else
-		memset((char *)n, 0, sizeof(n));
-	ktrace_enter(xfs_dir2_trace_buf,
-		(void *)(long)type, (void *)where,
-		(void *)a0, (void *)a1, (void *)a2, (void *)a3,
-		(void *)a4, (void *)a5, (void *)a6, (void *)a7,
-		(void *)(long)namelen,
-		(void *)n[0], (void *)n[1], (void *)n[2],
-		(void *)n[3], (void *)n[4]);
-	ktrace_enter(dp->i_dir_trace,
-		(void *)(long)type, (void *)where,
-		(void *)a0, (void *)a1, (void *)a2, (void *)a3,
-		(void *)a4, (void *)a5, (void *)a6, (void *)a7,
-		(void *)(long)namelen,
-		(void *)n[0], (void *)n[1], (void *)n[2],
-		(void *)n[3], (void *)n[4]);
-}
-
-void
-xfs_dir2_trace_args(
-	char		*where,
-	xfs_da_args_t	*args)
-{
-	xfs_dir2_trace_enter(args->dp, XFS_DIR2_KTRACE_ARGS, where,
-		(char *)args->name, (int)args->namelen,
-		(void *)(unsigned long)args->hashval,
-		(void *)((unsigned long)(args->inumber >> 32)),
-		(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
-		(void *)args->dp, (void *)args->trans,
-		(void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
-		NULL, NULL);
-}
-
-void
-xfs_dir2_trace_args_b(
-	char		*where,
-	xfs_da_args_t	*args,
-	xfs_dabuf_t	*bp)
-{
-	xfs_dir2_trace_enter(args->dp, XFS_DIR2_KTRACE_ARGS_B, where,
-		(char *)args->name, (int)args->namelen,
-		(void *)(unsigned long)args->hashval,
-		(void *)((unsigned long)(args->inumber >> 32)),
-		(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
-		(void *)args->dp, (void *)args->trans,
-		(void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
-		(void *)(bp ? bp->bps[0] : NULL), NULL);
-}
-
-void
-xfs_dir2_trace_args_bb(
-	char		*where,
-	xfs_da_args_t	*args,
-	xfs_dabuf_t	*lbp,
-	xfs_dabuf_t	*dbp)
-{
-	xfs_dir2_trace_enter(args->dp, XFS_DIR2_KTRACE_ARGS_BB, where,
-		(char *)args->name, (int)args->namelen,
-		(void *)(unsigned long)args->hashval,
-		(void *)((unsigned long)(args->inumber >> 32)),
-		(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
-		(void *)args->dp, (void *)args->trans,
-		(void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
-		(void *)(lbp ? lbp->bps[0] : NULL),
-		(void *)(dbp ? dbp->bps[0] : NULL));
-}
-
-void
-xfs_dir2_trace_args_bibii(
-	char		*where,
-	xfs_da_args_t	*args,
-	xfs_dabuf_t	*bs,
-	int		ss,
-	xfs_dabuf_t	*bd,
-	int		sd,
-	int		c)
-{
-	xfs_buf_t	*bpbs = bs ? bs->bps[0] : NULL;
-	xfs_buf_t	*bpbd = bd ? bd->bps[0] : NULL;
-
-	xfs_dir2_trace_enter(args->dp, XFS_DIR2_KTRACE_ARGS_BIBII, where,
-		(char *)args->name, (int)args->namelen,
-		(void *)args->dp, (void *)args->trans,
-		(void *)bpbs, (void *)(long)ss, (void *)bpbd, (void *)(long)sd,
-		(void *)(long)c, NULL);
-}
-
-void
-xfs_dir2_trace_args_db(
-	char		*where,
-	xfs_da_args_t	*args,
-	xfs_dir2_db_t	db,
-	xfs_dabuf_t	*bp)
-{
-	xfs_buf_t	*dbp = bp ? bp->bps[0] : NULL;
-
-	xfs_dir2_trace_enter(args->dp, XFS_DIR2_KTRACE_ARGS_DB, where,
-		(char *)args->name, (int)args->namelen,
-		(void *)(unsigned long)args->hashval,
-		(void *)((unsigned long)(args->inumber >> 32)),
-		(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
-		(void *)args->dp, (void *)args->trans,
-		(void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
-		(void *)(long)db, (void *)dbp);
-}
-
-void
-xfs_dir2_trace_args_i(
-	char		*where,
-	xfs_da_args_t	*args,
-	xfs_ino_t	i)
-{
-	xfs_dir2_trace_enter(args->dp, XFS_DIR2_KTRACE_ARGS_I, where,
-		(char *)args->name, (int)args->namelen,
-		(void *)(unsigned long)args->hashval,
-		(void *)((unsigned long)(args->inumber >> 32)),
-		(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
-		(void *)args->dp, (void *)args->trans,
-		(void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
-		(void *)((unsigned long)(i >> 32)),
-		(void *)((unsigned long)(i & 0xFFFFFFFF)));
-}
-
-void
-xfs_dir2_trace_args_s(
-	char		*where,
-	xfs_da_args_t	*args,
-	int		s)
-{
-	xfs_dir2_trace_enter(args->dp, XFS_DIR2_KTRACE_ARGS_S, where,
-		(char *)args->name, (int)args->namelen,
-		(void *)(unsigned long)args->hashval,
-		(void *)((unsigned long)(args->inumber >> 32)),
-		(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
-		(void *)args->dp, (void *)args->trans,
-		(void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
-		(void *)(long)s, NULL);
-}
-
-void
-xfs_dir2_trace_args_sb(
-	char		*where,
-	xfs_da_args_t	*args,
-	int		s,
-	xfs_dabuf_t	*bp)
-{
-	xfs_buf_t	*dbp = bp ? bp->bps[0] : NULL;
-
-	xfs_dir2_trace_enter(args->dp, XFS_DIR2_KTRACE_ARGS_SB, where,
-		(char *)args->name, (int)args->namelen,
-		(void *)(unsigned long)args->hashval,
-		(void *)((unsigned long)(args->inumber >> 32)),
-		(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
-		(void *)args->dp, (void *)args->trans,
-		(void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
-		(void *)(long)s, (void *)dbp);
-}
-#endif	/* XFS_DIR2_TRACE */
diff --git a/fs/xfs/xfs_dir2_trace.h b/fs/xfs/xfs_dir2_trace.h
deleted file mode 100644
index ca3c754f4822..000000000000
--- a/fs/xfs/xfs_dir2_trace.h
+++ /dev/null
@@ -1,72 +0,0 @@
-/*
- * Copyright (c) 2000,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-#ifndef __XFS_DIR2_TRACE_H__
-#define __XFS_DIR2_TRACE_H__
-
-/*
- * Tracing for xfs v2 directories.
- */
-
-#if defined(XFS_DIR2_TRACE)
-
-struct ktrace;
-struct xfs_dabuf;
-struct xfs_da_args;
-
-#define	XFS_DIR2_GTRACE_SIZE		4096	/* global buffer */
-#define	XFS_DIR2_KTRACE_SIZE		32	/* per-inode buffer */
-extern struct ktrace *xfs_dir2_trace_buf;
-
-#define	XFS_DIR2_KTRACE_ARGS		1	/* args only */
-#define	XFS_DIR2_KTRACE_ARGS_B		2	/* args + buffer */
-#define	XFS_DIR2_KTRACE_ARGS_BB		3	/* args + 2 buffers */
-#define	XFS_DIR2_KTRACE_ARGS_DB		4	/* args, db, buffer */
-#define	XFS_DIR2_KTRACE_ARGS_I		5	/* args, inum */
-#define	XFS_DIR2_KTRACE_ARGS_S		6	/* args, int */
-#define	XFS_DIR2_KTRACE_ARGS_SB		7	/* args, int, buffer */
-#define	XFS_DIR2_KTRACE_ARGS_BIBII	8	/* args, buf/int/buf/int/int */
-
-void xfs_dir2_trace_args(char *where, struct xfs_da_args *args);
-void xfs_dir2_trace_args_b(char *where, struct xfs_da_args *args,
-			   struct xfs_dabuf *bp);
-void xfs_dir2_trace_args_bb(char *where, struct xfs_da_args *args,
-			    struct xfs_dabuf *lbp, struct xfs_dabuf *dbp);
-void xfs_dir2_trace_args_bibii(char *where, struct xfs_da_args *args,
-			       struct xfs_dabuf *bs, int ss,
-			       struct xfs_dabuf *bd, int sd, int c);
-void xfs_dir2_trace_args_db(char *where, struct xfs_da_args *args,
-			    xfs_dir2_db_t db, struct xfs_dabuf *bp);
-void xfs_dir2_trace_args_i(char *where, struct xfs_da_args *args, xfs_ino_t i);
-void xfs_dir2_trace_args_s(char *where, struct xfs_da_args *args, int s);
-void xfs_dir2_trace_args_sb(char *where, struct xfs_da_args *args, int s,
-			    struct xfs_dabuf *bp);
-
-#else	/* XFS_DIR2_TRACE */
-
-#define	xfs_dir2_trace_args(where, args)
-#define	xfs_dir2_trace_args_b(where, args, bp)
-#define	xfs_dir2_trace_args_bb(where, args, lbp, dbp)
-#define	xfs_dir2_trace_args_bibii(where, args, bs, ss, bd, sd, c)
-#define	xfs_dir2_trace_args_db(where, args, db, bp)
-#define	xfs_dir2_trace_args_i(where, args, i)
-#define	xfs_dir2_trace_args_s(where, args, s)
-#define	xfs_dir2_trace_args_sb(where, args, s, bp)
-
-#endif	/* XFS_DIR2_TRACE */
-
-#endif	/* __XFS_DIR2_TRACE_H__ */
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index edf8bdf4141f..a631e1451abb 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -34,6 +34,7 @@
 #include "xfs_utils.h"
 #include "xfs_mru_cache.h"
 #include "xfs_filestream.h"
+#include "xfs_trace.h"
 
 #ifdef XFS_FILESTREAMS_TRACE
 
@@ -394,9 +395,7 @@ xfs_filestream_init(void)
 	item_zone = kmem_zone_init(sizeof(fstrm_item_t), "fstrm_item");
 	if (!item_zone)
 		return -ENOMEM;
-#ifdef XFS_FILESTREAMS_TRACE
-	xfs_filestreams_trace_buf = ktrace_alloc(XFS_FSTRM_KTRACE_SIZE, KM_NOFS);
-#endif
+
 	return 0;
 }
 
@@ -407,9 +406,6 @@ xfs_filestream_init(void)
 void
 xfs_filestream_uninit(void)
 {
-#ifdef XFS_FILESTREAMS_TRACE
-	ktrace_free(xfs_filestreams_trace_buf);
-#endif
 	kmem_zone_destroy(item_zone);
 }
 
diff --git a/fs/xfs/xfs_filestream.h b/fs/xfs/xfs_filestream.h
index f655f7dc334c..4aba67c5f64f 100644
--- a/fs/xfs/xfs_filestream.h
+++ b/fs/xfs/xfs_filestream.h
@@ -79,7 +79,7 @@ extern ktrace_t *xfs_filestreams_trace_buf;
  * the cache that reference per-ag array elements that have since been
  * reallocated.
  */
-STATIC_INLINE int
+static inline int
 xfs_filestream_peek_ag(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno)
@@ -87,7 +87,7 @@ xfs_filestream_peek_ag(
 	return atomic_read(&mp->m_perag[agno].pagf_fstrms);
 }
 
-STATIC_INLINE int
+static inline int
 xfs_filestream_get_ag(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno)
@@ -95,7 +95,7 @@ xfs_filestream_get_ag(
 	return atomic_inc_return(&mp->m_perag[agno].pagf_fstrms);
 }
 
-STATIC_INLINE int
+static inline int
 xfs_filestream_put_ag(
 	xfs_mount_t	*mp,
 	xfs_agnumber_t	agno)
@@ -122,7 +122,7 @@ int xfs_filestream_new_ag(struct xfs_bmalloca *ap, xfs_agnumber_t *agp);
 
 
 /* filestreams for the inode? */
-STATIC_INLINE int
+static inline int
 xfs_inode_is_filestream(
 	struct xfs_inode	*ip)
 {
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 2d0b3e1da9e6..a13919a6a364 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -45,6 +45,7 @@
 #include "xfs_rtalloc.h"
 #include "xfs_rw.h"
 #include "xfs_filestream.h"
+#include "xfs_trace.h"
 
 /*
  * File system operations
@@ -201,8 +202,8 @@ xfs_growfs_data_private(
 		 * AG freelist header block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-				  XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
-				  XFS_FSS_TO_BB(mp, 1), 0);
+				 XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
+				 XFS_FSS_TO_BB(mp, 1), XBF_LOCK | XBF_MAPPED);
 		agf = XFS_BUF_TO_AGF(bp);
 		memset(agf, 0, mp->m_sb.sb_sectsize);
 		agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC);
@@ -233,8 +234,8 @@ xfs_growfs_data_private(
 		 * AG inode header block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-				  XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
-				  XFS_FSS_TO_BB(mp, 1), 0);
+				 XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)),
+				 XFS_FSS_TO_BB(mp, 1), XBF_LOCK | XBF_MAPPED);
 		agi = XFS_BUF_TO_AGI(bp);
 		memset(agi, 0, mp->m_sb.sb_sectsize);
 		agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC);
@@ -257,8 +258,9 @@ xfs_growfs_data_private(
 		 * BNO btree root block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-			XFS_AGB_TO_DADDR(mp, agno, XFS_BNO_BLOCK(mp)),
-			BTOBB(mp->m_sb.sb_blocksize), 0);
+				 XFS_AGB_TO_DADDR(mp, agno, XFS_BNO_BLOCK(mp)),
+				 BTOBB(mp->m_sb.sb_blocksize),
+				 XBF_LOCK | XBF_MAPPED);
 		block = XFS_BUF_TO_BLOCK(bp);
 		memset(block, 0, mp->m_sb.sb_blocksize);
 		block->bb_magic = cpu_to_be32(XFS_ABTB_MAGIC);
@@ -278,8 +280,9 @@ xfs_growfs_data_private(
 		 * CNT btree root block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-			XFS_AGB_TO_DADDR(mp, agno, XFS_CNT_BLOCK(mp)),
-			BTOBB(mp->m_sb.sb_blocksize), 0);
+				 XFS_AGB_TO_DADDR(mp, agno, XFS_CNT_BLOCK(mp)),
+				 BTOBB(mp->m_sb.sb_blocksize),
+				 XBF_LOCK | XBF_MAPPED);
 		block = XFS_BUF_TO_BLOCK(bp);
 		memset(block, 0, mp->m_sb.sb_blocksize);
 		block->bb_magic = cpu_to_be32(XFS_ABTC_MAGIC);
@@ -300,8 +303,9 @@ xfs_growfs_data_private(
 		 * INO btree root block
 		 */
 		bp = xfs_buf_get(mp->m_ddev_targp,
-			XFS_AGB_TO_DADDR(mp, agno, XFS_IBT_BLOCK(mp)),
-			BTOBB(mp->m_sb.sb_blocksize), 0);
+				 XFS_AGB_TO_DADDR(mp, agno, XFS_IBT_BLOCK(mp)),
+				 BTOBB(mp->m_sb.sb_blocksize),
+				 XBF_LOCK | XBF_MAPPED);
 		block = XFS_BUF_TO_BLOCK(bp);
 		memset(block, 0, mp->m_sb.sb_blocksize);
 		block->bb_magic = cpu_to_be32(XFS_IBT_MAGIC);
@@ -344,6 +348,7 @@ xfs_growfs_data_private(
 		be32_add_cpu(&agf->agf_length, new);
 		ASSERT(be32_to_cpu(agf->agf_length) ==
 		       be32_to_cpu(agi->agi_length));
+
 		xfs_alloc_log_agf(tp, bp, XFS_AGF_LENGTH);
 		/*
 		 * Free the new space.
@@ -611,7 +616,7 @@ xfs_fs_log_dummy(
 	xfs_inode_t	*ip;
 	int		error;
 
-	tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1);
+	tp = _xfs_trans_alloc(mp, XFS_TRANS_DUMMY1, KM_SLEEP);
 	error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
 	if (error) {
 		xfs_trans_cancel(tp, 0);
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
index 0785797db828..cb907ba69c4c 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/xfs_ialloc.c
@@ -425,7 +425,7 @@ xfs_ialloc_ag_alloc(
 	return 0;
 }
 
-STATIC_INLINE xfs_agnumber_t
+STATIC xfs_agnumber_t
 xfs_ialloc_next_ag(
 	xfs_mount_t	*mp)
 {
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index 80e526489be5..155e798f30a1 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -43,7 +43,7 @@
 #include "xfs_inode_item.h"
 #include "xfs_bmap.h"
 #include "xfs_btree_trace.h"
-#include "xfs_dir2_trace.h"
+#include "xfs_trace.h"
 
 
 /*
@@ -74,6 +74,8 @@ xfs_inode_alloc(
 	ASSERT(!spin_is_locked(&ip->i_flags_lock));
 	ASSERT(completion_done(&ip->i_flush));
 
+	mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
+
 	/* initialise the xfs inode */
 	ip->i_ino = ino;
 	ip->i_mount = mp;
@@ -87,30 +89,8 @@ xfs_inode_alloc(
 	ip->i_size = 0;
 	ip->i_new_size = 0;
 
-	/*
-	 * Initialize inode's trace buffers.
-	 */
-#ifdef	XFS_INODE_TRACE
-	ip->i_trace = ktrace_alloc(INODE_TRACE_SIZE, KM_NOFS);
-#endif
-#ifdef XFS_BMAP_TRACE
-	ip->i_xtrace = ktrace_alloc(XFS_BMAP_KTRACE_SIZE, KM_NOFS);
-#endif
-#ifdef XFS_BTREE_TRACE
-	ip->i_btrace = ktrace_alloc(XFS_BMBT_KTRACE_SIZE, KM_NOFS);
-#endif
-#ifdef XFS_RW_TRACE
-	ip->i_rwtrace = ktrace_alloc(XFS_RW_KTRACE_SIZE, KM_NOFS);
-#endif
-#ifdef XFS_ILOCK_TRACE
-	ip->i_lock_trace = ktrace_alloc(XFS_ILOCK_KTRACE_SIZE, KM_NOFS);
-#endif
-#ifdef XFS_DIR2_TRACE
-	ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_NOFS);
-#endif
-
 	/* prevent anyone from using this yet */
-	VFS_I(ip)->i_state = I_NEW|I_LOCK;
+	VFS_I(ip)->i_state = I_NEW;
 
 	return ip;
 }
@@ -130,25 +110,6 @@ xfs_inode_free(
 	if (ip->i_afp)
 		xfs_idestroy_fork(ip, XFS_ATTR_FORK);
 
-#ifdef XFS_INODE_TRACE
-	ktrace_free(ip->i_trace);
-#endif
-#ifdef XFS_BMAP_TRACE
-	ktrace_free(ip->i_xtrace);
-#endif
-#ifdef XFS_BTREE_TRACE
-	ktrace_free(ip->i_btrace);
-#endif
-#ifdef XFS_RW_TRACE
-	ktrace_free(ip->i_rwtrace);
-#endif
-#ifdef XFS_ILOCK_TRACE
-	ktrace_free(ip->i_lock_trace);
-#endif
-#ifdef XFS_DIR2_TRACE
-	ktrace_free(ip->i_dir_trace);
-#endif
-
 	if (ip->i_itemp) {
 		/*
 		 * Only if we are shutting down the fs will we see an
@@ -207,6 +168,7 @@ xfs_iget_cache_hit(
 	 *	     instead of polling for it.
 	 */
 	if (ip->i_flags & (XFS_INEW|XFS_IRECLAIM)) {
+		trace_xfs_iget_skip(ip);
 		XFS_STATS_INC(xs_ig_frecycle);
 		error = EAGAIN;
 		goto out_error;
@@ -225,7 +187,7 @@ xfs_iget_cache_hit(
 	 * Need to carefully get it back into useable state.
 	 */
 	if (ip->i_flags & XFS_IRECLAIMABLE) {
-		xfs_itrace_exit_tag(ip, "xfs_iget.alloc");
+		trace_xfs_iget_reclaim(ip);
 
 		/*
 		 * We need to set XFS_INEW atomically with clearing the
@@ -251,9 +213,10 @@ xfs_iget_cache_hit(
 			ip->i_flags &= ~XFS_INEW;
 			ip->i_flags |= XFS_IRECLAIMABLE;
 			__xfs_inode_set_reclaim_tag(pag, ip);
+			trace_xfs_iget_reclaim(ip);
 			goto out_error;
 		}
-		inode->i_state = I_LOCK|I_NEW;
+		inode->i_state = I_NEW;
 	} else {
 		/* If the VFS inode is being torn down, pause and try again. */
 		if (!igrab(inode)) {
@@ -270,8 +233,9 @@ xfs_iget_cache_hit(
 		xfs_ilock(ip, lock_flags);
 
 	xfs_iflags_clear(ip, XFS_ISTALE);
-	xfs_itrace_exit_tag(ip, "xfs_iget.found");
 	XFS_STATS_INC(xs_ig_found);
+
+	trace_xfs_iget_found(ip);
 	return 0;
 
 out_error:
@@ -290,7 +254,7 @@ xfs_iget_cache_miss(
 	struct xfs_inode	**ipp,
 	xfs_daddr_t		bno,
 	int			flags,
-	int			lock_flags) __releases(pag->pag_ici_lock)
+	int			lock_flags)
 {
 	struct xfs_inode	*ip;
 	int			error;
@@ -305,7 +269,7 @@ xfs_iget_cache_miss(
 	if (error)
 		goto out_destroy;
 
-	xfs_itrace_exit_tag(ip, "xfs_iget.alloc");
+	xfs_itrace_entry(ip);
 
 	if ((ip->i_d.di_mode == 0) && !(flags & XFS_IGET_CREATE)) {
 		error = ENOENT;
@@ -350,6 +314,8 @@ xfs_iget_cache_miss(
 
 	write_unlock(&pag->pag_ici_lock);
 	radix_tree_preload_end();
+
+	trace_xfs_iget_alloc(ip);
 	*ipp = ip;
 	return 0;
 
@@ -511,17 +477,21 @@ xfs_ireclaim(
 {
 	struct xfs_mount	*mp = ip->i_mount;
 	struct xfs_perag	*pag;
+	xfs_agino_t		agino = XFS_INO_TO_AGINO(mp, ip->i_ino);
 
 	XFS_STATS_INC(xs_ig_reclaims);
 
 	/*
-	 * Remove the inode from the per-AG radix tree.  It doesn't matter
-	 * if it was never added to it because radix_tree_delete can deal
-	 * with that case just fine.
+	 * Remove the inode from the per-AG radix tree.
+	 *
+	 * Because radix_tree_delete won't complain even if the item was never
+	 * added to the tree assert that it's been there before to catch
+	 * problems with the inode life time early on.
 	 */
 	pag = xfs_get_perag(mp, ip->i_ino);
 	write_lock(&pag->pag_ici_lock);
-	radix_tree_delete(&pag->pag_ici_root, XFS_INO_TO_AGINO(mp, ip->i_ino));
+	if (!radix_tree_delete(&pag->pag_ici_root, agino))
+		ASSERT(0);
 	write_unlock(&pag->pag_ici_lock);
 	xfs_put_perag(mp, pag);
 
@@ -636,7 +606,7 @@ xfs_ilock(
 	else if (lock_flags & XFS_ILOCK_SHARED)
 		mraccess_nested(&ip->i_lock, XFS_ILOCK_DEP(lock_flags));
 
-	xfs_ilock_trace(ip, 1, lock_flags, (inst_t *)__return_address);
+	trace_xfs_ilock(ip, lock_flags, _RET_IP_);
 }
 
 /*
@@ -681,7 +651,7 @@ xfs_ilock_nowait(
 		if (!mrtryaccess(&ip->i_lock))
 			goto out_undo_iolock;
 	}
-	xfs_ilock_trace(ip, 2, lock_flags, (inst_t *)__return_address);
+	trace_xfs_ilock_nowait(ip, lock_flags, _RET_IP_);
 	return 1;
 
  out_undo_iolock:
@@ -743,7 +713,7 @@ xfs_iunlock(
 		xfs_trans_unlocked_item(ip->i_itemp->ili_item.li_ailp,
 					(xfs_log_item_t*)(ip->i_itemp));
 	}
-	xfs_ilock_trace(ip, 3, lock_flags, (inst_t *)__return_address);
+	trace_xfs_iunlock(ip, lock_flags, _RET_IP_);
 }
 
 /*
@@ -762,6 +732,8 @@ xfs_ilock_demote(
 		mrdemote(&ip->i_lock);
 	if (lock_flags & XFS_IOLOCK_EXCL)
 		mrdemote(&ip->i_iolock);
+
+	trace_xfs_ilock_demote(ip, lock_flags, _RET_IP_);
 }
 
 #ifdef DEBUG
@@ -792,52 +764,3 @@ xfs_isilocked(
 	return 1;
 }
 #endif
-
-#ifdef	XFS_INODE_TRACE
-
-#define KTRACE_ENTER(ip, vk, s, line, ra)			\
-	ktrace_enter((ip)->i_trace,				\
-/*  0 */		(void *)(__psint_t)(vk),		\
-/*  1 */		(void *)(s),				\
-/*  2 */		(void *)(__psint_t) line,		\
-/*  3 */		(void *)(__psint_t)atomic_read(&VFS_I(ip)->i_count), \
-/*  4 */		(void *)(ra),				\
-/*  5 */		NULL,					\
-/*  6 */		(void *)(__psint_t)current_cpu(),	\
-/*  7 */		(void *)(__psint_t)current_pid(),	\
-/*  8 */		(void *)__return_address,		\
-/*  9 */		NULL, NULL, NULL, NULL, NULL, NULL, NULL)
-
-/*
- * Vnode tracing code.
- */
-void
-_xfs_itrace_entry(xfs_inode_t *ip, const char *func, inst_t *ra)
-{
-	KTRACE_ENTER(ip, INODE_KTRACE_ENTRY, func, 0, ra);
-}
-
-void
-_xfs_itrace_exit(xfs_inode_t *ip, const char *func, inst_t *ra)
-{
-	KTRACE_ENTER(ip, INODE_KTRACE_EXIT, func, 0, ra);
-}
-
-void
-xfs_itrace_hold(xfs_inode_t *ip, char *file, int line, inst_t *ra)
-{
-	KTRACE_ENTER(ip, INODE_KTRACE_HOLD, file, line, ra);
-}
-
-void
-_xfs_itrace_ref(xfs_inode_t *ip, char *file, int line, inst_t *ra)
-{
-	KTRACE_ENTER(ip, INODE_KTRACE_REF, file, line, ra);
-}
-
-void
-xfs_itrace_rele(xfs_inode_t *ip, char *file, int line, inst_t *ra)
-{
-	KTRACE_ENTER(ip, INODE_KTRACE_RELE, file, line, ra);
-}
-#endif	/* XFS_INODE_TRACE */
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index b92a4fa2a0a1..ef77fd88c8e3 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -47,10 +47,10 @@
 #include "xfs_rw.h"
 #include "xfs_error.h"
 #include "xfs_utils.h"
-#include "xfs_dir2_trace.h"
 #include "xfs_quota.h"
 #include "xfs_filestream.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 kmem_zone_t *xfs_ifork_zone;
 kmem_zone_t *xfs_inode_zone;
@@ -1291,42 +1291,6 @@ xfs_file_last_byte(
 	return last_byte;
 }
 
-#if defined(XFS_RW_TRACE)
-STATIC void
-xfs_itrunc_trace(
-	int		tag,
-	xfs_inode_t	*ip,
-	int		flag,
-	xfs_fsize_t	new_size,
-	xfs_off_t	toss_start,
-	xfs_off_t	toss_finish)
-{
-	if (ip->i_rwtrace == NULL) {
-		return;
-	}
-
-	ktrace_enter(ip->i_rwtrace,
-		     (void*)((long)tag),
-		     (void*)ip,
-		     (void*)(unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff),
-		     (void*)(unsigned long)(ip->i_d.di_size & 0xffffffff),
-		     (void*)((long)flag),
-		     (void*)(unsigned long)((new_size >> 32) & 0xffffffff),
-		     (void*)(unsigned long)(new_size & 0xffffffff),
-		     (void*)(unsigned long)((toss_start >> 32) & 0xffffffff),
-		     (void*)(unsigned long)(toss_start & 0xffffffff),
-		     (void*)(unsigned long)((toss_finish >> 32) & 0xffffffff),
-		     (void*)(unsigned long)(toss_finish & 0xffffffff),
-		     (void*)(unsigned long)current_cpu(),
-		     (void*)(unsigned long)current_pid(),
-		     (void*)NULL,
-		     (void*)NULL,
-		     (void*)NULL);
-}
-#else
-#define	xfs_itrunc_trace(tag, ip, flag, new_size, toss_start, toss_finish)
-#endif
-
 /*
  * Start the truncation of the file to new_size.  The new size
  * must be smaller than the current size.  This routine will
@@ -1409,8 +1373,7 @@ xfs_itruncate_start(
 		return 0;
 	}
 	last_byte = xfs_file_last_byte(ip);
-	xfs_itrunc_trace(XFS_ITRUNC_START, ip, flags, new_size, toss_start,
-			 last_byte);
+	trace_xfs_itruncate_start(ip, flags, new_size, toss_start, last_byte);
 	if (last_byte > toss_start) {
 		if (flags & XFS_ITRUNC_DEFINITE) {
 			xfs_tosspages(ip, toss_start,
@@ -1514,7 +1477,8 @@ xfs_itruncate_finish(
 		new_size = 0LL;
 	}
 	first_unmap_block = XFS_B_TO_FSB(mp, (xfs_ufsize_t)new_size);
-	xfs_itrunc_trace(XFS_ITRUNC_FINISH1, ip, 0, new_size, 0, 0);
+	trace_xfs_itruncate_finish_start(ip, new_size);
+
 	/*
 	 * The first thing we do is set the size to new_size permanently
 	 * on disk.  This way we don't have to worry about anyone ever
@@ -1731,7 +1695,7 @@ xfs_itruncate_finish(
 	ASSERT((new_size != 0) ||
 	       (fork == XFS_ATTR_FORK) ||
 	       (ip->i_d.di_nextents == 0));
-	xfs_itrunc_trace(XFS_ITRUNC_FINISH2, ip, 0, new_size, 0, 0);
+	trace_xfs_itruncate_finish_end(ip, new_size);
 	return 0;
 }
 
@@ -2877,8 +2841,8 @@ xfs_iflush(
 	mp = ip->i_mount;
 
 	/*
-	 * If the inode isn't dirty, then just release the inode
-	 * flush lock and do nothing.
+	 * If the inode isn't dirty, then just release the inode flush lock and
+	 * do nothing.
 	 */
 	if (xfs_inode_clean(ip)) {
 		xfs_ifunlock(ip);
@@ -2904,6 +2868,19 @@ xfs_iflush(
 	xfs_iunpin_wait(ip);
 
 	/*
+	 * For stale inodes we cannot rely on the backing buffer remaining
+	 * stale in cache for the remaining life of the stale inode and so
+	 * xfs_itobp() below may give us a buffer that no longer contains
+	 * inodes below. We have to check this after ensuring the inode is
+	 * unpinned so that it is safe to reclaim the stale inode after the
+	 * flush call.
+	 */
+	if (xfs_iflags_test(ip, XFS_ISTALE)) {
+		xfs_ifunlock(ip);
+		return 0;
+	}
+
+	/*
 	 * This may have been unpinned because the filesystem is shutting
 	 * down forcibly. If that's the case we must not write this inode
 	 * to disk, because the log record didn't make it to disk!
@@ -3252,23 +3229,6 @@ corrupt_out:
 	return XFS_ERROR(EFSCORRUPTED);
 }
 
-
-
-#ifdef XFS_ILOCK_TRACE
-void
-xfs_ilock_trace(xfs_inode_t *ip, int lock, unsigned int lockflags, inst_t *ra)
-{
-	ktrace_enter(ip->i_lock_trace,
-		     (void *)ip,
-		     (void *)(unsigned long)lock, /* 1 = LOCK, 3=UNLOCK, etc */
-		     (void *)(unsigned long)lockflags, /* XFS_ILOCK_EXCL etc */
-		     (void *)ra,		/* caller of ilock */
-		     (void *)(unsigned long)current_cpu(),
-		     (void *)(unsigned long)current_pid(),
-		     NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL);
-}
-#endif
-
 /*
  * Return a pointer to the extent record at file index idx.
  */
@@ -3300,13 +3260,17 @@ xfs_iext_get_ext(
  */
 void
 xfs_iext_insert(
-	xfs_ifork_t	*ifp,		/* inode fork pointer */
+	xfs_inode_t	*ip,		/* incore inode pointer */
 	xfs_extnum_t	idx,		/* starting index of new items */
 	xfs_extnum_t	count,		/* number of inserted items */
-	xfs_bmbt_irec_t	*new)		/* items to insert */
+	xfs_bmbt_irec_t	*new,		/* items to insert */
+	int		state)		/* type of extent conversion */
 {
+	xfs_ifork_t	*ifp = (state & BMAP_ATTRFORK) ? ip->i_afp : &ip->i_df;
 	xfs_extnum_t	i;		/* extent record index */
 
+	trace_xfs_iext_insert(ip, idx, new, state, _RET_IP_);
+
 	ASSERT(ifp->if_flags & XFS_IFEXTENTS);
 	xfs_iext_add(ifp, idx, count);
 	for (i = idx; i < idx + count; i++, new++)
@@ -3549,13 +3513,17 @@ xfs_iext_add_indirect_multi(
  */
 void
 xfs_iext_remove(
-	xfs_ifork_t	*ifp,		/* inode fork pointer */
+	xfs_inode_t	*ip,		/* incore inode pointer */
 	xfs_extnum_t	idx,		/* index to begin removing exts */
-	int		ext_diff)	/* number of extents to remove */
+	int		ext_diff,	/* number of extents to remove */
+	int		state)		/* type of extent conversion */
 {
+	xfs_ifork_t	*ifp = (state & BMAP_ATTRFORK) ? ip->i_afp : &ip->i_df;
 	xfs_extnum_t	nextents;	/* number of extents in file */
 	int		new_size;	/* size of extents after removal */
 
+	trace_xfs_iext_remove(ip, idx, state, _RET_IP_);
+
 	ASSERT(ext_diff > 0);
 	nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
 	new_size = (nextents - ext_diff) * sizeof(xfs_bmbt_rec_t);
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 41555de1d1db..ec1f28c4fc4f 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -213,7 +213,6 @@ typedef struct xfs_icdinode {
 
 struct bhv_desc;
 struct cred;
-struct ktrace;
 struct xfs_buf;
 struct xfs_bmap_free;
 struct xfs_bmbt_irec;
@@ -222,13 +221,6 @@ struct xfs_mount;
 struct xfs_trans;
 struct xfs_dquot;
 
-#if defined(XFS_ILOCK_TRACE)
-#define XFS_ILOCK_KTRACE_SIZE	32
-extern void xfs_ilock_trace(struct xfs_inode *, int, unsigned int, inst_t *);
-#else
-#define	xfs_ilock_trace(i,n,f,ra)
-#endif
-
 typedef struct dm_attrs_s {
 	__uint32_t	da_dmevmask;	/* DMIG event mask */
 	__uint16_t	da_dmstate;	/* DMIG state info */
@@ -271,26 +263,6 @@ typedef struct xfs_inode {
 
 	/* VFS inode */
 	struct inode		i_vnode;	/* embedded VFS inode */
-
-	/* Trace buffers per inode. */
-#ifdef XFS_INODE_TRACE
-	struct ktrace		*i_trace;	/* general inode trace */
-#endif
-#ifdef XFS_BMAP_TRACE
-	struct ktrace		*i_xtrace;	/* inode extent list trace */
-#endif
-#ifdef XFS_BTREE_TRACE
-	struct ktrace		*i_btrace;	/* inode bmap btree trace */
-#endif
-#ifdef XFS_RW_TRACE
-	struct ktrace		*i_rwtrace;	/* inode read/write trace */
-#endif
-#ifdef XFS_ILOCK_TRACE
-	struct ktrace		*i_lock_trace;	/* inode lock/unlock trace */
-#endif
-#ifdef XFS_DIR2_TRACE
-	struct ktrace		*i_dir_trace;	/* inode directory trace */
-#endif
 } xfs_inode_t;
 
 #define XFS_ISIZE(ip)	(((ip)->i_d.di_mode & S_IFMT) == S_IFREG) ? \
@@ -406,6 +378,14 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
 #define XFS_LOCK_MASK		(XFS_IOLOCK_EXCL | XFS_IOLOCK_SHARED \
 				| XFS_ILOCK_EXCL | XFS_ILOCK_SHARED)
 
+#define XFS_LOCK_FLAGS \
+	{ XFS_IOLOCK_EXCL,	"IOLOCK_EXCL" }, \
+	{ XFS_IOLOCK_SHARED,	"IOLOCK_SHARED" }, \
+	{ XFS_ILOCK_EXCL,	"ILOCK_EXCL" }, \
+	{ XFS_ILOCK_SHARED,	"ILOCK_SHARED" }, \
+	{ XFS_IUNLOCK_NONOTIFY,	"IUNLOCK_NONOTIFY" }
+
+
 /*
  * Flags for lockdep annotations.
  *
@@ -455,6 +435,10 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
 #define	XFS_ITRUNC_DEFINITE	0x1
 #define	XFS_ITRUNC_MAYBE	0x2
 
+#define XFS_ITRUNC_FLAGS \
+	{ XFS_ITRUNC_DEFINITE,	"DEFINITE" }, \
+	{ XFS_ITRUNC_MAYBE,	"MAYBE" }
+
 /*
  * For multiple groups support: if S_ISGID bit is set in the parent
  * directory, group of new file is set to that of the parent, and
@@ -507,48 +491,16 @@ void		xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint);
 void		xfs_synchronize_times(xfs_inode_t *);
 void		xfs_mark_inode_dirty_sync(xfs_inode_t *);
 
-#if defined(XFS_INODE_TRACE)
-
-#define	INODE_TRACE_SIZE	16		/* number of trace entries */
-#define	INODE_KTRACE_ENTRY	1
-#define	INODE_KTRACE_EXIT	2
-#define	INODE_KTRACE_HOLD	3
-#define	INODE_KTRACE_REF	4
-#define	INODE_KTRACE_RELE	5
-
-extern void _xfs_itrace_entry(struct xfs_inode *, const char *, inst_t *);
-extern void _xfs_itrace_exit(struct xfs_inode *, const char *, inst_t *);
-extern void xfs_itrace_hold(struct xfs_inode *, char *, int, inst_t *);
-extern void _xfs_itrace_ref(struct xfs_inode *, char *, int, inst_t *);
-extern void xfs_itrace_rele(struct xfs_inode *, char *, int, inst_t *);
-#define xfs_itrace_entry(ip)	\
-	_xfs_itrace_entry(ip, __func__, (inst_t *)__return_address)
-#define xfs_itrace_exit(ip)	\
-	_xfs_itrace_exit(ip, __func__, (inst_t *)__return_address)
-#define xfs_itrace_exit_tag(ip, tag)	\
-	_xfs_itrace_exit(ip, tag, (inst_t *)__return_address)
-#define xfs_itrace_ref(ip)	\
-	_xfs_itrace_ref(ip, __FILE__, __LINE__, (inst_t *)__return_address)
-
-#else
-#define	xfs_itrace_entry(a)
-#define	xfs_itrace_exit(a)
-#define	xfs_itrace_exit_tag(a, b)
-#define	xfs_itrace_hold(a, b, c, d)
-#define	xfs_itrace_ref(a)
-#define	xfs_itrace_rele(a, b, c, d)
-#endif
-
 #define IHOLD(ip) \
 do { \
 	ASSERT(atomic_read(&VFS_I(ip)->i_count) > 0) ; \
 	atomic_inc(&(VFS_I(ip)->i_count)); \
-	xfs_itrace_hold((ip), __FILE__, __LINE__, (inst_t *)__return_address); \
+	trace_xfs_ihold(ip, _THIS_IP_); \
 } while (0)
 
 #define IRELE(ip) \
 do { \
-	xfs_itrace_rele((ip), __FILE__, __LINE__, (inst_t *)__return_address); \
+	trace_xfs_irele(ip, _THIS_IP_); \
 	iput(VFS_I(ip)); \
 } while (0)
 
@@ -577,11 +529,11 @@ int		xfs_iread_extents(struct xfs_trans *, struct xfs_inode *, int);
 int		xfs_iextents_copy(struct xfs_inode *, xfs_bmbt_rec_t *, int);
 
 xfs_bmbt_rec_host_t *xfs_iext_get_ext(xfs_ifork_t *, xfs_extnum_t);
-void		xfs_iext_insert(xfs_ifork_t *, xfs_extnum_t, xfs_extnum_t,
-				xfs_bmbt_irec_t *);
+void		xfs_iext_insert(xfs_inode_t *, xfs_extnum_t, xfs_extnum_t,
+				xfs_bmbt_irec_t *, int);
 void		xfs_iext_add(xfs_ifork_t *, xfs_extnum_t, int);
 void		xfs_iext_add_indirect_multi(xfs_ifork_t *, int, xfs_extnum_t, int);
-void		xfs_iext_remove(xfs_ifork_t *, xfs_extnum_t, int);
+void		xfs_iext_remove(xfs_inode_t *, xfs_extnum_t, int, int);
 void		xfs_iext_remove_inline(xfs_ifork_t *, xfs_extnum_t, int);
 void		xfs_iext_remove_direct(xfs_ifork_t *, xfs_extnum_t, int);
 void		xfs_iext_remove_indirect(xfs_ifork_t *, xfs_extnum_t, int);
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 9794b876d6ff..f38855d21ea5 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -41,6 +41,7 @@
 #include "xfs_ialloc.h"
 #include "xfs_rw.h"
 #include "xfs_error.h"
+#include "xfs_trace.h"
 
 
 kmem_zone_t	*xfs_ili_zone;		/* inode log item zone */
@@ -800,7 +801,9 @@ xfs_inode_item_pushbuf(
 				  !completion_done(&ip->i_flush));
 			iip->ili_pushbuf_flag = 0;
 			xfs_iunlock(ip, XFS_ILOCK_SHARED);
-			xfs_buftrace("INODE ITEM PUSH", bp);
+
+			trace_xfs_inode_item_push(bp, _RET_IP_);
+
 			if (XFS_BUF_ISPINNED(bp)) {
 				xfs_log_force(mp, (xfs_lsn_t)0,
 					      XFS_LOG_FORCE);
diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h
index 65bae4c9b8bf..cc8df1ac7783 100644
--- a/fs/xfs/xfs_inode_item.h
+++ b/fs/xfs/xfs_inode_item.h
@@ -127,7 +127,7 @@ static inline int xfs_ilog_fdata(int w)
 #ifdef __KERNEL__
 
 struct xfs_buf;
-struct xfs_bmbt_rec_64;
+struct xfs_bmbt_rec;
 struct xfs_inode;
 struct xfs_mount;
 
@@ -140,9 +140,9 @@ typedef struct xfs_inode_log_item {
 	unsigned short		ili_flags;	   /* misc flags */
 	unsigned short		ili_logged;	   /* flushed logged data */
 	unsigned int		ili_last_fields;   /* fields when flushed */
-	struct xfs_bmbt_rec_64	*ili_extents_buf;  /* array of logged
+	struct xfs_bmbt_rec	*ili_extents_buf;  /* array of logged
 						      data exts */
-	struct xfs_bmbt_rec_64	*ili_aextents_buf; /* array of logged
+	struct xfs_bmbt_rec	*ili_aextents_buf; /* array of logged
 						      attr exts */
 	unsigned int            ili_pushbuf_flag;  /* one bit used in push_ail */
 
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 67ae5555a30a..0b65039951a0 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -47,72 +47,8 @@
 #include "xfs_trans_space.h"
 #include "xfs_utils.h"
 #include "xfs_iomap.h"
+#include "xfs_trace.h"
 
-#if defined(XFS_RW_TRACE)
-void
-xfs_iomap_enter_trace(
-	int		tag,
-	xfs_inode_t	*ip,
-	xfs_off_t	offset,
-	ssize_t		count)
-{
-	if (!ip->i_rwtrace)
-		return;
-
-	ktrace_enter(ip->i_rwtrace,
-		(void *)((unsigned long)tag),
-		(void *)ip,
-		(void *)((unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(ip->i_d.di_size & 0xffffffff)),
-		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(offset & 0xffffffff)),
-		(void *)((unsigned long)count),
-		(void *)((unsigned long)((ip->i_new_size >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(ip->i_new_size & 0xffffffff)),
-		(void *)((unsigned long)current_pid()),
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL,
-		(void *)NULL);
-}
-
-void
-xfs_iomap_map_trace(
-	int		tag,
-	xfs_inode_t	*ip,
-	xfs_off_t	offset,
-	ssize_t		count,
-	xfs_iomap_t	*iomapp,
-	xfs_bmbt_irec_t	*imapp,
-	int		flags)
-{
-	if (!ip->i_rwtrace)
-		return;
-
-	ktrace_enter(ip->i_rwtrace,
-		(void *)((unsigned long)tag),
-		(void *)ip,
-		(void *)((unsigned long)((ip->i_d.di_size >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(ip->i_d.di_size & 0xffffffff)),
-		(void *)((unsigned long)((offset >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(offset & 0xffffffff)),
-		(void *)((unsigned long)count),
-		(void *)((unsigned long)flags),
-		(void *)((unsigned long)((iomapp->iomap_offset >> 32) & 0xffffffff)),
-		(void *)((unsigned long)(iomapp->iomap_offset & 0xffffffff)),
-		(void *)((unsigned long)(iomapp->iomap_delta)),
-		(void *)((unsigned long)(iomapp->iomap_bsize)),
-		(void *)((unsigned long)(iomapp->iomap_bn)),
-		(void *)(__psint_t)(imapp->br_startoff),
-		(void *)((unsigned long)(imapp->br_blockcount)),
-		(void *)(__psint_t)(imapp->br_startblock));
-}
-#else
-#define xfs_iomap_enter_trace(tag, io, offset, count)
-#define xfs_iomap_map_trace(tag, io, offset, count, iomapp, imapp, flags)
-#endif
 
 #define XFS_WRITEIO_ALIGN(mp,off)	(((off) >> mp->m_writeio_log) \
 						<< mp->m_writeio_log)
@@ -187,21 +123,20 @@ xfs_iomap(
 	if (XFS_FORCED_SHUTDOWN(mp))
 		return XFS_ERROR(EIO);
 
+	trace_xfs_iomap_enter(ip, offset, count, flags, NULL);
+
 	switch (flags & (BMAPI_READ | BMAPI_WRITE | BMAPI_ALLOCATE)) {
 	case BMAPI_READ:
-		xfs_iomap_enter_trace(XFS_IOMAP_READ_ENTER, ip, offset, count);
 		lockmode = xfs_ilock_map_shared(ip);
 		bmapi_flags = XFS_BMAPI_ENTIRE;
 		break;
 	case BMAPI_WRITE:
-		xfs_iomap_enter_trace(XFS_IOMAP_WRITE_ENTER, ip, offset, count);
 		lockmode = XFS_ILOCK_EXCL;
 		if (flags & BMAPI_IGNSTATE)
 			bmapi_flags |= XFS_BMAPI_IGSTATE|XFS_BMAPI_ENTIRE;
 		xfs_ilock(ip, lockmode);
 		break;
 	case BMAPI_ALLOCATE:
-		xfs_iomap_enter_trace(XFS_IOMAP_ALLOC_ENTER, ip, offset, count);
 		lockmode = XFS_ILOCK_SHARED;
 		bmapi_flags = XFS_BMAPI_ENTIRE;
 
@@ -237,8 +172,7 @@ xfs_iomap(
 		if (nimaps &&
 		    (imap.br_startblock != HOLESTARTBLOCK) &&
 		    (imap.br_startblock != DELAYSTARTBLOCK)) {
-			xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, ip,
-					offset, count, iomapp, &imap, flags);
+			trace_xfs_iomap_found(ip, offset, count, flags, &imap);
 			break;
 		}
 
@@ -250,8 +184,7 @@ xfs_iomap(
 						      &imap, &nimaps);
 		}
 		if (!error) {
-			xfs_iomap_map_trace(XFS_IOMAP_ALLOC_MAP, ip,
-					offset, count, iomapp, &imap, flags);
+			trace_xfs_iomap_alloc(ip, offset, count, flags, &imap);
 		}
 		iomap_flags = IOMAP_NEW;
 		break;
@@ -261,8 +194,7 @@ xfs_iomap(
 		lockmode = 0;
 
 		if (nimaps && !isnullstartblock(imap.br_startblock)) {
-			xfs_iomap_map_trace(XFS_IOMAP_WRITE_MAP, ip,
-					offset, count, iomapp, &imap, flags);
+			trace_xfs_iomap_found(ip, offset, count, flags, &imap);
 			break;
 		}
 
@@ -623,8 +555,7 @@ retry:
 	 * delalloc blocks and retry without EOF preallocation.
 	 */
 	if (nimaps == 0) {
-		xfs_iomap_enter_trace(XFS_IOMAP_WRITE_NOSPACE,
-					ip, offset, count);
+		trace_xfs_delalloc_enospc(ip, offset, count);
 		if (flushed)
 			return XFS_ERROR(ENOSPC);
 
@@ -837,7 +768,7 @@ xfs_iomap_write_unwritten(
 	int		committed;
 	int		error;
 
-	xfs_iomap_enter_trace(XFS_IOMAP_UNWRITTEN, ip, offset, count);
+	trace_xfs_unwritten_convert(ip, offset, count);
 
 	offset_fsb = XFS_B_TO_FSBT(mp, offset);
 	count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
@@ -860,8 +791,15 @@ xfs_iomap_write_unwritten(
 		 * set up a transaction to convert the range of extents
 		 * from unwritten to real. Do allocations in a loop until
 		 * we have covered the range passed in.
+		 *
+		 * Note that we open code the transaction allocation here
+		 * to pass KM_NOFS--we can't risk to recursing back into
+		 * the filesystem here as we might be asked to write out
+		 * the same inode that we complete here and might deadlock
+		 * on the iolock.
 		 */
-		tp = xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE);
+		xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
+		tp = _xfs_trans_alloc(mp, XFS_TRANS_STRAT_WRITE, KM_NOFS);
 		tp->t_flags |= XFS_TRANS_RESERVE;
 		error = xfs_trans_reserve(tp, resblks,
 				XFS_WRITE_LOG_RES(mp), 0,
diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h
index fdcf7b82747f..174f29990991 100644
--- a/fs/xfs/xfs_iomap.h
+++ b/fs/xfs/xfs_iomap.h
@@ -43,6 +43,14 @@ typedef enum {
 	BMAPI_TRYLOCK = (1 << 7),	/* non-blocking request */
 } bmapi_flags_t;
 
+#define BMAPI_FLAGS \
+	{ BMAPI_READ,		"READ" }, \
+	{ BMAPI_WRITE,		"WRITE" }, \
+	{ BMAPI_ALLOCATE,	"ALLOCATE" }, \
+	{ BMAPI_IGNSTATE,	"IGNSTATE" }, \
+	{ BMAPI_DIRECT,		"DIRECT" }, \
+	{ BMAPI_MMAP,		"MMAP" }, \
+	{ BMAPI_TRYLOCK,	"TRYLOCK" }
 
 /*
  * xfs_iomap_t:  File system I/O map
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 9dbdff3ea484..600b5b06aaeb 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -40,6 +40,7 @@
 #include "xfs_dinode.h"
 #include "xfs_inode.h"
 #include "xfs_rw.h"
+#include "xfs_trace.h"
 
 kmem_zone_t	*xfs_log_ticket_zone;
 
@@ -122,85 +123,6 @@ STATIC void	xlog_verify_tail_lsn(xlog_t *log, xlog_in_core_t *iclog,
 
 STATIC int	xlog_iclogs_empty(xlog_t *log);
 
-#if defined(XFS_LOG_TRACE)
-
-#define XLOG_TRACE_LOGGRANT_SIZE	2048
-#define XLOG_TRACE_ICLOG_SIZE		256
-
-void
-xlog_trace_loggrant_alloc(xlog_t *log)
-{
-	log->l_grant_trace = ktrace_alloc(XLOG_TRACE_LOGGRANT_SIZE, KM_NOFS);
-}
-
-void
-xlog_trace_loggrant_dealloc(xlog_t *log)
-{
-	ktrace_free(log->l_grant_trace);
-}
-
-void
-xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string)
-{
-	unsigned long cnts;
-
-	/* ticket counts are 1 byte each */
-	cnts = ((unsigned long)tic->t_ocnt) | ((unsigned long)tic->t_cnt) << 8;
-
-	ktrace_enter(log->l_grant_trace,
-		     (void *)tic,
-		     (void *)log->l_reserve_headq,
-		     (void *)log->l_write_headq,
-		     (void *)((unsigned long)log->l_grant_reserve_cycle),
-		     (void *)((unsigned long)log->l_grant_reserve_bytes),
-		     (void *)((unsigned long)log->l_grant_write_cycle),
-		     (void *)((unsigned long)log->l_grant_write_bytes),
-		     (void *)((unsigned long)log->l_curr_cycle),
-		     (void *)((unsigned long)log->l_curr_block),
-		     (void *)((unsigned long)CYCLE_LSN(log->l_tail_lsn)),
-		     (void *)((unsigned long)BLOCK_LSN(log->l_tail_lsn)),
-		     (void *)string,
-		     (void *)((unsigned long)tic->t_trans_type),
-		     (void *)cnts,
-		     (void *)((unsigned long)tic->t_curr_res),
-		     (void *)((unsigned long)tic->t_unit_res));
-}
-
-void
-xlog_trace_iclog_alloc(xlog_in_core_t *iclog)
-{
-	iclog->ic_trace = ktrace_alloc(XLOG_TRACE_ICLOG_SIZE, KM_NOFS);
-}
-
-void
-xlog_trace_iclog_dealloc(xlog_in_core_t *iclog)
-{
-	ktrace_free(iclog->ic_trace);
-}
-
-void
-xlog_trace_iclog(xlog_in_core_t *iclog, uint state)
-{
-	ktrace_enter(iclog->ic_trace,
-		     (void *)((unsigned long)state),
-		     (void *)((unsigned long)current_pid()),
-		     (void *)NULL, (void *)NULL, (void *)NULL, (void *)NULL,
-		     (void *)NULL, (void *)NULL, (void *)NULL, (void *)NULL,
-		     (void *)NULL, (void *)NULL, (void *)NULL, (void *)NULL,
-		     (void *)NULL, (void *)NULL);
-}
-#else
-
-#define	xlog_trace_loggrant_alloc(log)
-#define	xlog_trace_loggrant_dealloc(log)
-#define	xlog_trace_loggrant(log,tic,string)
-
-#define	xlog_trace_iclog_alloc(iclog)
-#define	xlog_trace_iclog_dealloc(iclog)
-#define	xlog_trace_iclog(iclog,state)
-
-#endif /* XFS_LOG_TRACE */
-
 
 static void
 xlog_ins_ticketq(struct xlog_ticket **qp, struct xlog_ticket *tic)
@@ -353,15 +275,17 @@ xfs_log_done(xfs_mount_t	*mp,
 
 	if ((ticket->t_flags & XLOG_TIC_PERM_RESERV) == 0 ||
 	    (flags & XFS_LOG_REL_PERM_RESERV)) {
+		trace_xfs_log_done_nonperm(log, ticket);
+
 		/*
 		 * Release ticket if not permanent reservation or a specific
 		 * request has been made to release a permanent reservation.
 		 */
-		xlog_trace_loggrant(log, ticket, "xfs_log_done: (non-permanent)");
 		xlog_ungrant_log_space(log, ticket);
 		xfs_log_ticket_put(ticket);
 	} else {
-		xlog_trace_loggrant(log, ticket, "xfs_log_done: (permanent)");
+		trace_xfs_log_done_perm(log, ticket);
+
 		xlog_regrant_reserve_log_space(log, ticket);
 		/* If this ticket was a permanent reservation and we aren't
 		 * trying to release it, reset the inited flags; so next time
@@ -505,10 +429,13 @@ xfs_log_reserve(xfs_mount_t	 *mp,
 
 	XFS_STATS_INC(xs_try_logspace);
 
+
 	if (*ticket != NULL) {
 		ASSERT(flags & XFS_LOG_PERM_RESERV);
 		internal_ticket = (xlog_ticket_t *)*ticket;
-		xlog_trace_loggrant(log, internal_ticket, "xfs_log_reserve: existing ticket (permanent trans)");
+
+		trace_xfs_log_reserve(log, internal_ticket);
+
 		xlog_grant_push_ail(mp, internal_ticket->t_unit_res);
 		retval = xlog_regrant_write_log_space(log, internal_ticket);
 	} else {
@@ -519,10 +446,9 @@ xfs_log_reserve(xfs_mount_t	 *mp,
 			return XFS_ERROR(ENOMEM);
 		internal_ticket->t_trans_type = t_type;
 		*ticket = internal_ticket;
-		xlog_trace_loggrant(log, internal_ticket, 
-			(internal_ticket->t_flags & XLOG_TIC_PERM_RESERV) ?
-			"xfs_log_reserve: create new ticket (permanent trans)" :
-			"xfs_log_reserve: create new ticket");
+
+		trace_xfs_log_reserve(log, internal_ticket);
+
 		xlog_grant_push_ail(mp,
 				    (internal_ticket->t_unit_res *
 				     internal_ticket->t_cnt));
@@ -734,7 +660,7 @@ xfs_log_unmount_write(xfs_mount_t *mp)
 			spin_unlock(&log->l_icloglock);
 		}
 		if (tic) {
-			xlog_trace_loggrant(log, tic, "unmount rec");
+			trace_xfs_log_umount_write(log, tic);
 			xlog_ungrant_log_space(log, tic);
 			xfs_log_ticket_put(tic);
 		}
@@ -1030,7 +956,6 @@ xlog_iodone(xfs_buf_t *bp)
 		xfs_fs_cmn_err(CE_WARN, l->l_mp,
 				"xlog_iodone: Barriers are no longer supported"
 				" by device. Disabling barriers\n");
-		xfs_buftrace("XLOG_IODONE BARRIERS OFF", bp);
 	}
 
 	/*
@@ -1085,13 +1010,10 @@ xlog_bdstrat_cb(struct xfs_buf *bp)
 		return 0;
 	}
 
-	xfs_buftrace("XLOG__BDSTRAT IOERROR", bp);
 	XFS_BUF_ERROR(bp, EIO);
 	XFS_BUF_STALE(bp);
 	xfs_biodone(bp);
 	return XFS_ERROR(EIO);
-
-
 }
 
 /*
@@ -1246,7 +1168,6 @@ xlog_alloc_log(xfs_mount_t	*mp,
 	spin_lock_init(&log->l_grant_lock);
 	sv_init(&log->l_flush_wait, 0, "flush_wait");
 
-	xlog_trace_loggrant_alloc(log);
 	/* log record size must be multiple of BBSIZE; see xlog_rec_header_t */
 	ASSERT((XFS_BUF_SIZE(bp) & BBMASK) == 0);
 
@@ -1305,8 +1226,6 @@ xlog_alloc_log(xfs_mount_t	*mp,
 		sv_init(&iclog->ic_force_wait, SV_DEFAULT, "iclog-force");
 		sv_init(&iclog->ic_write_wait, SV_DEFAULT, "iclog-write");
 
-		xlog_trace_iclog_alloc(iclog);
-
 		iclogp = &iclog->ic_next;
 	}
 	*iclogp = log->l_iclog;			/* complete ring */
@@ -1321,13 +1240,11 @@ out_free_iclog:
 			sv_destroy(&iclog->ic_force_wait);
 			sv_destroy(&iclog->ic_write_wait);
 			xfs_buf_free(iclog->ic_bp);
-			xlog_trace_iclog_dealloc(iclog);
 		}
 		kmem_free(iclog);
 	}
 	spinlock_destroy(&log->l_icloglock);
 	spinlock_destroy(&log->l_grant_lock);
-	xlog_trace_loggrant_dealloc(log);
 	xfs_buf_free(log->l_xbuf);
 out_free_log:
 	kmem_free(log);
@@ -1524,6 +1441,7 @@ xlog_sync(xlog_t		*log,
 	XFS_BUF_ZEROFLAGS(bp);
 	XFS_BUF_BUSY(bp);
 	XFS_BUF_ASYNC(bp);
+	bp->b_flags |= XBF_LOG_BUFFER;
 	/*
 	 * Do an ordered write for the log block.
 	 * Its unnecessary to flush the first split block in the log wrap case.
@@ -1561,6 +1479,7 @@ xlog_sync(xlog_t		*log,
 		XFS_BUF_ZEROFLAGS(bp);
 		XFS_BUF_BUSY(bp);
 		XFS_BUF_ASYNC(bp);
+		bp->b_flags |= XBF_LOG_BUFFER;
 		if (log->l_mp->m_flags & XFS_MOUNT_BARRIER)
 			XFS_BUF_ORDERED(bp);
 		dptr = XFS_BUF_PTR(bp);
@@ -1607,7 +1526,6 @@ xlog_dealloc_log(xlog_t *log)
 		sv_destroy(&iclog->ic_force_wait);
 		sv_destroy(&iclog->ic_write_wait);
 		xfs_buf_free(iclog->ic_bp);
-		xlog_trace_iclog_dealloc(iclog);
 		next_iclog = iclog->ic_next;
 		kmem_free(iclog);
 		iclog = next_iclog;
@@ -1616,7 +1534,6 @@ xlog_dealloc_log(xlog_t *log)
 	spinlock_destroy(&log->l_grant_lock);
 
 	xfs_buf_free(log->l_xbuf);
-	xlog_trace_loggrant_dealloc(log);
 	log->l_mp->m_log = NULL;
 	kmem_free(log);
 }	/* xlog_dealloc_log */
@@ -2414,7 +2331,6 @@ restart:
 
 	iclog = log->l_iclog;
 	if (iclog->ic_state != XLOG_STATE_ACTIVE) {
-		xlog_trace_iclog(iclog, XLOG_TRACE_SLEEP_FLUSH);
 		XFS_STATS_INC(xs_log_noiclogs);
 
 		/* Wait for log writes to have flushed */
@@ -2520,13 +2436,15 @@ xlog_grant_log_space(xlog_t	   *log,
 
 	/* Is there space or do we need to sleep? */
 	spin_lock(&log->l_grant_lock);
-	xlog_trace_loggrant(log, tic, "xlog_grant_log_space: enter");
+
+	trace_xfs_log_grant_enter(log, tic);
 
 	/* something is already sleeping; insert new transaction at end */
 	if (log->l_reserve_headq) {
 		xlog_ins_ticketq(&log->l_reserve_headq, tic);
-		xlog_trace_loggrant(log, tic,
-				    "xlog_grant_log_space: sleep 1");
+
+		trace_xfs_log_grant_sleep1(log, tic);
+
 		/*
 		 * Gotta check this before going to sleep, while we're
 		 * holding the grant lock.
@@ -2540,8 +2458,7 @@ xlog_grant_log_space(xlog_t	   *log,
 		 * If we got an error, and the filesystem is shutting down,
 		 * we'll catch it down below. So just continue...
 		 */
-		xlog_trace_loggrant(log, tic,
-				    "xlog_grant_log_space: wake 1");
+		trace_xfs_log_grant_wake1(log, tic);
 		spin_lock(&log->l_grant_lock);
 	}
 	if (tic->t_flags & XFS_LOG_PERM_RESERV)
@@ -2558,8 +2475,9 @@ redo:
 	if (free_bytes < need_bytes) {
 		if ((tic->t_flags & XLOG_TIC_IN_Q) == 0)
 			xlog_ins_ticketq(&log->l_reserve_headq, tic);
-		xlog_trace_loggrant(log, tic,
-				    "xlog_grant_log_space: sleep 2");
+
+		trace_xfs_log_grant_sleep2(log, tic);
+
 		spin_unlock(&log->l_grant_lock);
 		xlog_grant_push_ail(log->l_mp, need_bytes);
 		spin_lock(&log->l_grant_lock);
@@ -2571,8 +2489,8 @@ redo:
 		if (XLOG_FORCED_SHUTDOWN(log))
 			goto error_return;
 
-		xlog_trace_loggrant(log, tic,
-				    "xlog_grant_log_space: wake 2");
+		trace_xfs_log_grant_wake2(log, tic);
+
 		goto redo;
 	} else if (tic->t_flags & XLOG_TIC_IN_Q)
 		xlog_del_ticketq(&log->l_reserve_headq, tic);
@@ -2592,7 +2510,7 @@ redo:
 		ASSERT(log->l_grant_write_bytes <= BBTOB(BLOCK_LSN(tail_lsn)));
 	}
 #endif
-	xlog_trace_loggrant(log, tic, "xlog_grant_log_space: exit");
+	trace_xfs_log_grant_exit(log, tic);
 	xlog_verify_grant_head(log, 1);
 	spin_unlock(&log->l_grant_lock);
 	return 0;
@@ -2600,7 +2518,9 @@ redo:
  error_return:
 	if (tic->t_flags & XLOG_TIC_IN_Q)
 		xlog_del_ticketq(&log->l_reserve_headq, tic);
-	xlog_trace_loggrant(log, tic, "xlog_grant_log_space: err_ret");
+
+	trace_xfs_log_grant_error(log, tic);
+
 	/*
 	 * If we are failing, make sure the ticket doesn't have any
 	 * current reservations. We don't want to add this back when
@@ -2640,7 +2560,8 @@ xlog_regrant_write_log_space(xlog_t	   *log,
 #endif
 
 	spin_lock(&log->l_grant_lock);
-	xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: enter");
+
+	trace_xfs_log_regrant_write_enter(log, tic);
 
 	if (XLOG_FORCED_SHUTDOWN(log))
 		goto error_return;
@@ -2669,8 +2590,8 @@ xlog_regrant_write_log_space(xlog_t	   *log,
 			if ((tic->t_flags & XLOG_TIC_IN_Q) == 0)
 				xlog_ins_ticketq(&log->l_write_headq, tic);
 
-			xlog_trace_loggrant(log, tic,
-				    "xlog_regrant_write_log_space: sleep 1");
+			trace_xfs_log_regrant_write_sleep1(log, tic);
+
 			spin_unlock(&log->l_grant_lock);
 			xlog_grant_push_ail(log->l_mp, need_bytes);
 			spin_lock(&log->l_grant_lock);
@@ -2685,8 +2606,7 @@ xlog_regrant_write_log_space(xlog_t	   *log,
 			if (XLOG_FORCED_SHUTDOWN(log))
 				goto error_return;
 
-			xlog_trace_loggrant(log, tic,
-				    "xlog_regrant_write_log_space: wake 1");
+			trace_xfs_log_regrant_write_wake1(log, tic);
 		}
 	}
 
@@ -2704,6 +2624,8 @@ redo:
 		spin_lock(&log->l_grant_lock);
 
 		XFS_STATS_INC(xs_sleep_logspace);
+		trace_xfs_log_regrant_write_sleep2(log, tic);
+
 		sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s);
 
 		/* If we're shutting down, this tic is already off the queue */
@@ -2711,8 +2633,7 @@ redo:
 		if (XLOG_FORCED_SHUTDOWN(log))
 			goto error_return;
 
-		xlog_trace_loggrant(log, tic,
-				    "xlog_regrant_write_log_space: wake 2");
+		trace_xfs_log_regrant_write_wake2(log, tic);
 		goto redo;
 	} else if (tic->t_flags & XLOG_TIC_IN_Q)
 		xlog_del_ticketq(&log->l_write_headq, tic);
@@ -2727,7 +2648,8 @@ redo:
 	}
 #endif
 
-	xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: exit");
+	trace_xfs_log_regrant_write_exit(log, tic);
+
 	xlog_verify_grant_head(log, 1);
 	spin_unlock(&log->l_grant_lock);
 	return 0;
@@ -2736,7 +2658,9 @@ redo:
  error_return:
 	if (tic->t_flags & XLOG_TIC_IN_Q)
 		xlog_del_ticketq(&log->l_reserve_headq, tic);
-	xlog_trace_loggrant(log, tic, "xlog_regrant_write_log_space: err_ret");
+
+	trace_xfs_log_regrant_write_error(log, tic);
+
 	/*
 	 * If we are failing, make sure the ticket doesn't have any
 	 * current reservations. We don't want to add this back when
@@ -2760,8 +2684,8 @@ STATIC void
 xlog_regrant_reserve_log_space(xlog_t	     *log,
 			       xlog_ticket_t *ticket)
 {
-	xlog_trace_loggrant(log, ticket,
-			    "xlog_regrant_reserve_log_space: enter");
+	trace_xfs_log_regrant_reserve_enter(log, ticket);
+
 	if (ticket->t_cnt > 0)
 		ticket->t_cnt--;
 
@@ -2769,8 +2693,9 @@ xlog_regrant_reserve_log_space(xlog_t	     *log,
 	xlog_grant_sub_space(log, ticket->t_curr_res);
 	ticket->t_curr_res = ticket->t_unit_res;
 	xlog_tic_reset_res(ticket);
-	xlog_trace_loggrant(log, ticket,
-			    "xlog_regrant_reserve_log_space: sub current res");
+
+	trace_xfs_log_regrant_reserve_sub(log, ticket);
+
 	xlog_verify_grant_head(log, 1);
 
 	/* just return if we still have some of the pre-reserved space */
@@ -2780,8 +2705,9 @@ xlog_regrant_reserve_log_space(xlog_t	     *log,
 	}
 
 	xlog_grant_add_space_reserve(log, ticket->t_unit_res);
-	xlog_trace_loggrant(log, ticket,
-			    "xlog_regrant_reserve_log_space: exit");
+
+	trace_xfs_log_regrant_reserve_exit(log, ticket);
+
 	xlog_verify_grant_head(log, 0);
 	spin_unlock(&log->l_grant_lock);
 	ticket->t_curr_res = ticket->t_unit_res;
@@ -2811,11 +2737,11 @@ xlog_ungrant_log_space(xlog_t	     *log,
 		ticket->t_cnt--;
 
 	spin_lock(&log->l_grant_lock);
-	xlog_trace_loggrant(log, ticket, "xlog_ungrant_log_space: enter");
+	trace_xfs_log_ungrant_enter(log, ticket);
 
 	xlog_grant_sub_space(log, ticket->t_curr_res);
 
-	xlog_trace_loggrant(log, ticket, "xlog_ungrant_log_space: sub current");
+	trace_xfs_log_ungrant_sub(log, ticket);
 
 	/* If this is a permanent reservation ticket, we may be able to free
 	 * up more space based on the remaining count.
@@ -2825,7 +2751,8 @@ xlog_ungrant_log_space(xlog_t	     *log,
 		xlog_grant_sub_space(log, ticket->t_unit_res*ticket->t_cnt);
 	}
 
-	xlog_trace_loggrant(log, ticket, "xlog_ungrant_log_space: exit");
+	trace_xfs_log_ungrant_exit(log, ticket);
+
 	xlog_verify_grant_head(log, 1);
 	spin_unlock(&log->l_grant_lock);
 	xfs_log_move_tail(log->l_mp, 1);
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index 679c7c4926a2..d55662db7077 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -19,7 +19,6 @@
 #define __XFS_LOG_PRIV_H__
 
 struct xfs_buf;
-struct ktrace;
 struct log;
 struct xlog_ticket;
 struct xfs_buf_cancel;
@@ -135,6 +134,12 @@ static inline uint xlog_get_client_id(__be32 i)
 #define XLOG_TIC_INITED		0x1	/* has been initialized */
 #define XLOG_TIC_PERM_RESERV	0x2	/* permanent reservation */
 #define XLOG_TIC_IN_Q		0x4
+
+#define XLOG_TIC_FLAGS \
+	{ XLOG_TIC_INITED,	"XLOG_TIC_INITED" }, \
+	{ XLOG_TIC_PERM_RESERV,	"XLOG_TIC_PERM_RESERV" }, \
+	{ XLOG_TIC_IN_Q,	"XLOG_TIC_IN_Q" }
+
 #endif	/* __KERNEL__ */
 
 #define XLOG_UNMOUNT_TYPE	0x556e	/* Un for Unmount */
@@ -361,9 +366,6 @@ typedef struct xlog_in_core {
 	int			ic_bwritecnt;
 	unsigned short		ic_state;
 	char			*ic_datap;	/* pointer to iclog data */
-#ifdef XFS_LOG_TRACE
-	struct ktrace		*ic_trace;
-#endif
 
 	/* Callback structures need their own cacheline */
 	spinlock_t		ic_callback_lock ____cacheline_aligned_in_smp;
@@ -429,10 +431,6 @@ typedef struct log {
 	int			l_grant_write_cycle;
 	int			l_grant_write_bytes;
 
-#ifdef XFS_LOG_TRACE
-	struct ktrace		*l_grant_trace;
-#endif
-
 	/* The following field are used for debugging; need to hold icloglock */
 #ifdef DEBUG
 	char			*l_iclog_bak[XLOG_MAX_ICLOGS];
@@ -456,12 +454,6 @@ extern void	 xlog_put_bp(struct xfs_buf *);
 
 extern kmem_zone_t	*xfs_log_ticket_zone;
 
-/* iclog tracing */
-#define XLOG_TRACE_GRAB_FLUSH  1
-#define XLOG_TRACE_REL_FLUSH   2
-#define XLOG_TRACE_SLEEP_FLUSH 3
-#define XLOG_TRACE_WAKE_FLUSH  4
-
 /*
  * Unmount record type is used as a pseudo transaction type for the ticket.
  * It's value must be outside the range of XFS_TRANS_* values.
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index fb17f8226b09..69ac2e5ef20c 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -46,6 +46,7 @@
 #include "xfs_quota.h"
 #include "xfs_rw.h"
 #include "xfs_utils.h"
+#include "xfs_trace.h"
 
 STATIC int	xlog_find_zeroed(xlog_t *, xfs_daddr_t *);
 STATIC int	xlog_clear_stale_blocks(xlog_t *, xfs_lsn_t);
@@ -225,16 +226,10 @@ xlog_header_check_dump(
 	xfs_mount_t		*mp,
 	xlog_rec_header_t	*head)
 {
-	int			b;
-
-	cmn_err(CE_DEBUG, "%s:  SB : uuid = ", __func__);
-	for (b = 0; b < 16; b++)
-		cmn_err(CE_DEBUG, "%02x", ((__uint8_t *)&mp->m_sb.sb_uuid)[b]);
-	cmn_err(CE_DEBUG, ", fmt = %d\n", XLOG_FMT);
-	cmn_err(CE_DEBUG, "    log : uuid = ");
-	for (b = 0; b < 16; b++)
-		cmn_err(CE_DEBUG, "%02x", ((__uint8_t *)&head->h_fs_uuid)[b]);
-	cmn_err(CE_DEBUG, ", fmt = %d\n", be32_to_cpu(head->h_fmt));
+	cmn_err(CE_DEBUG, "%s:  SB : uuid = %pU, fmt = %d\n",
+		__func__, &mp->m_sb.sb_uuid, XLOG_FMT);
+	cmn_err(CE_DEBUG, "    log : uuid = %pU, fmt = %d\n",
+		&head->h_fs_uuid, be32_to_cpu(head->h_fmt));
 }
 #else
 #define xlog_header_check_dump(mp, head)
@@ -2206,6 +2201,7 @@ xlog_recover_do_buffer_trans(
 	xfs_daddr_t		blkno;
 	int			len;
 	ushort			flags;
+	uint			buf_flags;
 
 	buf_f = (xfs_buf_log_format_t *)item->ri_buf[0].i_addr;
 
@@ -2246,12 +2242,11 @@ xlog_recover_do_buffer_trans(
 	}
 
 	mp = log->l_mp;
-	if (flags & XFS_BLI_INODE_BUF) {
-		bp = xfs_buf_read_flags(mp->m_ddev_targp, blkno, len,
-								XFS_BUF_LOCK);
-	} else {
-		bp = xfs_buf_read(mp->m_ddev_targp, blkno, len, 0);
-	}
+	buf_flags = XFS_BUF_LOCK;
+	if (!(flags & XFS_BLI_INODE_BUF))
+		buf_flags |= XFS_BUF_MAPPED;
+
+	bp = xfs_buf_read(mp->m_ddev_targp, blkno, len, buf_flags);
 	if (XFS_BUF_ISERROR(bp)) {
 		xfs_ioerror_alert("xlog_recover_do..(read#1)", log->l_mp,
 				  bp, blkno);
@@ -2350,8 +2345,8 @@ xlog_recover_do_inode_trans(
 		goto error;
 	}
 
-	bp = xfs_buf_read_flags(mp->m_ddev_targp, in_f->ilf_blkno,
-				in_f->ilf_len, XFS_BUF_LOCK);
+	bp = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len,
+			  XFS_BUF_LOCK);
 	if (XFS_BUF_ISERROR(bp)) {
 		xfs_ioerror_alert("xlog_recover_do..(read#2)", mp,
 				  bp, in_f->ilf_blkno);
@@ -3517,7 +3512,7 @@ xlog_do_recovery_pass(
 {
 	xlog_rec_header_t	*rhead;
 	xfs_daddr_t		blk_no;
-	xfs_caddr_t		bufaddr, offset;
+	xfs_caddr_t		offset;
 	xfs_buf_t		*hbp, *dbp;
 	int			error = 0, h_size;
 	int			bblks, split_bblks;
@@ -3610,7 +3605,7 @@ xlog_do_recovery_pass(
 			/*
 			 * Check for header wrapping around physical end-of-log
 			 */
-			offset = NULL;
+			offset = XFS_BUF_PTR(hbp);
 			split_hblks = 0;
 			wrapped_hblks = 0;
 			if (blk_no + hblks <= log->l_logBBsize) {
@@ -3646,9 +3641,8 @@ xlog_do_recovery_pass(
 				 *   - order is important.
 				 */
 				wrapped_hblks = hblks - split_hblks;
-				bufaddr = XFS_BUF_PTR(hbp);
 				error = XFS_BUF_SET_PTR(hbp,
-						bufaddr + BBTOB(split_hblks),
+						offset + BBTOB(split_hblks),
 						BBTOB(hblks - split_hblks));
 				if (error)
 					goto bread_err2;
@@ -3658,14 +3652,10 @@ xlog_do_recovery_pass(
 				if (error)
 					goto bread_err2;
 
-				error = XFS_BUF_SET_PTR(hbp, bufaddr,
+				error = XFS_BUF_SET_PTR(hbp, offset,
 							BBTOB(hblks));
 				if (error)
 					goto bread_err2;
-
-				if (!offset)
-					offset = xlog_align(log, 0,
-							wrapped_hblks, hbp);
 			}
 			rhead = (xlog_rec_header_t *)offset;
 			error = xlog_valid_rec_header(log, rhead,
@@ -3685,7 +3675,7 @@ xlog_do_recovery_pass(
 			} else {
 				/* This log record is split across the
 				 * physical end of log */
-				offset = NULL;
+				offset = XFS_BUF_PTR(dbp);
 				split_bblks = 0;
 				if (blk_no != log->l_logBBsize) {
 					/* some data is before the physical
@@ -3714,9 +3704,8 @@ xlog_do_recovery_pass(
 				 *   _first_, then the log start (LR header end)
 				 *   - order is important.
 				 */
-				bufaddr = XFS_BUF_PTR(dbp);
 				error = XFS_BUF_SET_PTR(dbp,
-						bufaddr + BBTOB(split_bblks),
+						offset + BBTOB(split_bblks),
 						BBTOB(bblks - split_bblks));
 				if (error)
 					goto bread_err2;
@@ -3727,13 +3716,9 @@ xlog_do_recovery_pass(
 				if (error)
 					goto bread_err2;
 
-				error = XFS_BUF_SET_PTR(dbp, bufaddr, h_size);
+				error = XFS_BUF_SET_PTR(dbp, offset, h_size);
 				if (error)
 					goto bread_err2;
-
-				if (!offset)
-					offset = xlog_align(log, wrapped_hblks,
-						bblks - split_bblks, dbp);
 			}
 			xlog_unpack_data(rhead, offset, log);
 			if ((error = xlog_recover_process_data(log, rhash,
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 8b6c9e807efb..eb403b40e120 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -44,6 +44,8 @@
 #include "xfs_quota.h"
 #include "xfs_fsops.h"
 #include "xfs_utils.h"
+#include "xfs_trace.h"
+
 
 STATIC void	xfs_unmountfs_wait(xfs_mount_t *);
 
@@ -583,8 +585,8 @@ xfs_readsb(xfs_mount_t *mp, int flags)
 	sector_size = xfs_getsize_buftarg(mp->m_ddev_targp);
 	extra_flags = XFS_BUF_LOCK | XFS_BUF_MANAGE | XFS_BUF_MAPPED;
 
-	bp = xfs_buf_read_flags(mp->m_ddev_targp, XFS_SB_DADDR,
-				BTOBB(sector_size), extra_flags);
+	bp = xfs_buf_read(mp->m_ddev_targp, XFS_SB_DADDR, BTOBB(sector_size),
+			  extra_flags);
 	if (!bp || XFS_BUF_ISERROR(bp)) {
 		xfs_fs_mount_cmn_err(flags, "SB read failed");
 		error = bp ? XFS_BUF_GETERROR(bp) : ENOMEM;
@@ -624,8 +626,8 @@ xfs_readsb(xfs_mount_t *mp, int flags)
 		XFS_BUF_UNMANAGE(bp);
 		xfs_buf_relse(bp);
 		sector_size = mp->m_sb.sb_sectsize;
-		bp = xfs_buf_read_flags(mp->m_ddev_targp, XFS_SB_DADDR,
-					BTOBB(sector_size), extra_flags);
+		bp = xfs_buf_read(mp->m_ddev_targp, XFS_SB_DADDR,
+				  BTOBB(sector_size), extra_flags);
 		if (!bp || XFS_BUF_ISERROR(bp)) {
 			xfs_fs_mount_cmn_err(flags, "SB re-read failed");
 			error = bp ? XFS_BUF_GETERROR(bp) : ENOMEM;
@@ -1471,7 +1473,7 @@ xfs_log_sbcount(
 	if (!xfs_sb_version_haslazysbcount(&mp->m_sb))
 		return 0;
 
-	tp = _xfs_trans_alloc(mp, XFS_TRANS_SB_COUNT);
+	tp = _xfs_trans_alloc(mp, XFS_TRANS_SB_COUNT, KM_SLEEP);
 	error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0,
 					XFS_DEFAULT_LOG_COUNT);
 	if (error) {
@@ -2123,7 +2125,7 @@ xfs_icsb_destroy_counters(
 	mutex_destroy(&mp->m_icsb_mutex);
 }
 
-STATIC_INLINE void
+STATIC void
 xfs_icsb_lock_cntr(
 	xfs_icsb_cnts_t	*icsbp)
 {
@@ -2132,7 +2134,7 @@ xfs_icsb_lock_cntr(
 	}
 }
 
-STATIC_INLINE void
+STATIC void
 xfs_icsb_unlock_cntr(
 	xfs_icsb_cnts_t	*icsbp)
 {
@@ -2140,7 +2142,7 @@ xfs_icsb_unlock_cntr(
 }
 
 
-STATIC_INLINE void
+STATIC void
 xfs_icsb_lock_all_counters(
 	xfs_mount_t	*mp)
 {
@@ -2153,7 +2155,7 @@ xfs_icsb_lock_all_counters(
 	}
 }
 
-STATIC_INLINE void
+STATIC void
 xfs_icsb_unlock_all_counters(
 	xfs_mount_t	*mp)
 {
@@ -2389,12 +2391,12 @@ xfs_icsb_modify_counters(
 {
 	xfs_icsb_cnts_t	*icsbp;
 	long long	lcounter;	/* long counter for 64 bit fields */
-	int		cpu, ret = 0;
+	int		ret = 0;
 
 	might_sleep();
 again:
-	cpu = get_cpu();
-	icsbp = (xfs_icsb_cnts_t *)per_cpu_ptr(mp->m_sb_cnts, cpu);
+	preempt_disable();
+	icsbp = this_cpu_ptr(mp->m_sb_cnts);
 
 	/*
 	 * if the counter is disabled, go to slow path
@@ -2438,11 +2440,11 @@ again:
 		break;
 	}
 	xfs_icsb_unlock_cntr(icsbp);
-	put_cpu();
+	preempt_enable();
 	return 0;
 
 slow_path:
-	put_cpu();
+	preempt_enable();
 
 	/*
 	 * serialise with a mutex so we don't burn lots of cpu on
@@ -2490,7 +2492,7 @@ slow_path:
 
 balance_counter:
 	xfs_icsb_unlock_cntr(icsbp);
-	put_cpu();
+	preempt_enable();
 
 	/*
 	 * We may have multiple threads here if multiple per-cpu
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index a6c023bc0fb2..1df7e4502967 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -93,6 +93,9 @@ typedef struct xfs_dmops {
 	xfs_send_unmount_t	xfs_send_unmount;
 } xfs_dmops_t;
 
+#define XFS_DMAPI_UNMOUNT_FLAGS(mp) \
+	(((mp)->m_dmevmask & (1 << DM_EVENT_UNMOUNT)) ? 0 : DM_FLAGS_UNWANTED)
+
 #define XFS_SEND_DATA(mp, ev,ip,off,len,fl,lock) \
 	(*(mp)->m_dm_ops->xfs_send_data)(ev,ip,off,len,fl,lock)
 #define XFS_SEND_MMAP(mp, vma,fl) \
@@ -101,12 +104,24 @@ typedef struct xfs_dmops {
 	(*(mp)->m_dm_ops->xfs_send_destroy)(ip,right)
 #define XFS_SEND_NAMESP(mp, ev,b1,r1,b2,r2,n1,n2,mode,rval,fl) \
 	(*(mp)->m_dm_ops->xfs_send_namesp)(ev,NULL,b1,r1,b2,r2,n1,n2,mode,rval,fl)
-#define XFS_SEND_PREUNMOUNT(mp,b1,r1,b2,r2,n1,n2,mode,rval,fl) \
-	(*(mp)->m_dm_ops->xfs_send_namesp)(DM_EVENT_PREUNMOUNT,mp,b1,r1,b2,r2,n1,n2,mode,rval,fl)
 #define XFS_SEND_MOUNT(mp,right,path,name) \
 	(*(mp)->m_dm_ops->xfs_send_mount)(mp,right,path,name)
-#define XFS_SEND_UNMOUNT(mp, ip,right,mode,rval,fl) \
-	(*(mp)->m_dm_ops->xfs_send_unmount)(mp,ip,right,mode,rval,fl)
+#define XFS_SEND_PREUNMOUNT(mp) \
+do { \
+	if (mp->m_flags & XFS_MOUNT_DMAPI) { \
+		(*(mp)->m_dm_ops->xfs_send_namesp)(DM_EVENT_PREUNMOUNT, mp, \
+			(mp)->m_rootip, DM_RIGHT_NULL, \
+			(mp)->m_rootip, DM_RIGHT_NULL, \
+			NULL, NULL, 0, 0, XFS_DMAPI_UNMOUNT_FLAGS(mp)); \
+	} \
+} while (0)
+#define XFS_SEND_UNMOUNT(mp) \
+do { \
+	if (mp->m_flags & XFS_MOUNT_DMAPI) { \
+		(*(mp)->m_dm_ops->xfs_send_unmount)(mp, (mp)->m_rootip, \
+			DM_RIGHT_NULL, 0, 0, XFS_DMAPI_UNMOUNT_FLAGS(mp)); \
+	} \
+} while (0)
 
 
 #ifdef HAVE_PERCPU_SB
@@ -387,13 +402,13 @@ xfs_put_perag(struct xfs_mount *mp, xfs_perag_t *pag)
  * Per-cpu superblock locking functions
  */
 #ifdef HAVE_PERCPU_SB
-STATIC_INLINE void
+static inline void
 xfs_icsb_lock(xfs_mount_t *mp)
 {
 	mutex_lock(&mp->m_icsb_mutex);
 }
 
-STATIC_INLINE void
+static inline void
 xfs_icsb_unlock(xfs_mount_t *mp)
 {
 	mutex_unlock(&mp->m_icsb_mutex);
diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h
index 3ec91ac74c2a..91bfd60f4c74 100644
--- a/fs/xfs/xfs_quota.h
+++ b/fs/xfs/xfs_quota.h
@@ -92,6 +92,14 @@ typedef struct xfs_dqblk {
 
 #define XFS_DQ_ALLTYPES		(XFS_DQ_USER|XFS_DQ_PROJ|XFS_DQ_GROUP)
 
+#define XFS_DQ_FLAGS \
+	{ XFS_DQ_USER,		"USER" }, \
+	{ XFS_DQ_PROJ,		"PROJ" }, \
+	{ XFS_DQ_GROUP,		"GROUP" }, \
+	{ XFS_DQ_DIRTY,		"DIRTY" }, \
+	{ XFS_DQ_WANT,		"WANT" }, \
+	{ XFS_DQ_INACTIVE,	"INACTIVE" }
+
 /*
  * In the worst case, when both user and group quotas are on,
  * we can have a max of three dquots changing in a single transaction.
diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c
index b81deea0ce19..fc1cda23b817 100644
--- a/fs/xfs/xfs_rename.c
+++ b/fs/xfs/xfs_rename.c
@@ -39,6 +39,7 @@
 #include "xfs_utils.h"
 #include "xfs_trans_space.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 
 /*
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 385f6dceba5d..6be05f756d59 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -45,6 +45,7 @@
 #include "xfs_inode_item.h"
 #include "xfs_trans_space.h"
 #include "xfs_utils.h"
+#include "xfs_trace.h"
 
 
 /*
@@ -1516,6 +1517,8 @@ xfs_rtfree_range(
 	 */
 	error = xfs_rtfind_forw(mp, tp, end, mp->m_sb.sb_rextents - 1,
 		&postblock);
+	if (error)
+		return error;
 	/*
 	 * If there are blocks not being freed at the front of the
 	 * old extent, add summary data for them to be allocated.
diff --git a/fs/xfs/xfs_rw.c b/fs/xfs/xfs_rw.c
index 3f816ad7ff19..5aa07caea5f1 100644
--- a/fs/xfs/xfs_rw.c
+++ b/fs/xfs/xfs_rw.c
@@ -44,6 +44,7 @@
 #include "xfs_error.h"
 #include "xfs_buf_item.h"
 #include "xfs_rw.h"
+#include "xfs_trace.h"
 
 /*
  * This is a subroutine for xfs_write() and other writers (xfs_ioctl)
@@ -171,7 +172,6 @@ xfs_bioerror(
 	 * No need to wait until the buffer is unpinned.
 	 * We aren't flushing it.
 	 */
-	xfs_buftrace("XFS IOERROR", bp);
 	XFS_BUF_ERROR(bp, EIO);
 	/*
 	 * We're calling biodone, so delete B_DONE flag. Either way
@@ -205,7 +205,6 @@ xfs_bioerror_relse(
 	ASSERT(XFS_BUF_IODONE_FUNC(bp) != xfs_buf_iodone_callbacks);
 	ASSERT(XFS_BUF_IODONE_FUNC(bp) != xlog_iodone);
 
-	xfs_buftrace("XFS IOERRELSE", bp);
 	fl = XFS_BUF_BFLAGS(bp);
 	/*
 	 * No need to wait until the buffer is unpinned.
@@ -277,10 +276,10 @@ xfs_read_buf(
 	xfs_buf_t	 *bp;
 	int		 error;
 
-	if (flags)
-		bp = xfs_buf_read_flags(target, blkno, len, flags);
-	else
-		bp = xfs_buf_read(target, blkno, len, flags);
+	if (!flags)
+		flags = XBF_LOCK | XBF_MAPPED;
+
+	bp = xfs_buf_read(target, blkno, len, flags);
 	if (!bp)
 		return XFS_ERROR(EIO);
 	error = XFS_BUF_GETERROR(bp);
@@ -336,3 +335,25 @@ xfs_bwrite(
 	}
 	return (error);
 }
+
+/*
+ * helper function to extract extent size hint from inode
+ */
+xfs_extlen_t
+xfs_get_extsz_hint(
+	struct xfs_inode	*ip)
+{
+	xfs_extlen_t		extsz;
+
+	if (unlikely(XFS_IS_REALTIME_INODE(ip))) {
+		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
+				? ip->i_d.di_extsize
+				: ip->i_mount->m_sb.sb_rextsize;
+		ASSERT(extsz);
+	} else {
+		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
+				? ip->i_d.di_extsize : 0;
+	}
+
+	return extsz;
+}
diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h
index f5e4874c37d8..571f2174435c 100644
--- a/fs/xfs/xfs_rw.h
+++ b/fs/xfs/xfs_rw.h
@@ -37,34 +37,6 @@ xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
 }
 
 /*
- * Flags for xfs_free_eofblocks
- */
-#define XFS_FREE_EOF_LOCK	(1<<0)
-#define XFS_FREE_EOF_NOLOCK	(1<<1)
-
-
-/*
- * helper function to extract extent size hint from inode
- */
-STATIC_INLINE xfs_extlen_t
-xfs_get_extsz_hint(
-	xfs_inode_t	*ip)
-{
-	xfs_extlen_t	extsz;
-
-	if (unlikely(XFS_IS_REALTIME_INODE(ip))) {
-		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
-				? ip->i_d.di_extsize
-				: ip->i_mount->m_sb.sb_rextsize;
-		ASSERT(extsz);
-	} else {
-		extsz = (ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)
-				? ip->i_d.di_extsize : 0;
-	}
-	return extsz;
-}
-
-/*
  * Prototypes for functions in xfs_rw.c.
  */
 extern int xfs_write_clear_setuid(struct xfs_inode *ip);
@@ -76,5 +48,6 @@ extern int xfs_read_buf(struct xfs_mount *mp, xfs_buftarg_t *btp,
 			struct xfs_buf **bpp);
 extern void xfs_ioerror_alert(char *func, struct xfs_mount *mp,
 				xfs_buf_t *bp, xfs_daddr_t blkno);
+extern xfs_extlen_t xfs_get_extsz_hint(struct xfs_inode *ip);
 
 #endif /* __XFS_RW_H__ */
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 66b849358e62..237badcbac3b 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -236,19 +236,20 @@ xfs_trans_alloc(
 	uint		type)
 {
 	xfs_wait_for_freeze(mp, SB_FREEZE_TRANS);
-	return _xfs_trans_alloc(mp, type);
+	return _xfs_trans_alloc(mp, type, KM_SLEEP);
 }
 
 xfs_trans_t *
 _xfs_trans_alloc(
 	xfs_mount_t	*mp,
-	uint		type)
+	uint		type,
+	uint		memflags)
 {
 	xfs_trans_t	*tp;
 
 	atomic_inc(&mp->m_active_trans);
 
-	tp = kmem_zone_zalloc(xfs_trans_zone, KM_SLEEP);
+	tp = kmem_zone_zalloc(xfs_trans_zone, memflags);
 	tp->t_magic = XFS_TRANS_MAGIC;
 	tp->t_type = type;
 	tp->t_mountp = mp;
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index ed47fc77759c..ca64f33c63a3 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -100,6 +100,49 @@ typedef struct xfs_trans_header {
 #define	XFS_TRANS_TYPE_MAX		41
 /* new transaction types need to be reflected in xfs_logprint(8) */
 
+#define XFS_TRANS_TYPES \
+	{ XFS_TRANS_SETATTR_NOT_SIZE,	"SETATTR_NOT_SIZE" }, \
+	{ XFS_TRANS_SETATTR_SIZE,	"SETATTR_SIZE" }, \
+	{ XFS_TRANS_INACTIVE,		"INACTIVE" }, \
+	{ XFS_TRANS_CREATE,		"CREATE" }, \
+	{ XFS_TRANS_CREATE_TRUNC,	"CREATE_TRUNC" }, \
+	{ XFS_TRANS_TRUNCATE_FILE,	"TRUNCATE_FILE" }, \
+	{ XFS_TRANS_REMOVE,		"REMOVE" }, \
+	{ XFS_TRANS_LINK,		"LINK" }, \
+	{ XFS_TRANS_RENAME,		"RENAME" }, \
+	{ XFS_TRANS_MKDIR,		"MKDIR" }, \
+	{ XFS_TRANS_RMDIR,		"RMDIR" }, \
+	{ XFS_TRANS_SYMLINK,		"SYMLINK" }, \
+	{ XFS_TRANS_SET_DMATTRS,	"SET_DMATTRS" }, \
+	{ XFS_TRANS_GROWFS,		"GROWFS" }, \
+	{ XFS_TRANS_STRAT_WRITE,	"STRAT_WRITE" }, \
+	{ XFS_TRANS_DIOSTRAT,		"DIOSTRAT" }, \
+	{ XFS_TRANS_WRITEID,		"WRITEID" }, \
+	{ XFS_TRANS_ADDAFORK,		"ADDAFORK" }, \
+	{ XFS_TRANS_ATTRINVAL,		"ATTRINVAL" }, \
+	{ XFS_TRANS_ATRUNCATE,		"ATRUNCATE" }, \
+	{ XFS_TRANS_ATTR_SET,		"ATTR_SET" }, \
+	{ XFS_TRANS_ATTR_RM,		"ATTR_RM" }, \
+	{ XFS_TRANS_ATTR_FLAG,		"ATTR_FLAG" }, \
+	{ XFS_TRANS_CLEAR_AGI_BUCKET,	"CLEAR_AGI_BUCKET" }, \
+	{ XFS_TRANS_QM_SBCHANGE,	"QM_SBCHANGE" }, \
+	{ XFS_TRANS_QM_QUOTAOFF,	"QM_QUOTAOFF" }, \
+	{ XFS_TRANS_QM_DQALLOC,		"QM_DQALLOC" }, \
+	{ XFS_TRANS_QM_SETQLIM,		"QM_SETQLIM" }, \
+	{ XFS_TRANS_QM_DQCLUSTER,	"QM_DQCLUSTER" }, \
+	{ XFS_TRANS_QM_QINOCREATE,	"QM_QINOCREATE" }, \
+	{ XFS_TRANS_QM_QUOTAOFF_END,	"QM_QOFF_END" }, \
+	{ XFS_TRANS_SB_UNIT,		"SB_UNIT" }, \
+	{ XFS_TRANS_FSYNC_TS,		"FSYNC_TS" }, \
+	{ XFS_TRANS_GROWFSRT_ALLOC,	"GROWFSRT_ALLOC" }, \
+	{ XFS_TRANS_GROWFSRT_ZERO,	"GROWFSRT_ZERO" }, \
+	{ XFS_TRANS_GROWFSRT_FREE,	"GROWFSRT_FREE" }, \
+	{ XFS_TRANS_SWAPEXT,		"SWAPEXT" }, \
+	{ XFS_TRANS_SB_COUNT,		"SB_COUNT" }, \
+	{ XFS_TRANS_DUMMY1,		"DUMMY1" }, \
+	{ XFS_TRANS_DUMMY2,		"DUMMY2" }, \
+	{ XLOG_UNMOUNT_REC_TYPE,	"UNMOUNT" }
+
 /*
  * This structure is used to track log items associated with
  * a transaction.  It points to the log item and keeps some
@@ -782,6 +825,10 @@ typedef struct xfs_log_item {
 #define	XFS_LI_IN_AIL	0x1
 #define XFS_LI_ABORTED	0x2
 
+#define XFS_LI_FLAGS \
+	{ XFS_LI_IN_AIL,	"IN_AIL" }, \
+	{ XFS_LI_ABORTED,	"ABORTED" }
+
 typedef struct xfs_item_ops {
 	uint (*iop_size)(xfs_log_item_t *);
 	void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *);
@@ -924,7 +971,7 @@ typedef struct xfs_trans {
  * XFS transaction mechanism exported interfaces.
  */
 xfs_trans_t	*xfs_trans_alloc(struct xfs_mount *, uint);
-xfs_trans_t	*_xfs_trans_alloc(struct xfs_mount *, uint);
+xfs_trans_t	*_xfs_trans_alloc(struct xfs_mount *, uint, uint);
 xfs_trans_t	*xfs_trans_dup(xfs_trans_t *);
 int		xfs_trans_reserve(xfs_trans_t *, uint, uint, uint,
 				  uint, uint);
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index 218829e6a152..49130628d5ef 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -38,6 +38,7 @@
 #include "xfs_trans_priv.h"
 #include "xfs_error.h"
 #include "xfs_rw.h"
+#include "xfs_trace.h"
 
 
 STATIC xfs_buf_t *xfs_trans_buf_item_match(xfs_trans_t *, xfs_buftarg_t *,
@@ -79,11 +80,8 @@ xfs_trans_get_buf(xfs_trans_t	*tp,
 	/*
 	 * Default to a normal get_buf() call if the tp is NULL.
 	 */
-	if (tp == NULL) {
-		bp = xfs_buf_get_flags(target_dev, blkno, len,
-							flags | BUF_BUSY);
-		return(bp);
-	}
+	if (tp == NULL)
+		return xfs_buf_get(target_dev, blkno, len, flags | BUF_BUSY);
 
 	/*
 	 * If we find the buffer in the cache with this transaction
@@ -98,26 +96,23 @@ xfs_trans_get_buf(xfs_trans_t	*tp,
 	}
 	if (bp != NULL) {
 		ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
-		if (XFS_FORCED_SHUTDOWN(tp->t_mountp)) {
-			xfs_buftrace("TRANS GET RECUR SHUT", bp);
+		if (XFS_FORCED_SHUTDOWN(tp->t_mountp))
 			XFS_BUF_SUPER_STALE(bp);
-		}
+
 		/*
 		 * If the buffer is stale then it was binval'ed
 		 * since last read.  This doesn't matter since the
 		 * caller isn't allowed to use the data anyway.
 		 */
-		else if (XFS_BUF_ISSTALE(bp)) {
-			xfs_buftrace("TRANS GET RECUR STALE", bp);
+		else if (XFS_BUF_ISSTALE(bp))
 			ASSERT(!XFS_BUF_ISDELAYWRITE(bp));
-		}
+
 		ASSERT(XFS_BUF_FSPRIVATE2(bp, xfs_trans_t *) == tp);
 		bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
 		ASSERT(bip != NULL);
 		ASSERT(atomic_read(&bip->bli_refcount) > 0);
 		bip->bli_recur++;
-		xfs_buftrace("TRANS GET RECUR", bp);
-		xfs_buf_item_trace("GET RECUR", bip);
+		trace_xfs_trans_get_buf_recur(bip);
 		return (bp);
 	}
 
@@ -129,7 +124,7 @@ xfs_trans_get_buf(xfs_trans_t	*tp,
 	 * easily deadlock with our current transaction as well as cause
 	 * us to run out of stack space.
 	 */
-	bp = xfs_buf_get_flags(target_dev, blkno, len, flags | BUF_BUSY);
+	bp = xfs_buf_get(target_dev, blkno, len, flags | BUF_BUSY);
 	if (bp == NULL) {
 		return NULL;
 	}
@@ -169,8 +164,7 @@ xfs_trans_get_buf(xfs_trans_t	*tp,
 	 */
 	XFS_BUF_SET_FSPRIVATE2(bp, tp);
 
-	xfs_buftrace("TRANS GET", bp);
-	xfs_buf_item_trace("GET", bip);
+	trace_xfs_trans_get_buf(bip);
 	return (bp);
 }
 
@@ -210,7 +204,7 @@ xfs_trans_getsb(xfs_trans_t	*tp,
 		ASSERT(bip != NULL);
 		ASSERT(atomic_read(&bip->bli_refcount) > 0);
 		bip->bli_recur++;
-		xfs_buf_item_trace("GETSB RECUR", bip);
+		trace_xfs_trans_getsb_recur(bip);
 		return (bp);
 	}
 
@@ -252,7 +246,7 @@ xfs_trans_getsb(xfs_trans_t	*tp,
 	 */
 	XFS_BUF_SET_FSPRIVATE2(bp, tp);
 
-	xfs_buf_item_trace("GETSB", bip);
+	trace_xfs_trans_getsb(bip);
 	return (bp);
 }
 
@@ -302,7 +296,7 @@ xfs_trans_read_buf(
 	 * Default to a normal get_buf() call if the tp is NULL.
 	 */
 	if (tp == NULL) {
-		bp = xfs_buf_read_flags(target, blkno, len, flags | BUF_BUSY);
+		bp = xfs_buf_read(target, blkno, len, flags | BUF_BUSY);
 		if (!bp)
 			return (flags & XFS_BUF_TRYLOCK) ?
 					EAGAIN : XFS_ERROR(ENOMEM);
@@ -350,7 +344,7 @@ xfs_trans_read_buf(
 		ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
 		ASSERT((XFS_BUF_ISERROR(bp)) == 0);
 		if (!(XFS_BUF_ISDONE(bp))) {
-			xfs_buftrace("READ_BUF_INCORE !DONE", bp);
+			trace_xfs_trans_read_buf_io(bp, _RET_IP_);
 			ASSERT(!XFS_BUF_ISASYNC(bp));
 			XFS_BUF_READ(bp);
 			xfsbdstrat(tp->t_mountp, bp);
@@ -375,7 +369,7 @@ xfs_trans_read_buf(
 		 * brelse it either. Just get out.
 		 */
 		if (XFS_FORCED_SHUTDOWN(mp)) {
-			xfs_buftrace("READ_BUF_INCORE XFSSHUTDN", bp);
+			trace_xfs_trans_read_buf_shut(bp, _RET_IP_);
 			*bpp = NULL;
 			return XFS_ERROR(EIO);
 		}
@@ -385,7 +379,7 @@ xfs_trans_read_buf(
 		bip->bli_recur++;
 
 		ASSERT(atomic_read(&bip->bli_refcount) > 0);
-		xfs_buf_item_trace("READ RECUR", bip);
+		trace_xfs_trans_read_buf_recur(bip);
 		*bpp = bp;
 		return 0;
 	}
@@ -398,14 +392,13 @@ xfs_trans_read_buf(
 	 * easily deadlock with our current transaction as well as cause
 	 * us to run out of stack space.
 	 */
-	bp = xfs_buf_read_flags(target, blkno, len, flags | BUF_BUSY);
+	bp = xfs_buf_read(target, blkno, len, flags | BUF_BUSY);
 	if (bp == NULL) {
 		*bpp = NULL;
 		return 0;
 	}
 	if (XFS_BUF_GETERROR(bp) != 0) {
 	    XFS_BUF_SUPER_STALE(bp);
-		xfs_buftrace("READ ERROR", bp);
 		error = XFS_BUF_GETERROR(bp);
 
 		xfs_ioerror_alert("xfs_trans_read_buf", mp,
@@ -464,8 +457,7 @@ xfs_trans_read_buf(
 	 */
 	XFS_BUF_SET_FSPRIVATE2(bp, tp);
 
-	xfs_buftrace("TRANS READ", bp);
-	xfs_buf_item_trace("READ", bip);
+	trace_xfs_trans_read_buf(bip);
 	*bpp = bp;
 	return 0;
 
@@ -483,7 +475,7 @@ shutdown_abort:
 	ASSERT((XFS_BUF_BFLAGS(bp) & (XFS_B_STALE|XFS_B_DELWRI)) !=
 						(XFS_B_STALE|XFS_B_DELWRI));
 
-	xfs_buftrace("READ_BUF XFSSHUTDN", bp);
+	trace_xfs_trans_read_buf_shut(bp, _RET_IP_);
 	xfs_buf_relse(bp);
 	*bpp = NULL;
 	return XFS_ERROR(EIO);
@@ -549,13 +541,14 @@ xfs_trans_brelse(xfs_trans_t	*tp,
 	lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)bip);
 	ASSERT(lidp != NULL);
 
+	trace_xfs_trans_brelse(bip);
+
 	/*
 	 * If the release is just for a recursive lock,
 	 * then decrement the count and return.
 	 */
 	if (bip->bli_recur > 0) {
 		bip->bli_recur--;
-		xfs_buf_item_trace("RELSE RECUR", bip);
 		return;
 	}
 
@@ -563,10 +556,8 @@ xfs_trans_brelse(xfs_trans_t	*tp,
 	 * If the buffer is dirty within this transaction, we can't
 	 * release it until we commit.
 	 */
-	if (lidp->lid_flags & XFS_LID_DIRTY) {
-		xfs_buf_item_trace("RELSE DIRTY", bip);
+	if (lidp->lid_flags & XFS_LID_DIRTY)
 		return;
-	}
 
 	/*
 	 * If the buffer has been invalidated, then we can't release
@@ -574,13 +565,10 @@ xfs_trans_brelse(xfs_trans_t	*tp,
 	 * as part of this transaction.  This prevents us from pulling
 	 * the item from the AIL before we should.
 	 */
-	if (bip->bli_flags & XFS_BLI_STALE) {
-		xfs_buf_item_trace("RELSE STALE", bip);
+	if (bip->bli_flags & XFS_BLI_STALE)
 		return;
-	}
 
 	ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
-	xfs_buf_item_trace("RELSE", bip);
 
 	/*
 	 * Free up the log item descriptor tracking the released item.
@@ -677,7 +665,7 @@ xfs_trans_bjoin(xfs_trans_t	*tp,
 	 */
 	XFS_BUF_SET_FSPRIVATE2(bp, tp);
 
-	xfs_buf_item_trace("BJOIN", bip);
+	trace_xfs_trans_bjoin(bip);
 }
 
 /*
@@ -701,7 +689,7 @@ xfs_trans_bhold(xfs_trans_t	*tp,
 	ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
 	ASSERT(atomic_read(&bip->bli_refcount) > 0);
 	bip->bli_flags |= XFS_BLI_HOLD;
-	xfs_buf_item_trace("BHOLD", bip);
+	trace_xfs_trans_bhold(bip);
 }
 
 /*
@@ -724,7 +712,8 @@ xfs_trans_bhold_release(xfs_trans_t	*tp,
 	ASSERT(atomic_read(&bip->bli_refcount) > 0);
 	ASSERT(bip->bli_flags & XFS_BLI_HOLD);
 	bip->bli_flags &= ~XFS_BLI_HOLD;
-	xfs_buf_item_trace("BHOLD RELEASE", bip);
+
+	trace_xfs_trans_bhold_release(bip);
 }
 
 /*
@@ -770,6 +759,8 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
 	XFS_BUF_SET_IODONE_FUNC(bp, xfs_buf_iodone_callbacks);
 	bip->bli_item.li_cb = (void(*)(xfs_buf_t*,xfs_log_item_t*))xfs_buf_iodone;
 
+	trace_xfs_trans_log_buf(bip);
+
 	/*
 	 * If we invalidated the buffer within this transaction, then
 	 * cancel the invalidation now that we're dirtying the buffer
@@ -777,7 +768,6 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
 	 * because we have a reference to the buffer this entire time.
 	 */
 	if (bip->bli_flags & XFS_BLI_STALE) {
-		xfs_buf_item_trace("BLOG UNSTALE", bip);
 		bip->bli_flags &= ~XFS_BLI_STALE;
 		ASSERT(XFS_BUF_ISSTALE(bp));
 		XFS_BUF_UNSTALE(bp);
@@ -792,7 +782,6 @@ xfs_trans_log_buf(xfs_trans_t	*tp,
 	lidp->lid_flags &= ~XFS_LID_BUF_STALE;
 	bip->bli_flags |= XFS_BLI_LOGGED;
 	xfs_buf_item_log(bip, first, last);
-	xfs_buf_item_trace("BLOG", bip);
 }
 
 
@@ -831,6 +820,8 @@ xfs_trans_binval(
 	ASSERT(lidp != NULL);
 	ASSERT(atomic_read(&bip->bli_refcount) > 0);
 
+	trace_xfs_trans_binval(bip);
+
 	if (bip->bli_flags & XFS_BLI_STALE) {
 		/*
 		 * If the buffer is already invalidated, then
@@ -843,8 +834,6 @@ xfs_trans_binval(
 		ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
 		ASSERT(lidp->lid_flags & XFS_LID_DIRTY);
 		ASSERT(tp->t_flags & XFS_TRANS_DIRTY);
-		xfs_buftrace("XFS_BINVAL RECUR", bp);
-		xfs_buf_item_trace("BINVAL RECUR", bip);
 		return;
 	}
 
@@ -878,8 +867,6 @@ xfs_trans_binval(
 	      (bip->bli_format.blf_map_size * sizeof(uint)));
 	lidp->lid_flags |= XFS_LID_DIRTY|XFS_LID_BUF_STALE;
 	tp->t_flags |= XFS_TRANS_DIRTY;
-	xfs_buftrace("XFS_BINVAL", bp);
-	xfs_buf_item_trace("BINVAL", bip);
 }
 
 /*
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index b572f7e840e0..6f268756bf36 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -53,6 +53,7 @@
 #include "xfs_log_priv.h"
 #include "xfs_filestream.h"
 #include "xfs_vnodeops.h"
+#include "xfs_trace.h"
 
 int
 xfs_setattr(
@@ -69,7 +70,6 @@ xfs_setattr(
 	uint			commit_flags=0;
 	uid_t			uid=0, iuid=0;
 	gid_t			gid=0, igid=0;
-	int			timeflags = 0;
 	struct xfs_dquot	*udqp, *gdqp, *olddquot1, *olddquot2;
 	int			need_iolock = 1;
 
@@ -134,16 +134,13 @@ xfs_setattr(
 	if (flags & XFS_ATTR_NOLOCK)
 		need_iolock = 0;
 	if (!(mask & ATTR_SIZE)) {
-		if ((mask != (ATTR_CTIME|ATTR_ATIME|ATTR_MTIME)) ||
-		    (mp->m_flags & XFS_MOUNT_WSYNC)) {
-			tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
-			commit_flags = 0;
-			if ((code = xfs_trans_reserve(tp, 0,
-						     XFS_ICHANGE_LOG_RES(mp), 0,
-						     0, 0))) {
-				lock_flags = 0;
-				goto error_return;
-			}
+		tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
+		commit_flags = 0;
+		code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp),
+					 0, 0, 0);
+		if (code) {
+			lock_flags = 0;
+			goto error_return;
 		}
 	} else {
 		if (DM_EVENT_ENABLED(ip, DM_EVENT_TRUNCATE) &&
@@ -294,15 +291,23 @@ xfs_setattr(
 		 * or we are explicitly asked to change it. This handles
 		 * the semantic difference between truncate() and ftruncate()
 		 * as implemented in the VFS.
+		 *
+		 * The regular truncate() case without ATTR_CTIME and ATTR_MTIME
+		 * is a special case where we need to update the times despite
+		 * not having these flags set.  For all other operations the
+		 * VFS set these flags explicitly if it wants a timestamp
+		 * update.
 		 */
-		if (iattr->ia_size != ip->i_size || (mask & ATTR_CTIME))
-			timeflags |= XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG;
+		if (iattr->ia_size != ip->i_size &&
+		    (!(mask & (ATTR_CTIME | ATTR_MTIME)))) {
+			iattr->ia_ctime = iattr->ia_mtime =
+				current_fs_time(inode->i_sb);
+			mask |= ATTR_CTIME | ATTR_MTIME;
+		}
 
 		if (iattr->ia_size > ip->i_size) {
 			ip->i_d.di_size = iattr->ia_size;
 			ip->i_size = iattr->ia_size;
-			if (!(flags & XFS_ATTR_DMI))
-				xfs_ichgtime(ip, XFS_ICHGTIME_CHG);
 			xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 		} else if (iattr->ia_size <= ip->i_size ||
 			   (iattr->ia_size == 0 && ip->i_d.di_nextents)) {
@@ -373,9 +378,6 @@ xfs_setattr(
 			ip->i_d.di_gid = gid;
 			inode->i_gid = gid;
 		}
-
-		xfs_trans_log_inode (tp, ip, XFS_ILOG_CORE);
-		timeflags |= XFS_ICHGTIME_CHG;
 	}
 
 	/*
@@ -392,51 +394,37 @@ xfs_setattr(
 
 		inode->i_mode &= S_IFMT;
 		inode->i_mode |= mode & ~S_IFMT;
-
-		xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-		timeflags |= XFS_ICHGTIME_CHG;
 	}
 
 	/*
 	 * Change file access or modified times.
 	 */
-	if (mask & (ATTR_ATIME|ATTR_MTIME)) {
-		if (mask & ATTR_ATIME) {
-			inode->i_atime = iattr->ia_atime;
-			ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec;
-			ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec;
-			ip->i_update_core = 1;
-		}
-		if (mask & ATTR_MTIME) {
-			inode->i_mtime = iattr->ia_mtime;
-			ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
-			ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
-			timeflags &= ~XFS_ICHGTIME_MOD;
-			timeflags |= XFS_ICHGTIME_CHG;
-		}
-		if (tp && (mask & (ATTR_MTIME_SET|ATTR_ATIME_SET)))
-			xfs_trans_log_inode (tp, ip, XFS_ILOG_CORE);
+	if (mask & ATTR_ATIME) {
+		inode->i_atime = iattr->ia_atime;
+		ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec;
+		ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec;
+		ip->i_update_core = 1;
 	}
-
-	/*
-	 * Change file inode change time only if ATTR_CTIME set
-	 * AND we have been called by a DMI function.
-	 */
-
-	if ((flags & XFS_ATTR_DMI) && (mask & ATTR_CTIME)) {
+	if (mask & ATTR_CTIME) {
 		inode->i_ctime = iattr->ia_ctime;
 		ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
 		ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
 		ip->i_update_core = 1;
-		timeflags &= ~XFS_ICHGTIME_CHG;
+	}
+	if (mask & ATTR_MTIME) {
+		inode->i_mtime = iattr->ia_mtime;
+		ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
+		ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
+		ip->i_update_core = 1;
 	}
 
 	/*
-	 * Send out timestamp changes that need to be set to the
-	 * current time.  Not done when called by a DMI function.
+	 * And finally, log the inode core if any attribute in it
+	 * has been changed.
 	 */
-	if (timeflags && !(flags & XFS_ATTR_DMI))
-		xfs_ichgtime(ip, timeflags);
+	if (mask & (ATTR_UID|ATTR_GID|ATTR_MODE|
+		    ATTR_ATIME|ATTR_CTIME|ATTR_MTIME))
+		xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 
 	XFS_STATS_INC(xs_ig_attrchg);
 
@@ -451,12 +439,10 @@ xfs_setattr(
 	 * mix so this probably isn't worth the trouble to optimize.
 	 */
 	code = 0;
-	if (tp) {
-		if (mp->m_flags & XFS_MOUNT_WSYNC)
-			xfs_trans_set_sync(tp);
+	if (mp->m_flags & XFS_MOUNT_WSYNC)
+		xfs_trans_set_sync(tp);
 
-		code = xfs_trans_commit(tp, commit_flags);
-	}
+	code = xfs_trans_commit(tp, commit_flags);
 
 	xfs_iunlock(ip, lock_flags);
 
@@ -538,9 +524,8 @@ xfs_readlink_bmap(
 		d = XFS_FSB_TO_DADDR(mp, mval[n].br_startblock);
 		byte_cnt = XFS_FSB_TO_B(mp, mval[n].br_blockcount);
 
-		bp = xfs_buf_read_flags(mp->m_ddev_targp, d, BTOBB(byte_cnt),
-					XBF_LOCK | XBF_MAPPED |
-					XBF_DONT_BLOCK);
+		bp = xfs_buf_read(mp->m_ddev_targp, d, BTOBB(byte_cnt),
+				  XBF_LOCK | XBF_MAPPED | XBF_DONT_BLOCK);
 		error = XFS_BUF_GETERROR(bp);
 		if (error) {
 			xfs_ioerror_alert("xfs_readlink",
@@ -709,6 +694,11 @@ xfs_fsync(
 }
 
 /*
+ * Flags for xfs_free_eofblocks
+ */
+#define XFS_FREE_EOF_TRYLOCK	(1<<0)
+
+/*
  * This is called by xfs_inactive to free any blocks beyond eof
  * when the link count isn't zero and by xfs_dm_punch_hole() when
  * punching a hole to EOF.
@@ -726,7 +716,6 @@ xfs_free_eofblocks(
 	xfs_filblks_t	map_len;
 	int		nimaps;
 	xfs_bmbt_irec_t	imap;
-	int		use_iolock = (flags & XFS_FREE_EOF_LOCK);
 
 	/*
 	 * Figure out if there are any blocks beyond the end
@@ -768,14 +757,19 @@ xfs_free_eofblocks(
 		 * cache and we can't
 		 * do that within a transaction.
 		 */
-		if (use_iolock)
+		if (flags & XFS_FREE_EOF_TRYLOCK) {
+			if (!xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) {
+				xfs_trans_cancel(tp, 0);
+				return 0;
+			}
+		} else {
 			xfs_ilock(ip, XFS_IOLOCK_EXCL);
+		}
 		error = xfs_itruncate_start(ip, XFS_ITRUNC_DEFINITE,
 				    ip->i_size);
 		if (error) {
 			xfs_trans_cancel(tp, 0);
-			if (use_iolock)
-				xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+			xfs_iunlock(ip, XFS_IOLOCK_EXCL);
 			return error;
 		}
 
@@ -812,8 +806,7 @@ xfs_free_eofblocks(
 			error = xfs_trans_commit(tp,
 						XFS_TRANS_RELEASE_LOG_RES);
 		}
-		xfs_iunlock(ip, (use_iolock ? (XFS_IOLOCK_EXCL|XFS_ILOCK_EXCL)
-					    : XFS_ILOCK_EXCL));
+		xfs_iunlock(ip, XFS_IOLOCK_EXCL|XFS_ILOCK_EXCL);
 	}
 	return error;
 }
@@ -1113,7 +1106,17 @@ xfs_release(
 		     (ip->i_df.if_flags & XFS_IFEXTENTS))  &&
 		    (!(ip->i_d.di_flags &
 				(XFS_DIFLAG_PREALLOC | XFS_DIFLAG_APPEND)))) {
-			error = xfs_free_eofblocks(mp, ip, XFS_FREE_EOF_LOCK);
+
+			/*
+			 * If we can't get the iolock just skip truncating
+			 * the blocks past EOF because we could deadlock
+			 * with the mmap_sem otherwise.  We'll get another
+			 * chance to drop them once the last reference to
+			 * the inode is dropped, so we'll never leak blocks
+			 * permanently.
+			 */
+			error = xfs_free_eofblocks(mp, ip,
+						   XFS_FREE_EOF_TRYLOCK);
 			if (error)
 				return error;
 		}
@@ -1184,7 +1187,7 @@ xfs_inactive(
 		     (!(ip->i_d.di_flags &
 				(XFS_DIFLAG_PREALLOC | XFS_DIFLAG_APPEND)) ||
 		      (ip->i_delayed_blks != 0)))) {
-			error = xfs_free_eofblocks(mp, ip, XFS_FREE_EOF_LOCK);
+			error = xfs_free_eofblocks(mp, ip, 0);
 			if (error)
 				return VN_INACTIVE_CACHE;
 		}
@@ -1380,7 +1383,6 @@ xfs_lookup(
 	if (error)
 		goto out_free_name;
 
-	xfs_itrace_ref(*ipp);
 	return 0;
 
 out_free_name:
@@ -1526,7 +1528,6 @@ xfs_create(
 	 * At this point, we've gotten a newly allocated inode.
 	 * It is locked (and joined to the transaction).
 	 */
-	xfs_itrace_ref(ip);
 	ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
 
 	/*
@@ -1986,9 +1987,6 @@ xfs_remove(
 	if (!is_dir && link_zero && xfs_inode_is_filestream(ip))
 		xfs_filestream_deassociate(ip);
 
-	xfs_itrace_exit(ip);
-	xfs_itrace_exit(dp);
-
  std_return:
 	if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTREMOVE)) {
 		XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE, dp, DM_RIGHT_NULL,
@@ -2285,7 +2283,6 @@ xfs_symlink(
 			goto error_return;
 		goto error1;
 	}
-	xfs_itrace_ref(ip);
 
 	/*
 	 * An error after we've joined dp to the transaction will result in the
@@ -2456,46 +2453,6 @@ xfs_set_dmattrs(
 	return error;
 }
 
-int
-xfs_reclaim(
-	xfs_inode_t	*ip)
-{
-
-	xfs_itrace_entry(ip);
-
-	ASSERT(!VN_MAPPED(VFS_I(ip)));
-
-	/* bad inode, get out here ASAP */
-	if (is_bad_inode(VFS_I(ip))) {
-		xfs_ireclaim(ip);
-		return 0;
-	}
-
-	xfs_ioend_wait(ip);
-
-	ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0);
-
-	/*
-	 * If we have nothing to flush with this inode then complete the
-	 * teardown now, otherwise break the link between the xfs inode and the
-	 * linux inode and clean up the xfs inode later. This avoids flushing
-	 * the inode to disk during the delete operation itself.
-	 *
-	 * When breaking the link, we need to set the XFS_IRECLAIMABLE flag
-	 * first to ensure that xfs_iunpin() will never see an xfs inode
-	 * that has a linux inode being reclaimed. Synchronisation is provided
-	 * by the i_flags_lock.
-	 */
-	if (!ip->i_update_core && (ip->i_itemp == NULL)) {
-		xfs_ilock(ip, XFS_ILOCK_EXCL);
-		xfs_iflock(ip);
-		xfs_iflags_set(ip, XFS_IRECLAIMABLE);
-		return xfs_reclaim_inode(ip, 1, XFS_IFLUSH_DELWRI_ELSE_SYNC);
-	}
-	xfs_inode_set_reclaim_tag(ip);
-	return 0;
-}
-
 /*
  * xfs_alloc_file_space()
  *      This routine allocates disk space for the given file.
@@ -2868,7 +2825,6 @@ xfs_free_file_space(
 	ioffset = offset & ~(rounding - 1);
 
 	if (VN_CACHED(VFS_I(ip)) != 0) {
-		xfs_inval_cached_trace(ip, ioffset, -1, ioffset, -1);
 		error = xfs_flushinval_pages(ip, ioffset, -1, FI_REMAPF_LOCKED);
 		if (error)
 			goto out_unlock_iolock;
diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h
index a9e102de71a1..167a467403a5 100644
--- a/fs/xfs/xfs_vnodeops.h
+++ b/fs/xfs/xfs_vnodeops.h
@@ -38,7 +38,6 @@ int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name,
 		const char *target_path, mode_t mode, struct xfs_inode **ipp,
 		cred_t *credp);
 int xfs_set_dmattrs(struct xfs_inode *ip, u_int evmask, u_int16_t state);
-int xfs_reclaim(struct xfs_inode *ip);
 int xfs_change_file_space(struct xfs_inode *ip, int cmd,
 		xfs_flock64_t *bf, xfs_off_t offset, int attr_flags);
 int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name,