Merge commit 'linus/master' into x86/urgent

93 files changed, 2265 insertions, 1242 deletions

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index b9b3bb51b1e..d15ea1790bf 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -767,7 +767,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 	
 	current->mm->start_stack = bprm->p;
 
-	/* Now we do a little grungy work by mmaping the ELF image into
+	/* Now we do a little grungy work by mmapping the ELF image into
 	   the correct location in memory. */
 	for(i = 0, elf_ppnt = elf_phdata;
 	    i < loc->elf_ex.e_phnum; i++, elf_ppnt++) {
diff --git a/fs/bio.c b/fs/bio.c
index e23a63f4f7d..76e6713abf9 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -272,7 +272,7 @@ EXPORT_SYMBOL(bio_init);
  *   for a &struct bio to become free. If a %NULL @bs is passed in, we will
  *   fall back to just using @kmalloc to allocate the required memory.
  *
- *   Note that the caller must set ->bi_destructor on succesful return
+ *   Note that the caller must set ->bi_destructor on successful return
  *   of a bio, to do the appropriate freeing of the bio once the reference
  *   count drops to zero.
  **/
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index ccbdcb54ec5..46bea0f4dc7 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -256,7 +256,7 @@ out:
  * Insert @em into @tree or perform a simple forward/backward merge with
  * existing mappings.  The extent_map struct passed in will be inserted
  * into the tree directly, with an additional reference taken, or a
- * reference dropped if the merge attempt was sucessfull.
+ * reference dropped if the merge attempt was successfull.
  */
 int add_extent_mapping(struct extent_map_tree *tree,
 		       struct extent_map *em)
diff --git a/fs/cifs/README b/fs/cifs/README
index 79c1a93400b..a727b7cb075 100644
--- a/fs/cifs/README
+++ b/fs/cifs/README
@@ -423,7 +423,7 @@ A partial list of the supported mount options follows:
 		source name to use to represent the client netbios machine 
 		name when doing the RFC1001 netbios session initialize.
   direct        Do not do inode data caching on files opened on this mount.
-		This precludes mmaping files on this mount. In some cases
+		This precludes mmapping files on this mount. In some cases
 		with fast networks and little or no caching benefits on the
 		client (e.g. when the application is doing large sequential
 		reads bigger than page size without rereading the same data) 
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index 5d0fde18039..4b35f7ec058 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -39,7 +39,7 @@
 
 /*
  * MAX_REQ is the maximum number of requests that WE will send
- * on one socket concurently. It also matches the most common
+ * on one socket concurrently. It also matches the most common
  * value of max multiplex returned by servers.  We may
  * eventually want to use the negotiated value (in case
  * future servers can handle more) when we are more confident that
diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h
index 2d07f890a84..3877737f96a 100644
--- a/fs/cifs/cifspdu.h
+++ b/fs/cifs/cifspdu.h
@@ -1227,7 +1227,7 @@ typedef struct smb_com_setattr_rsp {
 /* empty wct response to setattr */
 
 /*******************************************************/
-/* NT Transact structure defintions follow             */
+/* NT Transact structure definitions follow            */
 /* Currently only ioctl, acl (get security descriptor) */
 /* and notify are implemented                          */
 /*******************************************************/
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index cababd8a52d..cf18ee76559 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -914,8 +914,8 @@ undo_setattr:
 /*
  * If dentry->d_inode is null (usually meaning the cached dentry
  * is a negative dentry) then we would attempt a standard SMB delete, but
- * if that fails we can not attempt the fall back mechanisms on EACESS
- * but will return the EACESS to the caller.  Note that the VFS does not call
+ * if that fails we can not attempt the fall back mechanisms on EACCESS
+ * but will return the EACCESS to the caller. Note that the VFS does not call
  * unlink on negative dentries currently.
  */
 int cifs_unlink(struct inode *dir, struct dentry *dentry)
diff --git a/fs/cifs/smbdes.c b/fs/cifs/smbdes.c
index 224a1f47896..b6b6dcb500b 100644
--- a/fs/cifs/smbdes.c
+++ b/fs/cifs/smbdes.c
@@ -371,7 +371,7 @@ E_P24(unsigned char *p21, const unsigned char *c8, unsigned char *p24)
 	smbhash(p24 + 16, c8, p21 + 14, 1);
 }
 
-#if 0 /* currently unsued */
+#if 0 /* currently unused */
 static void
 D_P16(unsigned char *p14, unsigned char *in, unsigned char *out)
 {
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 229e7221816..2346895b3a7 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1920,7 +1920,7 @@ COMPATIBLE_IOCTL(TIOCSLTC)
 #endif
 #ifdef TIOCSTART
 /*
- * For these two we have defintions in ioctls.h and/or termios.h on
+ * For these two we have definitions in ioctls.h and/or termios.h on
  * some architectures but no actual implemention.  Some applications
  * like bash call them if they are defined in the headers, so we provide
  * entries here to avoid syslog message spew.
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index d22438ef767..0d23b52dd22 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -184,7 +184,7 @@ static int debugfs_create_by_name(const char *name, mode_t mode,
 /**
  * debugfs_create_file - create a file in the debugfs filesystem
  * @name: a pointer to a string containing the name of the file to create.
- * @mode: the permission that the file should have
+ * @mode: the permission that the file should have.
  * @parent: a pointer to the parent dentry for this file.  This should be a
  *          directory dentry if set.  If this paramater is NULL, then the
  *          file will be created in the root of the debugfs filesystem.
@@ -195,8 +195,8 @@ static int debugfs_create_by_name(const char *name, mode_t mode,
  *        this file.
  *
  * This is the basic "create a file" function for debugfs.  It allows for a
- * wide range of flexibility in createing a file, or a directory (if you
- * want to create a directory, the debugfs_create_dir() function is
+ * wide range of flexibility in creating a file, or a directory (if you want
+ * to create a directory, the debugfs_create_dir() function is
  * recommended to be used instead.)
  *
  * This function will return a pointer to a dentry if it succeeds.  This
diff --git a/fs/dlm/config.c b/fs/dlm/config.c
index fd9859f92fa..0df24385081 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 1c8bb8c3a82..375a2359b3b 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 c4dfa1dcc86..7b84c1dbc82 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 d01ca0a711d..826d3dc6e0a 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 eb507c453c5..9c0c1db1e10 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 d489fcc8671..c010ecfc0d2 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 70736eb4b51..52cab160893 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 b128775913b..84f70bfb0ba 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 c1775b84eba..8e0d00db004 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 55ea369f43a..052095cd592 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 16f682e26c0..b5f89aef3b2 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;
@@ -143,7 +143,7 @@ out:
 }
 EXPORT_SYMBOL_GPL(dlm_posix_lock);
 
-/* Returns failure iff a succesful lock operation should be canceled */
+/* Returns failure iff a successful lock operation should be canceled */
 static int dlm_plock_callback(struct plock_op *op)
 {
 	struct file *file;
@@ -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 67522c268c1..3c83a49a48a 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 7a2307c0891..a44fa22890e 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 ebce994ab0b..e73a4bb572a 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/exofs/Kbuild b/fs/exofs/Kbuild
index cc2d22db119..2d0f757fda3 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 c6718e4817f..b1b178e6171 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 5ec72e020b2..c35fd462398 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 6c10f747669..698a8636d39 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);
 
@@ -771,19 +778,28 @@ 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 +809,6 @@ 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 +816,7 @@ 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 +846,62 @@ 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 +910,7 @@ 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 +919,13 @@ 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 +943,12 @@ 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 +1014,30 @@ 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 +1052,7 @@ 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 +1061,8 @@ 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 +1089,28 @@ 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 +1128,11 @@ 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 +1148,7 @@ 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 +1185,16 @@ 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 +1203,19 @@ 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 +1232,12 @@ 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 +1251,7 @@ 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 +1268,26 @@ 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 00000000000..5bad01fa1f9
--- /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 4372542df28..00000000000
--- 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 00000000000..423033addd1
--- /dev/null
+++ b/fs/exofs/pnfs.h
@@ -0,0 +1,51 @@
+/*
+ * 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(CONFIG_PNFS)
+
+
+/* FIXME: move this file to: linux/exportfs/pnfs_osd_xdr.h */
+#include "../nfs/objlayout/pnfs_osd_xdr.h"
+
+#else /* defined(CONFIG_PNFS) */
+
+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 /* else defined(CONFIG_PNFS) */
+
+#endif /* __EXOFS_PNFS_H__ */
diff --git a/fs/exofs/super.c b/fs/exofs/super.c
index 9f500dec3b5..a1d1e77b12e 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/ext3/inode.c b/fs/ext3/inode.c
index 354ed3b47b3..2db95777890 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -2033,7 +2033,7 @@ static Indirect *ext3_find_shared(struct inode *inode, int depth,
 	int k, err;
 
 	*top = 0;
-	/* Make k index the deepest non-null offest + 1 */
+	/* Make k index the deepest non-null offset + 1 */
 	for (k = depth; k > 1 && !offsets[k-1]; k--)
 		;
 	partial = ext3_get_branch(inode, k, offsets, chain, &err);
diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig
index 9f2d45d75b1..9acf7e80813 100644
--- a/fs/ext4/Kconfig
+++ b/fs/ext4/Kconfig
@@ -26,6 +26,16 @@ config EXT4_FS
 
 	  If unsure, say N.
 
+config EXT4_USE_FOR_EXT23
+	bool "Use ext4 for ext2/ext3 file systems"
+	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/balloc.c b/fs/ext4/balloc.c
index 1d0418980f8..22bc7435d91 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 50784ef0756..4df8621ec31 100644
--- a/fs/ext4/block_validity.c
+++ b/fs/ext4/block_validity.c
@@ -160,7 +160,7 @@ 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 +228,7 @@ 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 8825515eedd..ab31e65d46d 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -376,6 +376,12 @@ 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
@@ -703,6 +709,13 @@ struct ext4_inode_info {
 	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 +763,7 @@ 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 +1338,6 @@ 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 +1396,15 @@ 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 *,
diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c
index 6a9409920de..b57e5c711b6 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 a2865980342..05eca817d70 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 715264b4bae..3a7928f825e 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -1007,7 +1007,8 @@ 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 +1762,9 @@ 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 +1960,6 @@ 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 +1975,8 @@ 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 +2043,11 @@ 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 +2072,7 @@ 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 +2163,7 @@ 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)
@@ -3064,6 +3060,8 @@ 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 +3089,8 @@ 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;
@@ -3319,8 +3319,8 @@ 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;
 	}
 
@@ -3329,10 +3329,16 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 	allocated = ext4_ext_get_actual_len(&newex);
 	set_buffer_new(bh_result);
 
-	/* Cache only when it is _not_ an uninitialized extent */
-	if ((flags & EXT4_GET_BLOCKS_UNINIT_EXT) == 0)
+	/*
+	 * 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 +3726,8 @@ 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 2b1531266ee..0b22497d92e 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 +84,13 @@ 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))
+		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 2c8caa51add..5352db1a308 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)
@@ -1052,7 +1012,7 @@ qsize_t ext4_get_reserved_space(struct inode *inode)
 		EXT4_I(inode)->i_reserved_meta_blocks;
 	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 
-	return total;
+	return (total << inode->i_blkbits);
 }
 /*
  * Calculate the number of metadata blocks need to reserve
@@ -1534,6 +1494,16 @@ 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 +1569,7 @@ 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 +1679,7 @@ 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 +1721,7 @@ 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 +1784,7 @@ 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
@@ -2600,7 +2570,6 @@ 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 +2727,7 @@ 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 +2757,7 @@ 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;
 
@@ -2933,7 +2902,7 @@ retry:
 		ret = write_cache_pages(mapping, wbc, __mpage_da_writepage,
 					&mpd);
 		/*
-		 * If we have a contigous extent of pages and we
+		 * If we have a contiguous extent of pages and we
 		 * haven't done the I/O yet, map the blocks and submit
 		 * them for I/O.
 		 */
@@ -3091,7 +3060,7 @@ 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))
@@ -4064,7 +4033,7 @@ static Indirect *ext4_find_shared(struct inode *inode, int depth,
 	int k, err;
 
 	*top = 0;
-	/* Make k index the deepest non-null offest + 1 */
+	/* Make k index the deepest non-null offset + 1 */
 	for (k = depth; k > 1 && !offsets[k-1]; k--)
 		;
 	partial = ext4_get_branch(inode, k, offsets, chain, &err);
@@ -4120,6 +4089,11 @@ 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 +4108,10 @@ 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 +4299,8 @@ 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 +4739,8 @@ 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 +4751,11 @@ 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 +4778,6 @@ 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;
 		}
@@ -4848,11 +4805,35 @@ 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 +4865,7 @@ 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 +4883,8 @@ 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 +4912,6 @@ 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 +4924,7 @@ 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 +5084,7 @@ 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 +5204,8 @@ 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;
@@ -5376,7 +5353,7 @@ static int ext4_index_trans_blocks(struct inode *inode, int nrblocks, int chunk)
  * worse case, the indexs blocks spread over different block groups
  *
  * If datablocks are discontiguous, they are possible to spread over
- * different block groups too. If they are contiugous, with flexbg,
+ * different block groups too. If they are contiuguous, with flexbg,
  * they could still across block group boundary.
  *
  * Also account for superblock, inode, quota and xattr blocks
@@ -5452,7 +5429,7 @@ int ext4_writepage_trans_blocks(struct inode *inode)
  * Calculate the journal credits for a chunk of data modification.
  *
  * This is called from DIO, fallocate or whoever calling
- * ext4_get_blocks() to map/allocate a chunk of contigous disk blocks.
+ * ext4_get_blocks() to map/allocate a chunk of contiguous disk blocks.
  *
  * journal buffers for data blocks are not included here, as DIO
  * and fallocate do no need to journal data buffers.
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index c1cdf613e72..b63d193126d 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 bba12824def..c1e19d5b598 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -142,7 +142,7 @@
  * 2 blocks and the order of allocation is >= sbi->s_mb_order2_reqs. The
  * value of s_mb_order2_reqs can be tuned via
  * /sys/fs/ext4/<partition>/mb_order2_req.  If the request len is equal to
- * stripe size (sbi->s_stripe), we try to search for contigous block in
+ * stripe size (sbi->s_stripe), we try to search for contiguous block in
  * stripe size. This should result in better allocation on RAID setups. If
  * not, we search in the specific group using bitmap for best extents. The
  * tunable min_to_scan and max_to_scan control the behaviour here.
@@ -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);
 	}
@@ -3006,6 +3011,24 @@ 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,
@@ -4290,6 +4313,7 @@ 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 +4446,24 @@ 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 +4473,16 @@ 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 +4490,32 @@ 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 +4591,8 @@ 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 +4631,7 @@ 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 +4651,8 @@ 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/migrate.c b/fs/ext4/migrate.c
index a93d5b80f3e..81415814b00 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 25b6b145736..82c415be87a 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 6d2c1b897fc..17a17e10dd6 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 3cfc343c41b..3b2c5541d8a 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 d4ca92aab51..8b58a144c31 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);
@@ -704,6 +706,8 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	spin_lock_init(&(ei->i_block_reservation_lock));
 	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;
 }
@@ -899,6 +903,12 @@ 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;
@@ -1079,7 +1089,8 @@ enum {
 	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 +1115,7 @@ 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"},
@@ -1144,6 +1156,8 @@ 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},
 };
 
@@ -1565,6 +1579,12 @@ 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 +1693,14 @@ 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) <<
@@ -2721,26 +2741,6 @@ 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 +3668,11 @@ 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 +3964,58 @@ 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);
+}
+#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);
+}
+#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 +4045,15 @@ 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 +4069,8 @@ 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 fed5b01d7a8..910bf9a59cb 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -482,9 +482,10 @@ 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 +833,8 @@ 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 +990,10 @@ 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 +1019,6 @@ 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);
diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c
index d4cfd6d2779..8896c1d4feb 100644
--- a/fs/jbd2/commit.c
+++ b/fs/jbd2/commit.c
@@ -636,6 +636,10 @@ 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);
 
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index fed85388ee8..b7ca3a92a4d 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);
@@ -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. */
diff --git a/fs/jffs2/compr.c b/fs/jffs2/compr.c
index f25e70c1b51..f0294410868 100644
--- a/fs/jffs2/compr.c
+++ b/fs/jffs2/compr.c
@@ -177,7 +177,7 @@ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f,
 		spin_unlock(&jffs2_compressor_list_lock);
 		break;
 	default:
-		printk(KERN_ERR "JFFS2: unknow compression mode.\n");
+		printk(KERN_ERR "JFFS2: unknown compression mode.\n");
 	}
  out:
 	if (ret == JFFS2_COMPR_NONE) {
diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c
index 1a80301004b..378991cfe40 100644
--- a/fs/jffs2/readinode.c
+++ b/fs/jffs2/readinode.c
@@ -931,7 +931,7 @@ static inline int read_unknown(struct jffs2_sb_info *c, struct jffs2_raw_node_re
  * Helper function for jffs2_get_inode_nodes().
  * The function detects whether more data should be read and reads it if yes.
  *
- * Returns: 0 on succes;
+ * Returns: 0 on success;
  * 	    negative error code on failure.
  */
 static int read_more(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *ref,
diff --git a/fs/jffs2/xattr.c b/fs/jffs2/xattr.c
index 082e844ab2d..4b107881acd 100644
--- a/fs/jffs2/xattr.c
+++ b/fs/jffs2/xattr.c
@@ -31,7 +31,7 @@
  *   is used to release xattr name/value pair and detach from c->xattrindex.
  * reclaim_xattr_datum(c)
  *   is used to reclaim xattr name/value pairs on the xattr name/value pair cache when
- *   memory usage by cache is over c->xdatum_mem_threshold. Currentry, this threshold 
+ *   memory usage by cache is over c->xdatum_mem_threshold. Currently, this threshold 
  *   is hard coded as 32KiB.
  * do_verify_xattr_datum(c, xd)
  *   is used to load the xdatum informations without name/value pair from the medium.
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index 2bc7d8aa574..d9b031cf69f 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -755,7 +755,7 @@ int dbAlloc(struct inode *ip, s64 hint, s64 nblocks, s64 * results)
 	 * allocation group.
 	 */
 	if ((blkno & (bmp->db_agsize - 1)) == 0)
-		/* check if the AG is currenly being written to.
+		/* check if the AG is currently being written to.
 		 * if so, call dbNextAG() to find a non-busy
 		 * AG with sufficient free space.
 		 */
@@ -3337,7 +3337,7 @@ int dbExtendFS(struct inode *ipbmap, s64 blkno,	s64 nblocks)
 	for (i = 0, n = 0; i < agno; n++) {
 		bmp->db_agfree[n] = 0;	/* init collection point */
 
-		/* coalesce cotiguous k AGs; */
+		/* coalesce contiguous k AGs; */
 		for (j = 0; j < k && i < agno; j++, i++) {
 			/* merge AGi to AGn */
 			bmp->db_agfree[n] += bmp->db_agfree[i];
diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c
index 0d58caf4a6e..ec8f45f12e0 100644
--- a/fs/ncpfs/ioctl.c
+++ b/fs/ncpfs/ioctl.c
@@ -835,7 +835,7 @@ static int ncp_ioctl_need_write(unsigned int cmd)
 	case NCP_IOC_SETROOT:
 		return 0;
 	default:
-		/* unkown IOCTL command, assume write */
+		/* unknown IOCTL command, assume write */
 		return 1;
 	}
 }
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 1d1d1a2765d..5ef5f365a5c 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -741,10 +741,6 @@ SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname,
 
 	/* create/update an inode mark */
 	ret = inotify_update_watch(group, inode, mask);
-	if (unlikely(ret))
-		goto path_put_and_out;
-
-path_put_and_out:
 	path_put(&path);
 fput_and_out:
 	fput_light(filp, fput_needed);
diff --git a/fs/ntfs/compress.c b/fs/ntfs/compress.c
index 9669541d011..08f7530e934 100644
--- a/fs/ntfs/compress.c
+++ b/fs/ntfs/compress.c
@@ -927,7 +927,7 @@ lock_retry_remap:
 		return 0;
 
 	ntfs_debug("Failed. Returning error code %s.", err == -EOVERFLOW ?
-			"EOVERFLOW" : (!err ? "EIO" : "unkown error"));
+			"EOVERFLOW" : (!err ? "EIO" : "unknown error"));
 	return err < 0 ? err : -EIO;
 
 read_err:
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index 663c0e341f8..43179ddd336 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -399,7 +399,7 @@ static inline void ntfs_fault_in_pages_readable_iovec(const struct iovec *iov,
  * @cached_page: allocated but as yet unused page
  * @lru_pvec:	lru-buffering pagevec of caller
  *
- * Obtain @nr_pages locked page cache pages from the mapping @maping and
+ * Obtain @nr_pages locked page cache pages from the mapping @mapping and
  * starting at index @index.
  *
  * If a page is newly created, increment its refcount and add it to the
@@ -1281,7 +1281,7 @@ rl_not_mapped_enoent:
 
 /*
  * Copy as much as we can into the pages and return the number of bytes which
- * were sucessfully copied.  If a fault is encountered then clear the pages
+ * were successfully copied.  If a fault is encountered then clear the pages
  * out to (ofs + bytes) and return the number of bytes which were copied.
  */
 static inline size_t ntfs_copy_from_user(struct page **pages,
diff --git a/fs/ntfs/logfile.c b/fs/ntfs/logfile.c
index 89b02985c05..4dadcdf3d45 100644
--- a/fs/ntfs/logfile.c
+++ b/fs/ntfs/logfile.c
@@ -338,7 +338,7 @@ err_out:
  * copy of the complete multi sector transfer deprotected page.  On failure,
  * *@wrp is undefined.
  *
- * Simillarly, if @lsn is not NULL, on succes *@lsn will be set to the current
+ * Simillarly, if @lsn is not NULL, on success *@lsn will be set to the current
  * logfile lsn according to this restart page.  On failure, *@lsn is undefined.
  *
  * The following error codes are defined:
diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c
index 38a42f5d59f..7c7198a5bc9 100644
--- a/fs/ocfs2/alloc.c
+++ b/fs/ocfs2/alloc.c
@@ -2398,7 +2398,7 @@ static int ocfs2_leftmost_rec_contains(struct ocfs2_extent_list *el, u32 cpos)
  *
  * The array is assumed to be large enough to hold an entire path (tree depth).
  *
- * Upon succesful return from this function:
+ * Upon successful return from this function:
  *
  * - The 'right_path' array will contain a path to the leaf block
  *   whose range contains e_cpos.
diff --git a/fs/ocfs2/blockcheck.c b/fs/ocfs2/blockcheck.c
index a1163b8b417..b7428c5d0d3 100644
--- a/fs/ocfs2/blockcheck.c
+++ b/fs/ocfs2/blockcheck.c
@@ -47,7 +47,7 @@
  * Calculate the bit offset in the hamming code buffer based on the bit's
  * offset in the data buffer.  Since the hamming code reserves all
  * power-of-two bits for parity, the data bit number and the code bit
- * number are offest by all the parity bits beforehand.
+ * number are offset by all the parity bits beforehand.
  *
  * Recall that bit numbers in hamming code are 1-based.  This function
  * takes the 0-based data bit from the caller.
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index 83bcaf266b3..03ccf9a7b1f 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -2586,7 +2586,7 @@ fail:
 	 * is complete everywhere.  if the target dies while this is
 	 * going on, some nodes could potentially see the target as the
 	 * master, so it is important that my recovery finds the migration
-	 * mle and sets the master to UNKNONWN. */
+	 * mle and sets the master to UNKNOWN. */
 
 
 	/* wait for new node to assert master */
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index 0d38d67194c..c5e4a49e3a1 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -1855,7 +1855,7 @@ int ocfs2_file_lock(struct file *file, int ex, int trylock)
 		 * outstanding lock request, so a cancel convert is
 		 * required. We intentionally overwrite 'ret' - if the
 		 * cancel fails and the lock was granted, it's easier
-		 * to just bubble sucess back up to the user.
+		 * to just bubble success back up to the user.
 		 */
 		ret = ocfs2_flock_handle_signal(lockres, level);
 	} else if (!ret && (level > lockres->l_level)) {
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 54c16b66327..bf34c491ae9 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -659,7 +659,7 @@ static int __ocfs2_journal_access(handle_t *handle,
 
 	default:
 		status = -EINVAL;
-		mlog(ML_ERROR, "Uknown access type!\n");
+		mlog(ML_ERROR, "Unknown access type!\n");
 	}
 	if (!status && ocfs2_meta_ecc(osb) && triggers)
 		jbd2_journal_set_triggers(bh, &triggers->ot_triggers);
diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c
index 3a0df7a1b81..30967e3f5e4 100644
--- a/fs/ocfs2/refcounttree.c
+++ b/fs/ocfs2/refcounttree.c
@@ -2431,7 +2431,7 @@ out:
  * we gonna touch and whether we need to create new blocks.
  *
  * Normally the refcount blocks store these refcount should be
- * continguous also, so that we can get the number easily.
+ * contiguous also, so that we can get the number easily.
  * As for meta_ac, we will at most add split 2 refcount record and
  * 2 more refcount block, so just check it in a rough way.
  *
diff --git a/fs/omfs/bitmap.c b/fs/omfs/bitmap.c
index e1c0ec0ae98..082234581d0 100644
--- a/fs/omfs/bitmap.c
+++ b/fs/omfs/bitmap.c
@@ -85,7 +85,7 @@ out:
 }
 
 /*
- * Tries to allocate exactly one block.  Returns true if sucessful.
+ * Tries to allocate exactly one block.  Returns true if successful.
  */
 int omfs_allocate_block(struct super_block *sb, u64 block)
 {
diff --git a/fs/qnx4/bitmap.c b/fs/qnx4/bitmap.c
index 0afba069d56..32f5d131a64 100644
--- a/fs/qnx4/bitmap.c
+++ b/fs/qnx4/bitmap.c
@@ -67,7 +67,7 @@ unsigned long qnx4_count_free_blocks(struct super_block *sb)
 
 	while (total < size) {
 		if ((bh = sb_bread(sb, start + offset)) == NULL) {
-			printk("qnx4: I/O error in counting free blocks\n");
+			printk(KERN_ERR "qnx4: I/O error in counting free blocks\n");
 			break;
 		}
 		count_bits(bh->b_data, size - total, &total_free);
diff --git a/fs/qnx4/dir.c b/fs/qnx4/dir.c
index 86cc39cb139..6f30c3d5bcb 100644
--- a/fs/qnx4/dir.c
+++ b/fs/qnx4/dir.c
@@ -26,8 +26,8 @@ static int qnx4_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	int ix, ino;
 	int size;
 
-	QNX4DEBUG(("qnx4_readdir:i_size = %ld\n", (long) inode->i_size));
-	QNX4DEBUG(("filp->f_pos         = %ld\n", (long) filp->f_pos));
+	QNX4DEBUG((KERN_INFO "qnx4_readdir:i_size = %ld\n", (long) inode->i_size));
+	QNX4DEBUG((KERN_INFO "filp->f_pos         = %ld\n", (long) filp->f_pos));
 
 	lock_kernel();
 
@@ -50,7 +50,7 @@ static int qnx4_readdir(struct file *filp, void *dirent, filldir_t filldir)
 					size = QNX4_NAME_MAX;
 
 				if ( ( de->di_status & (QNX4_FILE_USED|QNX4_FILE_LINK) ) != 0 ) {
-					QNX4DEBUG(("qnx4_readdir:%.*s\n", size, de->di_fname));
+					QNX4DEBUG((KERN_INFO "qnx4_readdir:%.*s\n", size, de->di_fname));
 					if ( ( de->di_status & QNX4_FILE_LINK ) == 0 )
 						ino = blknum * QNX4_INODES_PER_BLOCK + ix - 1;
 					else {
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index d2cd1798d8c..449f5a66dd3 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -107,7 +107,7 @@ static int qnx4_get_block( struct inode *inode, sector_t iblock, struct buffer_h
 {
 	unsigned long phys;
 
-	QNX4DEBUG(("qnx4: qnx4_get_block inode=[%ld] iblock=[%ld]\n",inode->i_ino,iblock));
+	QNX4DEBUG((KERN_INFO "qnx4: qnx4_get_block inode=[%ld] iblock=[%ld]\n",inode->i_ino,iblock));
 
 	phys = qnx4_block_map( inode, iblock );
 	if ( phys ) {
@@ -142,12 +142,12 @@ unsigned long qnx4_block_map( struct inode *inode, long iblock )
 				// read next xtnt block.
 				bh = sb_bread(inode->i_sb, i_xblk - 1);
 				if ( !bh ) {
-					QNX4DEBUG(("qnx4: I/O error reading xtnt block [%ld])\n", i_xblk - 1));
+					QNX4DEBUG((KERN_ERR "qnx4: I/O error reading xtnt block [%ld])\n", i_xblk - 1));
 					return -EIO;
 				}
 				xblk = (struct qnx4_xblk*)bh->b_data;
 				if ( memcmp( xblk->xblk_signature, "IamXblk", 7 ) ) {
-					QNX4DEBUG(("qnx4: block at %ld is not a valid xtnt\n", qnx4_inode->i_xblk));
+					QNX4DEBUG((KERN_ERR "qnx4: block at %ld is not a valid xtnt\n", qnx4_inode->i_xblk));
 					return -EIO;
 				}
 			}
@@ -168,7 +168,7 @@ unsigned long qnx4_block_map( struct inode *inode, long iblock )
 			brelse( bh );
 	}
 
-	QNX4DEBUG(("qnx4: mapping block %ld of inode %ld = %ld\n",iblock,inode->i_ino,block));
+	QNX4DEBUG((KERN_INFO "qnx4: mapping block %ld of inode %ld = %ld\n",iblock,inode->i_ino,block));
 	return block;
 }
 
@@ -209,7 +209,7 @@ static const char *qnx4_checkroot(struct super_block *sb)
 	if (*(qnx4_sb(sb)->sb->RootDir.di_fname) != '/') {
 		return "no qnx4 filesystem (no root dir).";
 	} else {
-		QNX4DEBUG(("QNX4 filesystem found on dev %s.\n", sb->s_id));
+		QNX4DEBUG((KERN_NOTICE "QNX4 filesystem found on dev %s.\n", sb->s_id));
 		rd = le32_to_cpu(qnx4_sb(sb)->sb->RootDir.di_first_xtnt.xtnt_blk) - 1;
 		rl = le32_to_cpu(qnx4_sb(sb)->sb->RootDir.di_first_xtnt.xtnt_size);
 		for (j = 0; j < rl; j++) {
@@ -220,7 +220,7 @@ static const char *qnx4_checkroot(struct super_block *sb)
 			for (i = 0; i < QNX4_INODES_PER_BLOCK; i++) {
 				rootdir = (struct qnx4_inode_entry *) (bh->b_data + i * QNX4_DIR_ENTRY_SIZE);
 				if (rootdir->di_fname != NULL) {
-					QNX4DEBUG(("Rootdir entry found : [%s]\n", rootdir->di_fname));
+					QNX4DEBUG((KERN_INFO "rootdir entry found : [%s]\n", rootdir->di_fname));
 					if (!strncmp(rootdir->di_fname, QNX4_BMNAME, sizeof QNX4_BMNAME)) {
 						found = 1;
 						qnx4_sb(sb)->BitMap = kmalloc( sizeof( struct qnx4_inode_entry ), GFP_KERNEL );
@@ -265,12 +265,12 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent)
 	   if we don't belong here... */
 	bh = sb_bread(s, 1);
 	if (!bh) {
-		printk("qnx4: unable to read the superblock\n");
+		printk(KERN_ERR "qnx4: unable to read the superblock\n");
 		goto outnobh;
 	}
 	if ( le32_to_cpup((__le32*) bh->b_data) != QNX4_SUPER_MAGIC ) {
 		if (!silent)
-			printk("qnx4: wrong fsid in superblock.\n");
+			printk(KERN_ERR "qnx4: wrong fsid in superblock.\n");
 		goto out;
 	}
 	s->s_op = &qnx4_sops;
@@ -284,14 +284,14 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent)
 	errmsg = qnx4_checkroot(s);
 	if (errmsg != NULL) {
  		if (!silent)
- 			printk("qnx4: %s\n", errmsg);
+			printk(KERN_ERR "qnx4: %s\n", errmsg);
 		goto out;
 	}
 
  	/* does root not have inode number QNX4_ROOT_INO ?? */
 	root = qnx4_iget(s, QNX4_ROOT_INO * QNX4_INODES_PER_BLOCK);
 	if (IS_ERR(root)) {
- 		printk("qnx4: get inode failed\n");
+		printk(KERN_ERR "qnx4: get inode failed\n");
 		ret = PTR_ERR(root);
  		goto out;
  	}
@@ -374,7 +374,7 @@ struct inode *qnx4_iget(struct super_block *sb, unsigned long ino)
 	qnx4_inode = qnx4_raw_inode(inode);
 	inode->i_mode = 0;
 
-	QNX4DEBUG(("Reading inode : [%d]\n", ino));
+	QNX4DEBUG((KERN_INFO "reading inode : [%d]\n", ino));
 	if (!ino) {
 		printk(KERN_ERR "qnx4: bad inode number on dev %s: %lu is "
 				"out of range\n",
@@ -385,7 +385,7 @@ struct inode *qnx4_iget(struct super_block *sb, unsigned long ino)
 	block = ino / QNX4_INODES_PER_BLOCK;
 
 	if (!(bh = sb_bread(sb, block))) {
-		printk("qnx4: major problem: unable to read inode from dev "
+		printk(KERN_ERR "qnx4: major problem: unable to read inode from dev "
 		       "%s\n", sb->s_id);
 		iget_failed(inode);
 		return ERR_PTR(-EIO);
@@ -499,7 +499,7 @@ static int __init init_qnx4_fs(void)
 		return err;
 	}
 
-	printk("QNX4 filesystem 0.2.3 registered.\n");
+	printk(KERN_INFO "QNX4 filesystem 0.2.3 registered.\n");
 	return 0;
 }
 
diff --git a/fs/qnx4/namei.c b/fs/qnx4/namei.c
index ae1e7edbacd..58703ebba87 100644
--- a/fs/qnx4/namei.c
+++ b/fs/qnx4/namei.c
@@ -30,7 +30,7 @@ static int qnx4_match(int len, const char *name,
 	int namelen, thislen;
 
 	if (bh == NULL) {
-		printk("qnx4: matching unassigned buffer !\n");
+		printk(KERN_WARNING "qnx4: matching unassigned buffer !\n");
 		return 0;
 	}
 	de = (struct qnx4_inode_entry *) (bh->b_data + *offset);
@@ -66,7 +66,7 @@ static struct buffer_head *qnx4_find_entry(int len, struct inode *dir,
 
 	*res_dir = NULL;
 	if (!dir->i_sb) {
-		printk("qnx4: no superblock on dir.\n");
+		printk(KERN_WARNING "qnx4: no superblock on dir.\n");
 		return NULL;
 	}
 	bh = NULL;
@@ -124,7 +124,7 @@ struct dentry * qnx4_lookup(struct inode *dir, struct dentry *dentry, struct nam
 	foundinode = qnx4_iget(dir->i_sb, ino);
 	if (IS_ERR(foundinode)) {
 		unlock_kernel();
-		QNX4DEBUG(("qnx4: lookup->iget -> error %ld\n",
+		QNX4DEBUG((KERN_ERR "qnx4: lookup->iget -> error %ld\n",
 			   PTR_ERR(foundinode)));
 		return ERR_CAST(foundinode);
 	}
diff --git a/fs/reiserfs/Makefile b/fs/reiserfs/Makefile
index 7c5ab6330dd..6a9e30c041d 100644
--- a/fs/reiserfs/Makefile
+++ b/fs/reiserfs/Makefile
@@ -7,7 +7,7 @@ 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
+		 item_ops.o ioctl.o procfs.o xattr.o lock.o
 
 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 e716161ab32..68549570718 100644
--- a/fs/reiserfs/bitmap.c
+++ b/fs/reiserfs/bitmap.c
@@ -1249,14 +1249,18 @@ struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb,
 	else if (bitmap == 0)
 		block = (REISERFS_DISK_OFFSET_IN_BYTES >> sb->s_blocksize_bits) + 1;
 
+	reiserfs_write_unlock(sb);
 	bh = sb_bread(sb, block);
+	reiserfs_write_lock(sb);
 	if (bh == NULL)
 		reiserfs_warning(sb, "sh-2029: %s: bitmap block (#%u) "
 		                 "reading failed", __func__, block);
 	else {
 		if (buffer_locked(bh)) {
 			PROC_INFO_INC(sb, scan_bitmap.wait);
+			reiserfs_write_unlock(sb);
 			__wait_on_buffer(bh);
+			reiserfs_write_lock(sb);
 		}
 		BUG_ON(!buffer_uptodate(bh));
 		BUG_ON(atomic_read(&bh->b_count) == 0);
diff --git a/fs/reiserfs/dir.c b/fs/reiserfs/dir.c
index 6d2668fdc38..c094f58c744 100644
--- a/fs/reiserfs/dir.c
+++ b/fs/reiserfs/dir.c
@@ -20,7 +20,7 @@ const struct file_operations reiserfs_dir_operations = {
 	.read = generic_read_dir,
 	.readdir = reiserfs_readdir,
 	.fsync = reiserfs_dir_fsync,
-	.ioctl = reiserfs_ioctl,
+	.unlocked_ioctl = reiserfs_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl = reiserfs_compat_ioctl,
 #endif
@@ -174,14 +174,22 @@ int reiserfs_readdir_dentry(struct dentry *dentry, void *dirent,
 				// user space buffer is swapped out. At that time
 				// entry can move to somewhere else
 				memcpy(local_buf, d_name, d_reclen);
+
+				/*
+				 * Since filldir might sleep, we can release
+				 * the write lock here for other waiters
+				 */
+				reiserfs_write_unlock(inode->i_sb);
 				if (filldir
 				    (dirent, local_buf, d_reclen, d_off, d_ino,
 				     DT_UNKNOWN) < 0) {
+					reiserfs_write_lock(inode->i_sb);
 					if (local_buf != small_buf) {
 						kfree(local_buf);
 					}
 					goto end;
 				}
+				reiserfs_write_lock(inode->i_sb);
 				if (local_buf != small_buf) {
 					kfree(local_buf);
 				}
diff --git a/fs/reiserfs/do_balan.c b/fs/reiserfs/do_balan.c
index 128d3f7c8aa..60c08044066 100644
--- a/fs/reiserfs/do_balan.c
+++ b/fs/reiserfs/do_balan.c
@@ -21,14 +21,6 @@
 #include <linux/buffer_head.h>
 #include <linux/kernel.h>
 
-#ifdef CONFIG_REISERFS_CHECK
-
-struct tree_balance *cur_tb = NULL;	/* detects whether more than one
-					   copy of tb exists as a means
-					   of checking whether schedule
-					   is interrupting do_balance */
-#endif
-
 static inline void buffer_info_init_left(struct tree_balance *tb,
                                          struct buffer_info *bi)
 {
@@ -1840,11 +1832,12 @@ static int check_before_balancing(struct tree_balance *tb)
 {
 	int retval = 0;
 
-	if (cur_tb) {
+	if (REISERFS_SB(tb->tb_sb)->cur_tb) {
 		reiserfs_panic(tb->tb_sb, "vs-12335", "suspect that schedule "
 			       "occurred based on cur_tb not being null at "
 			       "this point in code. do_balance cannot properly "
-			       "handle schedule occurring while it runs.");
+			       "handle concurrent tree accesses on a same "
+			       "mount point.");
 	}
 
 	/* double check that buffers that we will modify are unlocked. (fix_nodes should already have
@@ -1986,7 +1979,7 @@ static inline void do_balance_starts(struct tree_balance *tb)
 	     "check");*/
 	RFALSE(check_before_balancing(tb), "PAP-12340: locked buffers in TB");
 #ifdef CONFIG_REISERFS_CHECK
-	cur_tb = tb;
+	REISERFS_SB(tb->tb_sb)->cur_tb = tb;
 #endif
 }
 
@@ -1996,7 +1989,7 @@ static inline void do_balance_completed(struct tree_balance *tb)
 #ifdef CONFIG_REISERFS_CHECK
 	check_leaf_level(tb);
 	check_internal_levels(tb);
-	cur_tb = NULL;
+	REISERFS_SB(tb->tb_sb)->cur_tb = NULL;
 #endif
 
 	/* reiserfs_free_block is no longer schedule safe.  So, we need to
diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c
index 9f436668b7f..da2dba082e2 100644
--- a/fs/reiserfs/file.c
+++ b/fs/reiserfs/file.c
@@ -284,7 +284,7 @@ static ssize_t reiserfs_file_write(struct file *file,	/* the file we are going t
 const struct file_operations reiserfs_file_operations = {
 	.read = do_sync_read,
 	.write = reiserfs_file_write,
-	.ioctl = reiserfs_ioctl,
+	.unlocked_ioctl = reiserfs_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl = reiserfs_compat_ioctl,
 #endif
diff --git a/fs/reiserfs/fix_node.c b/fs/reiserfs/fix_node.c
index 5e5a4e6fbaf..6591cb21edf 100644
--- a/fs/reiserfs/fix_node.c
+++ b/fs/reiserfs/fix_node.c
@@ -563,9 +563,6 @@ static int get_num_ver(int mode, struct tree_balance *tb, int h,
 	return needed_nodes;
 }
 
-#ifdef CONFIG_REISERFS_CHECK
-extern struct tree_balance *cur_tb;
-#endif
 
 /* Set parameters for balancing.
  * Performs write of results of analysis of balancing into structure tb,
@@ -834,7 +831,7 @@ static int get_empty_nodes(struct tree_balance *tb, int h)
 		RFALSE(buffer_dirty(new_bh) ||
 		       buffer_journaled(new_bh) ||
 		       buffer_journal_dirty(new_bh),
-		       "PAP-8140: journlaled or dirty buffer %b for the new block",
+		       "PAP-8140: journaled or dirty buffer %b for the new block",
 		       new_bh);
 
 		/* Put empty buffers into the array. */
@@ -1022,7 +1019,11 @@ static int get_far_parent(struct tree_balance *tb,
 	/* Check whether the common parent is locked. */
 
 	if (buffer_locked(*pcom_father)) {
+
+		/* Release the write lock while the buffer is busy */
+		reiserfs_write_unlock(tb->tb_sb);
 		__wait_on_buffer(*pcom_father);
+		reiserfs_write_lock(tb->tb_sb);
 		if (FILESYSTEM_CHANGED_TB(tb)) {
 			brelse(*pcom_father);
 			return REPEAT_SEARCH;
@@ -1927,7 +1928,9 @@ static int get_direct_parent(struct tree_balance *tb, int h)
 		return REPEAT_SEARCH;
 
 	if (buffer_locked(bh)) {
+		reiserfs_write_unlock(tb->tb_sb);
 		__wait_on_buffer(bh);
+		reiserfs_write_lock(tb->tb_sb);
 		if (FILESYSTEM_CHANGED_TB(tb))
 			return REPEAT_SEARCH;
 	}
@@ -1965,7 +1968,9 @@ static int get_neighbors(struct tree_balance *tb, int h)
 		     tb->FL[h]) ? tb->lkey[h] : B_NR_ITEMS(tb->
 								       FL[h]);
 		son_number = B_N_CHILD_NUM(tb->FL[h], child_position);
+		reiserfs_write_unlock(sb);
 		bh = sb_bread(sb, son_number);
+		reiserfs_write_lock(sb);
 		if (!bh)
 			return IO_ERROR;
 		if (FILESYSTEM_CHANGED_TB(tb)) {
@@ -2003,7 +2008,9 @@ static int get_neighbors(struct tree_balance *tb, int h)
 		child_position =
 		    (bh == tb->FR[h]) ? tb->rkey[h] + 1 : 0;
 		son_number = B_N_CHILD_NUM(tb->FR[h], child_position);
+		reiserfs_write_unlock(sb);
 		bh = sb_bread(sb, son_number);
+		reiserfs_write_lock(sb);
 		if (!bh)
 			return IO_ERROR;
 		if (FILESYSTEM_CHANGED_TB(tb)) {
@@ -2278,7 +2285,9 @@ static int wait_tb_buffers_until_unlocked(struct tree_balance *tb)
 				    REPEAT_SEARCH : CARRY_ON;
 			}
 #endif
+			reiserfs_write_unlock(tb->tb_sb);
 			__wait_on_buffer(locked);
+			reiserfs_write_lock(tb->tb_sb);
 			if (FILESYSTEM_CHANGED_TB(tb))
 				return REPEAT_SEARCH;
 		}
@@ -2349,12 +2358,14 @@ int fix_nodes(int op_mode, struct tree_balance *tb,
 
 	/* if it possible in indirect_to_direct conversion */
 	if (buffer_locked(tbS0)) {
+		reiserfs_write_unlock(tb->tb_sb);
 		__wait_on_buffer(tbS0);
+		reiserfs_write_lock(tb->tb_sb);
 		if (FILESYSTEM_CHANGED_TB(tb))
 			return REPEAT_SEARCH;
 	}
 #ifdef CONFIG_REISERFS_CHECK
-	if (cur_tb) {
+	if (REISERFS_SB(tb->tb_sb)->cur_tb) {
 		print_cur_tb("fix_nodes");
 		reiserfs_panic(tb->tb_sb, "PAP-8305",
 			       "there is pending do_balance");
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index a14d6cd9eed..3a28e7751b3 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -251,7 +251,6 @@ static int _get_block_create_0(struct inode *inode, sector_t block,
 	struct cpu_key key;
 	struct buffer_head *bh;
 	struct item_head *ih, tmp_ih;
-	int fs_gen;
 	b_blocknr_t blocknr;
 	char *p = NULL;
 	int chars;
@@ -265,7 +264,6 @@ static int _get_block_create_0(struct inode *inode, sector_t block,
 		     (loff_t) block * inode->i_sb->s_blocksize + 1, TYPE_ANY,
 		     3);
 
-      research:
 	result = search_for_position_by_key(inode->i_sb, &key, &path);
 	if (result != POSITION_FOUND) {
 		pathrelse(&path);
@@ -340,7 +338,6 @@ static int _get_block_create_0(struct inode *inode, sector_t block,
 	}
 	// read file tail into part of page
 	offset = (cpu_key_k_offset(&key) - 1) & (PAGE_CACHE_SIZE - 1);
-	fs_gen = get_generation(inode->i_sb);
 	copy_item_head(&tmp_ih, ih);
 
 	/* we only want to kmap if we are reading the tail into the page.
@@ -348,13 +345,9 @@ static int _get_block_create_0(struct inode *inode, sector_t block,
 	 ** sure we need to.  But, this means the item might move if
 	 ** kmap schedules
 	 */
-	if (!p) {
+	if (!p)
 		p = (char *)kmap(bh_result->b_page);
-		if (fs_changed(fs_gen, inode->i_sb)
-		    && item_moved(&tmp_ih, &path)) {
-			goto research;
-		}
-	}
+
 	p += offset;
 	memset(p, 0, inode->i_sb->s_blocksize);
 	do {
@@ -489,10 +482,14 @@ static int reiserfs_get_blocks_direct_io(struct inode *inode,
 	   disappeared */
 	if (REISERFS_I(inode)->i_flags & i_pack_on_close_mask) {
 		int err;
-		lock_kernel();
+
+		reiserfs_write_lock(inode->i_sb);
+
 		err = reiserfs_commit_for_inode(inode);
 		REISERFS_I(inode)->i_flags &= ~i_pack_on_close_mask;
-		unlock_kernel();
+
+		reiserfs_write_unlock(inode->i_sb);
+
 		if (err < 0)
 			ret = err;
 	}
@@ -601,6 +598,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 	__le32 *item;
 	int done;
 	int fs_gen;
+	int lock_depth;
 	struct reiserfs_transaction_handle *th = NULL;
 	/* space reserved in transaction batch:
 	   . 3 balancings in direct->indirect conversion
@@ -616,12 +614,11 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 	loff_t new_offset =
 	    (((loff_t) block) << inode->i_sb->s_blocksize_bits) + 1;
 
-	/* bad.... */
-	reiserfs_write_lock(inode->i_sb);
+	lock_depth = reiserfs_write_lock_once(inode->i_sb);
 	version = get_inode_item_key_version(inode);
 
 	if (!file_capable(inode, block)) {
-		reiserfs_write_unlock(inode->i_sb);
+		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
 		return -EFBIG;
 	}
 
@@ -633,7 +630,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 		/* find number of block-th logical block of the file */
 		ret = _get_block_create_0(inode, block, bh_result,
 					  create | GET_BLOCK_READ_DIRECT);
-		reiserfs_write_unlock(inode->i_sb);
+		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
 		return ret;
 	}
 	/*
@@ -751,7 +748,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 		if (!dangle && th)
 			retval = reiserfs_end_persistent_transaction(th);
 
-		reiserfs_write_unlock(inode->i_sb);
+		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
 
 		/* the item was found, so new blocks were not added to the file
 		 ** there is no need to make sure the inode is updated with this
@@ -935,7 +932,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 			if (blocks_needed == 1) {
 				un = &unf_single;
 			} else {
-				un = kzalloc(min(blocks_needed, max_to_insert) * UNFM_P_SIZE, GFP_ATOMIC);	// We need to avoid scheduling.
+				un = kzalloc(min(blocks_needed, max_to_insert) * UNFM_P_SIZE, GFP_NOFS);
 				if (!un) {
 					un = &unf_single;
 					blocks_needed = 1;
@@ -997,10 +994,16 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 			if (retval)
 				goto failure;
 		}
-		/* inserting indirect pointers for a hole can take a
-		 ** long time.  reschedule if needed
+		/*
+		 * inserting indirect pointers for a hole can take a
+		 * long time.  reschedule if needed and also release the write
+		 * lock for others.
 		 */
-		cond_resched();
+		if (need_resched()) {
+			reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+			schedule();
+			lock_depth = reiserfs_write_lock_once(inode->i_sb);
+		}
 
 		retval = search_for_position_by_key(inode->i_sb, &key, &path);
 		if (retval == IO_ERROR) {
@@ -1035,7 +1038,7 @@ int reiserfs_get_block(struct inode *inode, sector_t block,
 			retval = err;
 	}
 
-	reiserfs_write_unlock(inode->i_sb);
+	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
 	reiserfs_check_path(&path);
 	return retval;
 }
@@ -2072,8 +2075,9 @@ int reiserfs_truncate_file(struct inode *inode, int update_timestamps)
 	int error;
 	struct buffer_head *bh = NULL;
 	int err2;
+	int lock_depth;
 
-	reiserfs_write_lock(inode->i_sb);
+	lock_depth = reiserfs_write_lock_once(inode->i_sb);
 
 	if (inode->i_size > 0) {
 		error = grab_tail_page(inode, &page, &bh);
@@ -2142,14 +2146,17 @@ int reiserfs_truncate_file(struct inode *inode, int update_timestamps)
 		page_cache_release(page);
 	}
 
-	reiserfs_write_unlock(inode->i_sb);
+	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+
 	return 0;
       out:
 	if (page) {
 		unlock_page(page);
 		page_cache_release(page);
 	}
-	reiserfs_write_unlock(inode->i_sb);
+
+	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+
 	return error;
 }
 
@@ -2608,7 +2615,10 @@ int reiserfs_prepare_write(struct file *f, struct page *page,
 	int ret;
 	int old_ref = 0;
 
+	reiserfs_write_unlock(inode->i_sb);
 	reiserfs_wait_on_write_block(inode->i_sb);
+	reiserfs_write_lock(inode->i_sb);
+
 	fix_tail_page_for_writing(page);
 	if (reiserfs_transaction_running(inode->i_sb)) {
 		struct reiserfs_transaction_handle *th;
@@ -2664,6 +2674,8 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 	int update_sd = 0;
 	struct reiserfs_transaction_handle *th;
 	unsigned start;
+	int lock_depth = 0;
+	bool locked = false;
 
 	if ((unsigned long)fsdata & AOP_FLAG_CONT_EXPAND)
 		pos ++;
@@ -2690,9 +2702,11 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 	 ** to do the i_size updates here.
 	 */
 	pos += copied;
+
 	if (pos > inode->i_size) {
 		struct reiserfs_transaction_handle myth;
-		reiserfs_write_lock(inode->i_sb);
+		lock_depth = reiserfs_write_lock_once(inode->i_sb);
+		locked = true;
 		/* If the file have grown beyond the border where it
 		   can have a tail, unmark it as needing a tail
 		   packing */
@@ -2703,10 +2717,9 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 			REISERFS_I(inode)->i_flags &= ~i_pack_on_close_mask;
 
 		ret = journal_begin(&myth, inode->i_sb, 1);
-		if (ret) {
-			reiserfs_write_unlock(inode->i_sb);
+		if (ret)
 			goto journal_error;
-		}
+
 		reiserfs_update_inode_transaction(inode);
 		inode->i_size = pos;
 		/*
@@ -2718,34 +2731,36 @@ static int reiserfs_write_end(struct file *file, struct address_space *mapping,
 		reiserfs_update_sd(&myth, inode);
 		update_sd = 1;
 		ret = journal_end(&myth, inode->i_sb, 1);
-		reiserfs_write_unlock(inode->i_sb);
 		if (ret)
 			goto journal_error;
 	}
 	if (th) {
-		reiserfs_write_lock(inode->i_sb);
+		if (!locked) {
+			lock_depth = reiserfs_write_lock_once(inode->i_sb);
+			locked = true;
+		}
 		if (!update_sd)
 			mark_inode_dirty(inode);
 		ret = reiserfs_end_persistent_transaction(th);
-		reiserfs_write_unlock(inode->i_sb);
 		if (ret)
 			goto out;
 	}
 
       out:
+	if (locked)
+		reiserfs_write_unlock_once(inode->i_sb, lock_depth);
 	unlock_page(page);
 	page_cache_release(page);
 	return ret == 0 ? copied : ret;
 
       journal_error:
+	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
+	locked = false;
 	if (th) {
-		reiserfs_write_lock(inode->i_sb);
 		if (!update_sd)
 			reiserfs_update_sd(th, inode);
 		ret = reiserfs_end_persistent_transaction(th);
-		reiserfs_write_unlock(inode->i_sb);
 	}
-
 	goto out;
 }
 
@@ -2758,7 +2773,10 @@ int reiserfs_commit_write(struct file *f, struct page *page,
 	int update_sd = 0;
 	struct reiserfs_transaction_handle *th = NULL;
 
+	reiserfs_write_unlock(inode->i_sb);
 	reiserfs_wait_on_write_block(inode->i_sb);
+	reiserfs_write_lock(inode->i_sb);
+
 	if (reiserfs_transaction_running(inode->i_sb)) {
 		th = current->journal_info;
 	}
@@ -2770,7 +2788,6 @@ int reiserfs_commit_write(struct file *f, struct page *page,
 	 */
 	if (pos > inode->i_size) {
 		struct reiserfs_transaction_handle myth;
-		reiserfs_write_lock(inode->i_sb);
 		/* If the file have grown beyond the border where it
 		   can have a tail, unmark it as needing a tail
 		   packing */
@@ -2781,10 +2798,9 @@ int reiserfs_commit_write(struct file *f, struct page *page,
 			REISERFS_I(inode)->i_flags &= ~i_pack_on_close_mask;
 
 		ret = journal_begin(&myth, inode->i_sb, 1);
-		if (ret) {
-			reiserfs_write_unlock(inode->i_sb);
+		if (ret)
 			goto journal_error;
-		}
+
 		reiserfs_update_inode_transaction(inode);
 		inode->i_size = pos;
 		/*
@@ -2796,16 +2812,13 @@ int reiserfs_commit_write(struct file *f, struct page *page,
 		reiserfs_update_sd(&myth, inode);
 		update_sd = 1;
 		ret = journal_end(&myth, inode->i_sb, 1);
-		reiserfs_write_unlock(inode->i_sb);
 		if (ret)
 			goto journal_error;
 	}
 	if (th) {
-		reiserfs_write_lock(inode->i_sb);
 		if (!update_sd)
 			mark_inode_dirty(inode);
 		ret = reiserfs_end_persistent_transaction(th);
-		reiserfs_write_unlock(inode->i_sb);
 		if (ret)
 			goto out;
 	}
@@ -2815,11 +2828,9 @@ int reiserfs_commit_write(struct file *f, struct page *page,
 
       journal_error:
 	if (th) {
-		reiserfs_write_lock(inode->i_sb);
 		if (!update_sd)
 			reiserfs_update_sd(th, inode);
 		ret = reiserfs_end_persistent_transaction(th);
-		reiserfs_write_unlock(inode->i_sb);
 	}
 
 	return ret;
diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c
index 0ccc3fdda7b..ace77451ceb 100644
--- a/fs/reiserfs/ioctl.c
+++ b/fs/reiserfs/ioctl.c
@@ -13,44 +13,52 @@
 #include <linux/compat.h>
 
 /*
-** reiserfs_ioctl - handler for ioctl for inode
-** supported commands:
-**  1) REISERFS_IOC_UNPACK - try to unpack tail from direct item into indirect
-**                           and prevent packing file (argument arg has to be non-zero)
-**  2) REISERFS_IOC_[GS]ETFLAGS, REISERFS_IOC_[GS]ETVERSION
-**  3) That's all for a while ...
-*/
-int reiserfs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
-		   unsigned long arg)
+ * reiserfs_ioctl - handler for ioctl for inode
+ * supported commands:
+ *  1) REISERFS_IOC_UNPACK - try to unpack tail from direct item into indirect
+ *                           and prevent packing file (argument arg has to be non-zero)
+ *  2) REISERFS_IOC_[GS]ETFLAGS, REISERFS_IOC_[GS]ETVERSION
+ *  3) That's all for a while ...
+ */
+long reiserfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
+	struct inode *inode = filp->f_path.dentry->d_inode;
 	unsigned int flags;
 	int err = 0;
 
+	reiserfs_write_lock(inode->i_sb);
+
 	switch (cmd) {
 	case REISERFS_IOC_UNPACK:
 		if (S_ISREG(inode->i_mode)) {
 			if (arg)
-				return reiserfs_unpack(inode, filp);
-			else
-				return 0;
+				err = reiserfs_unpack(inode, filp);
 		} else
-			return -ENOTTY;
-		/* following two cases are taken from fs/ext2/ioctl.c by Remy
-		   Card (card@masi.ibp.fr) */
+			err = -ENOTTY;
+		break;
+		/*
+		 * following two cases are taken from fs/ext2/ioctl.c by Remy
+		 * Card (card@masi.ibp.fr)
+		 */
 	case REISERFS_IOC_GETFLAGS:
-		if (!reiserfs_attrs(inode->i_sb))
-			return -ENOTTY;
+		if (!reiserfs_attrs(inode->i_sb)) {
+			err = -ENOTTY;
+			break;
+		}
 
 		flags = REISERFS_I(inode)->i_attrs;
 		i_attrs_to_sd_attrs(inode, (__u16 *) & flags);
-		return put_user(flags, (int __user *)arg);
+		err = put_user(flags, (int __user *)arg);
+		break;
 	case REISERFS_IOC_SETFLAGS:{
-			if (!reiserfs_attrs(inode->i_sb))
-				return -ENOTTY;
+			if (!reiserfs_attrs(inode->i_sb)) {
+				err = -ENOTTY;
+				break;
+			}
 
 			err = mnt_want_write(filp->f_path.mnt);
 			if (err)
-				return err;
+				break;
 
 			if (!is_owner_or_cap(inode)) {
 				err = -EPERM;
@@ -90,16 +98,18 @@ int reiserfs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
 			mark_inode_dirty(inode);
 setflags_out:
 			mnt_drop_write(filp->f_path.mnt);
-			return err;
+			break;
 		}
 	case REISERFS_IOC_GETVERSION:
-		return put_user(inode->i_generation, (int __user *)arg);
+		err = put_user(inode->i_generation, (int __user *)arg);
+		break;
 	case REISERFS_IOC_SETVERSION:
 		if (!is_owner_or_cap(inode))
-			return -EPERM;
+			err = -EPERM;
+			break;
 		err = mnt_want_write(filp->f_path.mnt);
 		if (err)
-			return err;
+			break;
 		if (get_user(inode->i_generation, (int __user *)arg)) {
 			err = -EFAULT;
 			goto setversion_out;
@@ -108,19 +118,20 @@ setflags_out:
 		mark_inode_dirty(inode);
 setversion_out:
 		mnt_drop_write(filp->f_path.mnt);
-		return err;
+		break;
 	default:
-		return -ENOTTY;
+		err = -ENOTTY;
 	}
+
+	reiserfs_write_unlock(inode->i_sb);
+
+	return err;
 }
 
 #ifdef CONFIG_COMPAT
 long reiserfs_compat_ioctl(struct file *file, unsigned int cmd,
 				unsigned long arg)
 {
-	struct inode *inode = file->f_path.dentry->d_inode;
-	int ret;
-
 	/* These are just misnamed, they actually get/put from/to user an int */
 	switch (cmd) {
 	case REISERFS_IOC32_UNPACK:
@@ -141,10 +152,8 @@ long reiserfs_compat_ioctl(struct file *file, unsigned int cmd,
 	default:
 		return -ENOIOCTLCMD;
 	}
-	lock_kernel();
-	ret = reiserfs_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg));
-	unlock_kernel();
-	return ret;
+
+	return reiserfs_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
 }
 #endif
 
diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index 90622200b39..2f8a7e7b8da 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -429,21 +429,6 @@ static void clear_prepared_bits(struct buffer_head *bh)
 	clear_buffer_journal_restore_dirty(bh);
 }
 
-/* utility function to force a BUG if it is called without the big
-** kernel lock held.  caller is the string printed just before calling BUG()
-*/
-void reiserfs_check_lock_depth(struct super_block *sb, char *caller)
-{
-#ifdef CONFIG_SMP
-	if (current->lock_depth < 0) {
-		reiserfs_panic(sb, "journal-1", "%s called without kernel "
-			       "lock held", caller);
-	}
-#else
-	;
-#endif
-}
-
 /* return a cnode with same dev, block number and size in table, or null if not found */
 static inline struct reiserfs_journal_cnode *get_journal_hash_dev(struct
 								  super_block
@@ -556,7 +541,8 @@ static inline void insert_journal_hash(struct reiserfs_journal_cnode **table,
 static inline void lock_journal(struct super_block *sb)
 {
 	PROC_INFO_INC(sb, journal.lock_journal);
-	mutex_lock(&SB_JOURNAL(sb)->j_mutex);
+
+	reiserfs_mutex_lock_safe(&SB_JOURNAL(sb)->j_mutex, sb);
 }
 
 /* unlock the current transaction */
@@ -708,7 +694,9 @@ static void check_barrier_completion(struct super_block *s,
 		disable_barrier(s);
 		set_buffer_uptodate(bh);
 		set_buffer_dirty(bh);
+		reiserfs_write_unlock(s);
 		sync_dirty_buffer(bh);
+		reiserfs_write_lock(s);
 	}
 }
 
@@ -996,8 +984,13 @@ static int reiserfs_async_progress_wait(struct super_block *s)
 {
 	DEFINE_WAIT(wait);
 	struct reiserfs_journal *j = SB_JOURNAL(s);
-	if (atomic_read(&j->j_async_throttle))
+
+	if (atomic_read(&j->j_async_throttle)) {
+		reiserfs_write_unlock(s);
 		congestion_wait(BLK_RW_ASYNC, HZ / 10);
+		reiserfs_write_lock(s);
+	}
+
 	return 0;
 }
 
@@ -1043,7 +1036,8 @@ static int flush_commit_list(struct super_block *s,
 	}
 
 	/* make sure nobody is trying to flush this one at the same time */
-	mutex_lock(&jl->j_commit_mutex);
+	reiserfs_mutex_lock_safe(&jl->j_commit_mutex, s);
+
 	if (!journal_list_still_alive(s, trans_id)) {
 		mutex_unlock(&jl->j_commit_mutex);
 		goto put_jl;
@@ -1061,12 +1055,17 @@ static int flush_commit_list(struct super_block *s,
 
 	if (!list_empty(&jl->j_bh_list)) {
 		int ret;
-		unlock_kernel();
+
+		/*
+		 * We might sleep in numerous places inside
+		 * write_ordered_buffers. Relax the write lock.
+		 */
+		reiserfs_write_unlock(s);
 		ret = write_ordered_buffers(&journal->j_dirty_buffers_lock,
 					    journal, jl, &jl->j_bh_list);
 		if (ret < 0 && retval == 0)
 			retval = ret;
-		lock_kernel();
+		reiserfs_write_lock(s);
 	}
 	BUG_ON(!list_empty(&jl->j_bh_list));
 	/*
@@ -1085,8 +1084,11 @@ static int flush_commit_list(struct super_block *s,
 		    SB_ONDISK_JOURNAL_SIZE(s);
 		tbh = journal_find_get_block(s, bn);
 		if (tbh) {
-			if (buffer_dirty(tbh))
-			    ll_rw_block(WRITE, 1, &tbh) ;
+			if (buffer_dirty(tbh)) {
+		            reiserfs_write_unlock(s);
+			    ll_rw_block(WRITE, 1, &tbh);
+			    reiserfs_write_lock(s);
+			}
 			put_bh(tbh) ;
 		}
 	}
@@ -1114,12 +1116,19 @@ static int flush_commit_list(struct super_block *s,
 		bn = SB_ONDISK_JOURNAL_1st_BLOCK(s) +
 		    (jl->j_start + i) % SB_ONDISK_JOURNAL_SIZE(s);
 		tbh = journal_find_get_block(s, bn);
+
+		reiserfs_write_unlock(s);
 		wait_on_buffer(tbh);
+		reiserfs_write_lock(s);
 		// since we're using ll_rw_blk above, it might have skipped over
 		// a locked buffer.  Double check here
 		//
-		if (buffer_dirty(tbh))	/* redundant, sync_dirty_buffer() checks */
+		/* redundant, sync_dirty_buffer() checks */
+		if (buffer_dirty(tbh)) {
+			reiserfs_write_unlock(s);
 			sync_dirty_buffer(tbh);
+			reiserfs_write_lock(s);
+		}
 		if (unlikely(!buffer_uptodate(tbh))) {
 #ifdef CONFIG_REISERFS_CHECK
 			reiserfs_warning(s, "journal-601",
@@ -1143,10 +1152,15 @@ static int flush_commit_list(struct super_block *s,
 			if (buffer_dirty(jl->j_commit_bh))
 				BUG();
 			mark_buffer_dirty(jl->j_commit_bh) ;
+			reiserfs_write_unlock(s);
 			sync_dirty_buffer(jl->j_commit_bh) ;
+			reiserfs_write_lock(s);
 		}
-	} else
+	} else {
+		reiserfs_write_unlock(s);
 		wait_on_buffer(jl->j_commit_bh);
+		reiserfs_write_lock(s);
+	}
 
 	check_barrier_completion(s, jl->j_commit_bh);
 
@@ -1286,7 +1300,9 @@ static int _update_journal_header_block(struct super_block *sb,
 
 	if (trans_id >= journal->j_last_flush_trans_id) {
 		if (buffer_locked((journal->j_header_bh))) {
+			reiserfs_write_unlock(sb);
 			wait_on_buffer((journal->j_header_bh));
+			reiserfs_write_lock(sb);
 			if (unlikely(!buffer_uptodate(journal->j_header_bh))) {
 #ifdef CONFIG_REISERFS_CHECK
 				reiserfs_warning(sb, "journal-699",
@@ -1312,12 +1328,16 @@ static int _update_journal_header_block(struct super_block *sb,
 				disable_barrier(sb);
 				goto sync;
 			}
+			reiserfs_write_unlock(sb);
 			wait_on_buffer(journal->j_header_bh);
+			reiserfs_write_lock(sb);
 			check_barrier_completion(sb, journal->j_header_bh);
 		} else {
 		      sync:
 			set_buffer_dirty(journal->j_header_bh);
+			reiserfs_write_unlock(sb);
 			sync_dirty_buffer(journal->j_header_bh);
+			reiserfs_write_lock(sb);
 		}
 		if (!buffer_uptodate(journal->j_header_bh)) {
 			reiserfs_warning(sb, "journal-837",
@@ -1409,7 +1429,7 @@ static int flush_journal_list(struct super_block *s,
 
 	/* if flushall == 0, the lock is already held */
 	if (flushall) {
-		mutex_lock(&journal->j_flush_mutex);
+		reiserfs_mutex_lock_safe(&journal->j_flush_mutex, s);
 	} else if (mutex_trylock(&journal->j_flush_mutex)) {
 		BUG();
 	}
@@ -1553,7 +1573,11 @@ static int flush_journal_list(struct super_block *s,
 					reiserfs_panic(s, "journal-1011",
 						       "cn->bh is NULL");
 				}
+
+				reiserfs_write_unlock(s);
 				wait_on_buffer(cn->bh);
+				reiserfs_write_lock(s);
+
 				if (!cn->bh) {
 					reiserfs_panic(s, "journal-1012",
 						       "cn->bh is NULL");
@@ -1769,7 +1793,7 @@ static int kupdate_transactions(struct super_block *s,
 	struct reiserfs_journal *journal = SB_JOURNAL(s);
 	chunk.nr = 0;
 
-	mutex_lock(&journal->j_flush_mutex);
+	reiserfs_mutex_lock_safe(&journal->j_flush_mutex, s);
 	if (!journal_list_still_alive(s, orig_trans_id)) {
 		goto done;
 	}
@@ -1973,11 +1997,19 @@ static int do_journal_release(struct reiserfs_transaction_handle *th,
 	reiserfs_mounted_fs_count--;
 	/* wait for all commits to finish */
 	cancel_delayed_work(&SB_JOURNAL(sb)->j_work);
+
+	/*
+	 * We must release the write lock here because
+	 * the workqueue job (flush_async_commit) needs this lock
+	 */
+	reiserfs_write_unlock(sb);
 	flush_workqueue(commit_wq);
+
 	if (!reiserfs_mounted_fs_count) {
 		destroy_workqueue(commit_wq);
 		commit_wq = NULL;
 	}
+	reiserfs_write_lock(sb);
 
 	free_journal_ram(sb);
 
@@ -2243,7 +2275,11 @@ static int journal_read_transaction(struct super_block *sb,
 	/* read in the log blocks, memcpy to the corresponding real block */
 	ll_rw_block(READ, get_desc_trans_len(desc), log_blocks);
 	for (i = 0; i < get_desc_trans_len(desc); i++) {
+
+		reiserfs_write_unlock(sb);
 		wait_on_buffer(log_blocks[i]);
+		reiserfs_write_lock(sb);
+
 		if (!buffer_uptodate(log_blocks[i])) {
 			reiserfs_warning(sb, "journal-1212",
 					 "REPLAY FAILURE fsck required! "
@@ -2765,11 +2801,27 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 		goto free_and_return;
 	}
 
+	/*
+	 * We need to unlock here to avoid creating the following
+	 * dependency:
+	 * reiserfs_lock -> sysfs_mutex
+	 * Because the reiserfs mmap path creates the following dependency:
+	 * mm->mmap -> reiserfs_lock, hence we have
+	 * mm->mmap -> reiserfs_lock ->sysfs_mutex
+	 * This would ends up in a circular dependency with sysfs readdir path
+	 * which does sysfs_mutex -> mm->mmap_sem
+	 * This is fine because the reiserfs lock is useless in mount path,
+	 * at least until we call journal_begin. We keep it for paranoid
+	 * reasons.
+	 */
+	reiserfs_write_unlock(sb);
 	if (journal_init_dev(sb, journal, j_dev_name) != 0) {
+		reiserfs_write_lock(sb);
 		reiserfs_warning(sb, "sh-462",
 				 "unable to initialize jornal device");
 		goto free_and_return;
 	}
+	reiserfs_write_lock(sb);
 
 	rs = SB_DISK_SUPER_BLOCK(sb);
 
@@ -2881,8 +2933,11 @@ int journal_init(struct super_block *sb, const char *j_dev_name,
 	}
 
 	reiserfs_mounted_fs_count++;
-	if (reiserfs_mounted_fs_count <= 1)
+	if (reiserfs_mounted_fs_count <= 1) {
+		reiserfs_write_unlock(sb);
 		commit_wq = create_workqueue("reiserfs");
+		reiserfs_write_lock(sb);
+	}
 
 	INIT_DELAYED_WORK(&journal->j_work, flush_async_commits);
 	journal->j_work_sb = sb;
@@ -2964,8 +3019,11 @@ static void queue_log_writer(struct super_block *s)
 	init_waitqueue_entry(&wait, current);
 	add_wait_queue(&journal->j_join_wait, &wait);
 	set_current_state(TASK_UNINTERRUPTIBLE);
-	if (test_bit(J_WRITERS_QUEUED, &journal->j_state))
+	if (test_bit(J_WRITERS_QUEUED, &journal->j_state)) {
+		reiserfs_write_unlock(s);
 		schedule();
+		reiserfs_write_lock(s);
+	}
 	__set_current_state(TASK_RUNNING);
 	remove_wait_queue(&journal->j_join_wait, &wait);
 }
@@ -2982,7 +3040,9 @@ static void let_transaction_grow(struct super_block *sb, unsigned int trans_id)
 	struct reiserfs_journal *journal = SB_JOURNAL(sb);
 	unsigned long bcount = journal->j_bcount;
 	while (1) {
+		reiserfs_write_unlock(sb);
 		schedule_timeout_uninterruptible(1);
+		reiserfs_write_lock(sb);
 		journal->j_current_jl->j_state |= LIST_COMMIT_PENDING;
 		while ((atomic_read(&journal->j_wcount) > 0 ||
 			atomic_read(&journal->j_jlock)) &&
@@ -3033,7 +3093,9 @@ static int do_journal_begin_r(struct reiserfs_transaction_handle *th,
 
 	if (test_bit(J_WRITERS_BLOCKED, &journal->j_state)) {
 		unlock_journal(sb);
+		reiserfs_write_unlock(sb);
 		reiserfs_wait_on_write_block(sb);
+		reiserfs_write_lock(sb);
 		PROC_INFO_INC(sb, journal.journal_relock_writers);
 		goto relock;
 	}
@@ -3506,14 +3568,14 @@ static void flush_async_commits(struct work_struct *work)
 	struct reiserfs_journal_list *jl;
 	struct list_head *entry;
 
-	lock_kernel();
+	reiserfs_write_lock(sb);
 	if (!list_empty(&journal->j_journal_list)) {
 		/* last entry is the youngest, commit it and you get everything */
 		entry = journal->j_journal_list.prev;
 		jl = JOURNAL_LIST_ENTRY(entry);
 		flush_commit_list(sb, jl, 1);
 	}
-	unlock_kernel();
+	reiserfs_write_unlock(sb);
 }
 
 /*
@@ -4041,7 +4103,7 @@ static int do_journal_end(struct reiserfs_transaction_handle *th,
 	 * the new transaction is fully setup, and we've already flushed the
 	 * ordered bh list
 	 */
-	mutex_lock(&jl->j_commit_mutex);
+	reiserfs_mutex_lock_safe(&jl->j_commit_mutex, sb);
 
 	/* save the transaction id in case we need to commit it later */
 	commit_trans_id = jl->j_trans_id;
@@ -4156,7 +4218,9 @@ static int do_journal_end(struct reiserfs_transaction_handle *th,
 		next = cn->next;
 		free_cnode(sb, cn);
 		cn = next;
+		reiserfs_write_unlock(sb);
 		cond_resched();
+		reiserfs_write_lock(sb);
 	}
 
 	/* we are done  with both the c_bh and d_bh, but
@@ -4203,10 +4267,10 @@ static int do_journal_end(struct reiserfs_transaction_handle *th,
 	 * is lost.
 	 */
 	if (!list_empty(&jl->j_tail_bh_list)) {
-		unlock_kernel();
+		reiserfs_write_unlock(sb);
 		write_ordered_buffers(&journal->j_dirty_buffers_lock,
 				      journal, jl, &jl->j_tail_bh_list);
-		lock_kernel();
+		reiserfs_write_lock(sb);
 	}
 	BUG_ON(!list_empty(&jl->j_tail_bh_list));
 	mutex_unlock(&jl->j_commit_mutex);
diff --git a/fs/reiserfs/lock.c b/fs/reiserfs/lock.c
new file mode 100644
index 00000000000..ee2cfc0fd8a
--- /dev/null
+++ b/fs/reiserfs/lock.c
@@ -0,0 +1,88 @@
+#include <linux/reiserfs_fs.h>
+#include <linux/mutex.h>
+
+/*
+ * The previous reiserfs locking scheme was heavily based on
+ * the tricky properties of the Bkl:
+ *
+ * - it was acquired recursively by a same task
+ * - the performances relied on the release-while-schedule() property
+ *
+ * Now that we replace it by a mutex, we still want to keep the same
+ * recursive property to avoid big changes in the code structure.
+ * We use our own lock_owner here because the owner field on a mutex
+ * is only available in SMP or mutex debugging, also we only need this field
+ * for this mutex, no need for a system wide mutex facility.
+ *
+ * Also this lock is often released before a call that could block because
+ * reiserfs performances were partialy based on the release while schedule()
+ * property of the Bkl.
+ */
+void reiserfs_write_lock(struct super_block *s)
+{
+	struct reiserfs_sb_info *sb_i = REISERFS_SB(s);
+
+	if (sb_i->lock_owner != current) {
+		mutex_lock(&sb_i->lock);
+		sb_i->lock_owner = current;
+	}
+
+	/* No need to protect it, only the current task touches it */
+	sb_i->lock_depth++;
+}
+
+void reiserfs_write_unlock(struct super_block *s)
+{
+	struct reiserfs_sb_info *sb_i = REISERFS_SB(s);
+
+	/*
+	 * Are we unlocking without even holding the lock?
+	 * Such a situation must raise a BUG() if we don't want
+	 * to corrupt the data.
+	 */
+	BUG_ON(sb_i->lock_owner != current);
+
+	if (--sb_i->lock_depth == -1) {
+		sb_i->lock_owner = NULL;
+		mutex_unlock(&sb_i->lock);
+	}
+}
+
+/*
+ * If we already own the lock, just exit and don't increase the depth.
+ * Useful when we don't want to lock more than once.
+ *
+ * We always return the lock_depth we had before calling
+ * this function.
+ */
+int reiserfs_write_lock_once(struct super_block *s)
+{
+	struct reiserfs_sb_info *sb_i = REISERFS_SB(s);
+
+	if (sb_i->lock_owner != current) {
+		mutex_lock(&sb_i->lock);
+		sb_i->lock_owner = current;
+		return sb_i->lock_depth++;
+	}
+
+	return sb_i->lock_depth;
+}
+
+void reiserfs_write_unlock_once(struct super_block *s, int lock_depth)
+{
+	if (lock_depth == -1)
+		reiserfs_write_unlock(s);
+}
+
+/*
+ * Utility function to force a BUG if it is called without the superblock
+ * write lock held.  caller is the string printed just before calling BUG()
+ */
+void reiserfs_check_lock_depth(struct super_block *sb, char *caller)
+{
+	struct reiserfs_sb_info *sb_i = REISERFS_SB(sb);
+
+	if (sb_i->lock_depth < 0)
+		reiserfs_panic(sb, "%s called without kernel lock held %d",
+			       caller);
+}
diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c
index 27157912863..e296ff72a6c 100644
--- a/fs/reiserfs/namei.c
+++ b/fs/reiserfs/namei.c
@@ -324,6 +324,7 @@ static struct dentry *reiserfs_lookup(struct inode *dir, struct dentry *dentry,
 				      struct nameidata *nd)
 {
 	int retval;
+	int lock_depth;
 	struct inode *inode = NULL;
 	struct reiserfs_dir_entry de;
 	INITIALIZE_PATH(path_to_entry);
@@ -331,7 +332,13 @@ static struct dentry *reiserfs_lookup(struct inode *dir, struct dentry *dentry,
 	if (REISERFS_MAX_NAME(dir->i_sb->s_blocksize) < dentry->d_name.len)
 		return ERR_PTR(-ENAMETOOLONG);
 
-	reiserfs_write_lock(dir->i_sb);
+	/*
+	 * Might be called with or without the write lock, must be careful
+	 * to not recursively hold it in case we want to release the lock
+	 * before rescheduling.
+	 */
+	lock_depth = reiserfs_write_lock_once(dir->i_sb);
+
 	de.de_gen_number_bit_string = NULL;
 	retval =
 	    reiserfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len,
@@ -341,7 +348,7 @@ static struct dentry *reiserfs_lookup(struct inode *dir, struct dentry *dentry,
 		inode = reiserfs_iget(dir->i_sb,
 				      (struct cpu_key *)&(de.de_dir_id));
 		if (!inode || IS_ERR(inode)) {
-			reiserfs_write_unlock(dir->i_sb);
+			reiserfs_write_unlock_once(dir->i_sb, lock_depth);
 			return ERR_PTR(-EACCES);
 		}
 
@@ -350,7 +357,7 @@ static struct dentry *reiserfs_lookup(struct inode *dir, struct dentry *dentry,
 		if (IS_PRIVATE(dir))
 			inode->i_flags |= S_PRIVATE;
 	}
-	reiserfs_write_unlock(dir->i_sb);
+	reiserfs_write_unlock_once(dir->i_sb, lock_depth);
 	if (retval == IO_ERROR) {
 		return ERR_PTR(-EIO);
 	}
@@ -725,6 +732,7 @@ static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	struct inode *inode;
 	struct reiserfs_transaction_handle th;
 	struct reiserfs_security_handle security;
+	int lock_depth;
 	/* We need blocks for transaction + (user+group)*(quotas for new inode + update of quota for directory owner) */
 	int jbegin_count =
 	    JOURNAL_PER_BALANCE_CNT * 3 +
@@ -748,7 +756,7 @@ static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 		return retval;
 	}
 	jbegin_count += retval;
-	reiserfs_write_lock(dir->i_sb);
+	lock_depth = reiserfs_write_lock_once(dir->i_sb);
 
 	retval = journal_begin(&th, dir->i_sb, jbegin_count);
 	if (retval) {
@@ -798,8 +806,8 @@ static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	d_instantiate(dentry, inode);
 	unlock_new_inode(inode);
 	retval = journal_end(&th, dir->i_sb, jbegin_count);
-      out_failed:
-	reiserfs_write_unlock(dir->i_sb);
+out_failed:
+	reiserfs_write_unlock_once(dir->i_sb, lock_depth);
 	return retval;
 }
 
diff --git a/fs/reiserfs/prints.c b/fs/reiserfs/prints.c
index 536eacaeb71..adbc6f53851 100644
--- a/fs/reiserfs/prints.c
+++ b/fs/reiserfs/prints.c
@@ -349,10 +349,6 @@ void reiserfs_debug(struct super_block *s, int level, const char *fmt, ...)
 
    .  */
 
-#ifdef CONFIG_REISERFS_CHECK
-extern struct tree_balance *cur_tb;
-#endif
-
 void __reiserfs_panic(struct super_block *sb, const char *id,
 		      const char *function, const char *fmt, ...)
 {
diff --git a/fs/reiserfs/resize.c b/fs/reiserfs/resize.c
index 18b315d3d10..b3a94d20f0f 100644
--- a/fs/reiserfs/resize.c
+++ b/fs/reiserfs/resize.c
@@ -141,7 +141,9 @@ int reiserfs_resize(struct super_block *s, unsigned long block_count_new)
 
 			set_buffer_uptodate(bh);
 			mark_buffer_dirty(bh);
+			reiserfs_write_unlock(s);
 			sync_dirty_buffer(bh);
+			reiserfs_write_lock(s);
 			// update bitmap_info stuff
 			bitmap[i].free_count = sb_blocksize(sb) * 8 - 1;
 			brelse(bh);
diff --git a/fs/reiserfs/stree.c b/fs/reiserfs/stree.c
index d036ee5b1c8..5fa7118f04e 100644
--- a/fs/reiserfs/stree.c
+++ b/fs/reiserfs/stree.c
@@ -222,9 +222,6 @@ static inline int bin_search(const void *key,	/* Key to search for. */
 	return ITEM_NOT_FOUND;
 }
 
-#ifdef CONFIG_REISERFS_CHECK
-extern struct tree_balance *cur_tb;
-#endif
 
 /* Minimal possible key. It is never in the tree. */
 const struct reiserfs_key MIN_KEY = { 0, 0, {{0, 0},} };
@@ -519,25 +516,48 @@ static int is_tree_node(struct buffer_head *bh, int level)
 
 #define SEARCH_BY_KEY_READA 16
 
-/* The function is NOT SCHEDULE-SAFE! */
-static void search_by_key_reada(struct super_block *s,
+/*
+ * The function is NOT SCHEDULE-SAFE!
+ * It might unlock the write lock if we needed to wait for a block
+ * to be read. Note that in this case it won't recover the lock to avoid
+ * high contention resulting from too much lock requests, especially
+ * the caller (search_by_key) will perform other schedule-unsafe
+ * operations just after calling this function.
+ *
+ * @return true if we have unlocked
+ */
+static bool search_by_key_reada(struct super_block *s,
 				struct buffer_head **bh,
 				b_blocknr_t *b, int num)
 {
 	int i, j;
+	bool unlocked = false;
 
 	for (i = 0; i < num; i++) {
 		bh[i] = sb_getblk(s, b[i]);
 	}
+	/*
+	 * We are going to read some blocks on which we
+	 * have a reference. It's safe, though we might be
+	 * reading blocks concurrently changed if we release
+	 * the lock. But it's still fine because we check later
+	 * if the tree changed
+	 */
 	for (j = 0; j < i; j++) {
 		/*
 		 * note, this needs attention if we are getting rid of the BKL
 		 * you have to make sure the prepared bit isn't set on this buffer
 		 */
-		if (!buffer_uptodate(bh[j]))
+		if (!buffer_uptodate(bh[j])) {
+			if (!unlocked) {
+				reiserfs_write_unlock(s);
+				unlocked = true;
+			}
 			ll_rw_block(READA, 1, bh + j);
+		}
 		brelse(bh[j]);
 	}
+	return unlocked;
 }
 
 /**************************************************************************
@@ -625,11 +645,26 @@ int search_by_key(struct super_block *sb, const struct cpu_key *key,	/* Key to s
 		   have a pointer to it. */
 		if ((bh = last_element->pe_buffer =
 		     sb_getblk(sb, block_number))) {
+			bool unlocked = false;
+
 			if (!buffer_uptodate(bh) && reada_count > 1)
-				search_by_key_reada(sb, reada_bh,
+				/* may unlock the write lock */
+				unlocked = search_by_key_reada(sb, reada_bh,
 						    reada_blocks, reada_count);
+			/*
+			 * If we haven't already unlocked the write lock,
+			 * then we need to do that here before reading
+			 * the current block
+			 */
+			if (!buffer_uptodate(bh) && !unlocked) {
+				reiserfs_write_unlock(sb);
+				unlocked = true;
+			}
 			ll_rw_block(READ, 1, &bh);
 			wait_on_buffer(bh);
+
+			if (unlocked)
+				reiserfs_write_lock(sb);
 			if (!buffer_uptodate(bh))
 				goto io_error;
 		} else {
@@ -673,7 +708,7 @@ int search_by_key(struct super_block *sb, const struct cpu_key *key,	/* Key to s
 		       !key_in_buffer(search_path, key, sb),
 		       "PAP-5130: key is not in the buffer");
 #ifdef CONFIG_REISERFS_CHECK
-		if (cur_tb) {
+		if (REISERFS_SB(sb)->cur_tb) {
 			print_cur_tb("5140");
 			reiserfs_panic(sb, "PAP-5140",
 				       "schedule occurred in do_balance!");
@@ -1024,7 +1059,9 @@ static char prepare_for_delete_or_cut(struct reiserfs_transaction_handle *th, st
 			reiserfs_free_block(th, inode, block, 1);
 		    }
 
+		    reiserfs_write_unlock(sb);
 		    cond_resched();
+		    reiserfs_write_lock(sb);
 
 		    if (item_moved (&s_ih, path))  {
 			need_re_search = 1;
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index f0ad05f3802..339b0baf2af 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -465,7 +465,7 @@ static void reiserfs_put_super(struct super_block *s)
 	struct reiserfs_transaction_handle th;
 	th.t_trans_id = 0;
 
-	lock_kernel();
+	reiserfs_write_lock(s);
 
 	if (s->s_dirt)
 		reiserfs_write_super(s);
@@ -499,10 +499,10 @@ static void reiserfs_put_super(struct super_block *s)
 
 	reiserfs_proc_info_done(s);
 
+	reiserfs_write_unlock(s);
+	mutex_destroy(&REISERFS_SB(s)->lock);
 	kfree(s->s_fs_info);
 	s->s_fs_info = NULL;
-
-	unlock_kernel();
 }
 
 static struct kmem_cache *reiserfs_inode_cachep;
@@ -554,25 +554,28 @@ static void reiserfs_dirty_inode(struct inode *inode)
 	struct reiserfs_transaction_handle th;
 
 	int err = 0;
+	int lock_depth;
+
 	if (inode->i_sb->s_flags & MS_RDONLY) {
 		reiserfs_warning(inode->i_sb, "clm-6006",
 				 "writing inode %lu on readonly FS",
 				 inode->i_ino);
 		return;
 	}
-	reiserfs_write_lock(inode->i_sb);
+	lock_depth = reiserfs_write_lock_once(inode->i_sb);
 
 	/* this is really only used for atime updates, so they don't have
 	 ** to be included in O_SYNC or fsync
 	 */
 	err = journal_begin(&th, inode->i_sb, 1);
-	if (err) {
-		reiserfs_write_unlock(inode->i_sb);
-		return;
-	}
+	if (err)
+		goto out;
+
 	reiserfs_update_sd(&th, inode);
 	journal_end(&th, inode->i_sb, 1);
-	reiserfs_write_unlock(inode->i_sb);
+
+out:
+	reiserfs_write_unlock_once(inode->i_sb, lock_depth);
 }
 
 #ifdef CONFIG_QUOTA
@@ -1168,11 +1171,14 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg)
 	unsigned int qfmt = 0;
 #ifdef CONFIG_QUOTA
 	int i;
+#endif
+
+	reiserfs_write_lock(s);
 
+#ifdef CONFIG_QUOTA
 	memcpy(qf_names, REISERFS_SB(s)->s_qf_names, sizeof(qf_names));
 #endif
 
-	lock_kernel();
 	rs = SB_DISK_SUPER_BLOCK(s);
 
 	if (!reiserfs_parse_options
@@ -1295,12 +1301,12 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg)
 
 out_ok:
 	replace_mount_options(s, new_opts);
-	unlock_kernel();
+	reiserfs_write_unlock(s);
 	return 0;
 
 out_err:
 	kfree(new_opts);
-	unlock_kernel();
+	reiserfs_write_unlock(s);
 	return err;
 }
 
@@ -1404,7 +1410,9 @@ static int read_super_block(struct super_block *s, int offset)
 static int reread_meta_blocks(struct super_block *s)
 {
 	ll_rw_block(READ, 1, &(SB_BUFFER_WITH_SB(s)));
+	reiserfs_write_unlock(s);
 	wait_on_buffer(SB_BUFFER_WITH_SB(s));
+	reiserfs_write_lock(s);
 	if (!buffer_uptodate(SB_BUFFER_WITH_SB(s))) {
 		reiserfs_warning(s, "reiserfs-2504", "error reading the super");
 		return 1;
@@ -1613,7 +1621,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	sbi = kzalloc(sizeof(struct reiserfs_sb_info), GFP_KERNEL);
 	if (!sbi) {
 		errval = -ENOMEM;
-		goto error;
+		goto error_alloc;
 	}
 	s->s_fs_info = sbi;
 	/* Set default values for options: non-aggressive tails, RO on errors */
@@ -1627,6 +1635,20 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	/* setup default block allocator options */
 	reiserfs_init_alloc_options(s);
 
+	mutex_init(&REISERFS_SB(s)->lock);
+	REISERFS_SB(s)->lock_depth = -1;
+
+	/*
+	 * This function is called with the bkl, which also was the old
+	 * locking used here.
+	 * do_journal_begin() will soon check if we hold the lock (ie: was the
+	 * bkl). This is likely because do_journal_begin() has several another
+	 * callers because at this time, it doesn't seem to be necessary to
+	 * protect against anything.
+	 * Anyway, let's be conservative and lock for now.
+	 */
+	reiserfs_write_lock(s);
+
 	jdev_name = NULL;
 	if (reiserfs_parse_options
 	    (s, (char *)data, &(sbi->s_mount_opt), &blocks, &jdev_name,
@@ -1852,9 +1874,13 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	init_waitqueue_head(&(sbi->s_wait));
 	spin_lock_init(&sbi->bitmap_lock);
 
+	reiserfs_write_unlock(s);
+
 	return (0);
 
 error:
+	reiserfs_write_unlock(s);
+error_alloc:
 	if (jinit_done) {	/* kill the commit thread, free journal ram */
 		journal_release_error(NULL, s);
 	}
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index 6925b835a43..58aa8e75f7f 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -975,7 +975,7 @@ int reiserfs_lookup_privroot(struct super_block *s)
 	int err = 0;
 
 	/* If we don't have the privroot located yet - go find it */
-	mutex_lock(&s->s_root->d_inode->i_mutex);
+	reiserfs_mutex_lock_safe(&s->s_root->d_inode->i_mutex, s);
 	dentry = lookup_one_len(PRIVROOT_NAME, s->s_root,
 				strlen(PRIVROOT_NAME));
 	if (!IS_ERR(dentry)) {
@@ -1004,14 +1004,14 @@ int reiserfs_xattr_init(struct super_block *s, int mount_flags)
 		goto error;
 
 	if (!privroot->d_inode && !(mount_flags & MS_RDONLY)) {
-		mutex_lock(&s->s_root->d_inode->i_mutex);
+		reiserfs_mutex_lock_safe(&s->s_root->d_inode->i_mutex, s);
 		err = create_privroot(REISERFS_SB(s)->priv_root);
 		mutex_unlock(&s->s_root->d_inode->i_mutex);
 	}
 
 	if (privroot->d_inode) {
 		s->s_xattr = reiserfs_xattr_handlers;
-		mutex_lock(&privroot->d_inode->i_mutex);
+		reiserfs_mutex_lock_safe(&privroot->d_inode->i_mutex, s);
 		if (!REISERFS_SB(s)->xattr_root) {
 			struct dentry *dentry;
 			dentry = lookup_one_len(XAROOT_NAME, privroot,
diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c
index dbc093afd94..8a771c59ac3 100644
--- a/fs/ubifs/debug.c
+++ b/fs/ubifs/debug.c
@@ -2014,7 +2014,7 @@ 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 1009adc8d60..39849f887e7 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -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/recovery.c b/fs/ubifs/recovery.c
index f94ddf7efba..868a55ee080 100644
--- a/fs/ubifs/recovery.c
+++ b/fs/ubifs/recovery.c
@@ -23,7 +23,7 @@
 /*
  * This file implements functions needed to recover from unclean un-mounts.
  * When UBIFS is mounted, it checks a flag on the master node to determine if
- * an un-mount was completed sucessfully. If not, the process of mounting
+ * an un-mount was completed successfully. If not, the process of mounting
  * incorparates additional checking and fixing of on-flash data structures.
  * UBIFS always cleans away all remnants of an unclean un-mount, so that
  * errors do not accumulate. However UBIFS defers recovery if it is mounted
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 333e181ee98..943ad562453 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -1842,22 +1842,32 @@ 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/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h
index 6533ead9b88..a2c16bcee90 100644
--- a/fs/xfs/quota/xfs_dquot.h
+++ b/fs/xfs/quota/xfs_dquot.h
@@ -98,7 +98,7 @@ typedef struct xfs_dquot {
 #define dq_flags	q_lists.dqm_flags
 
 /*
- * Lock hierachy for q_qlock:
+ * Lock hierarchy for q_qlock:
  *	XFS_QLOCK_NORMAL is the implicit default,
  * 	XFS_QLOCK_NESTED is the dquot with the higher id in xfs_dqlock2
  */