ocfs2: Add extended attribute support

This patch implements storing extended attributes both in inode or a single
external block. We only store EA's in-inode when blocksize > 512 or that
inode block has free space for it. When an EA's value is larger than 80
bytes, we will store the value via b-tree outside inode or block.

Signed-off-by: Tiger Yang <tiger.yang@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>

16 files changed, 1927 insertions, 6 deletions

diff --git a/fs/ocfs2/Makefile b/fs/ocfs2/Makefile
index af63980319c..21323da4085 100644
--- a/fs/ocfs2/Makefile
+++ b/fs/ocfs2/Makefile
@@ -36,6 +36,8 @@ ocfs2-objs := \
 	uptodate.o		\
 	ver.o			\
 	xattr.o			\
+	xattr_user.o		\
+	xattr_trusted.o
 
 ocfs2_stackglue-objs := stackglue.o
 ocfs2_stack_o2cb-objs := stack_o2cb.o
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 89d8541f85b..f4273c2c209 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -55,6 +55,7 @@
 #include "mmap.h"
 #include "suballoc.h"
 #include "super.h"
+#include "xattr.h"
 
 #include "buffer_head_io.h"
 
@@ -2070,6 +2071,10 @@ const struct inode_operations ocfs2_file_iops = {
 	.setattr	= ocfs2_setattr,
 	.getattr	= ocfs2_getattr,
 	.permission	= ocfs2_permission,
+	.setxattr	= generic_setxattr,
+	.getxattr	= generic_getxattr,
+	.listxattr	= ocfs2_listxattr,
+	.removexattr	= generic_removexattr,
 	.fallocate	= ocfs2_fallocate,
 	.fiemap		= ocfs2_fiemap,
 };
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 99f012a0f20..4738dd25bb9 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -49,6 +49,7 @@
 #include "symlink.h"
 #include "sysfile.h"
 #include "uptodate.h"
+#include "xattr.h"
 
 #include "buffer_head_io.h"
 
@@ -741,6 +742,13 @@ static int ocfs2_wipe_inode(struct inode *inode,
 		goto bail_unlock_dir;
 	}
 
+	/*Free extended attribute resources associated with this inode.*/
+	status = ocfs2_xattr_remove(inode, di_bh);
+	if (status < 0) {
+		mlog_errno(status);
+		goto bail_unlock_dir;
+	}
+
 	status = ocfs2_remove_inode(inode, di_bh, orphan_dir_inode,
 				    orphan_dir_bh);
 	if (status < 0)
diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h
index 390a85596aa..499bc62e758 100644
--- a/fs/ocfs2/inode.h
+++ b/fs/ocfs2/inode.h
@@ -40,6 +40,9 @@ struct ocfs2_inode_info
 	/* protects allocation changes on this inode. */
 	struct rw_semaphore		ip_alloc_sem;
 
+	/* protects extended attribute changes on this inode */
+	struct rw_semaphore		ip_xattr_sem;
+
 	/* These fields are protected by ip_lock */
 	spinlock_t			ip_lock;
 	u32				ip_open_count;
diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h
index 9485f8037d9..08d1add1487 100644
--- a/fs/ocfs2/journal.h
+++ b/fs/ocfs2/journal.h
@@ -283,6 +283,9 @@ int                  ocfs2_journal_dirty_data(handle_t *handle,
 /* simple file updates like chmod, etc. */
 #define OCFS2_INODE_UPDATE_CREDITS 1
 
+/* extended attribute block update */
+#define OCFS2_XATTR_BLOCK_UPDATE_CREDITS 1
+
 /* group extend. inode update and last group update. */
 #define OCFS2_GROUP_EXTEND_CREDITS	(OCFS2_INODE_UPDATE_CREDITS + 1)
 
@@ -340,6 +343,13 @@ int                  ocfs2_journal_dirty_data(handle_t *handle,
 #define OCFS2_RENAME_CREDITS (3 * OCFS2_INODE_UPDATE_CREDITS + 3              \
 			     + OCFS2_UNLINK_CREDITS)
 
+/* global bitmap dinode, group desc., relinked group,
+ * suballocator dinode, group desc., relinked group,
+ * dinode, xattr block */
+#define OCFS2_XATTR_BLOCK_CREATE_CREDITS (OCFS2_SUBALLOC_ALLOC * 2 + \
+					  + OCFS2_INODE_UPDATE_CREDITS \
+					  + OCFS2_XATTR_BLOCK_UPDATE_CREDITS)
+
 /*
  * Please note that the caller must make sure that root_el is the root
  * of extent tree. So for an inode, it should be &fe->id2.i_list. Otherwise
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 2cd6f501755..76d1d131430 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -60,6 +60,7 @@
 #include "symlink.h"
 #include "sysfile.h"
 #include "uptodate.h"
+#include "xattr.h"
 
 #include "buffer_head_io.h"
 
@@ -1918,4 +1919,8 @@ const struct inode_operations ocfs2_dir_iops = {
 	.setattr	= ocfs2_setattr,
 	.getattr	= ocfs2_getattr,
 	.permission	= ocfs2_permission,
+	.setxattr	= generic_setxattr,
+	.getxattr	= generic_getxattr,
+	.listxattr	= ocfs2_listxattr,
+	.removexattr	= generic_removexattr,
 };
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index ce75ca312a2..cae0dd4b7f7 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -188,6 +188,7 @@ enum ocfs2_mount_options
 	OCFS2_MOUNT_ERRORS_PANIC = 1 << 3, /* Panic on errors */
 	OCFS2_MOUNT_DATA_WRITEBACK = 1 << 4, /* No data ordering */
 	OCFS2_MOUNT_LOCALFLOCKS = 1 << 5, /* No cluster aware user file locks */
+	OCFS2_MOUNT_NOUSERXATTR = 1 << 6, /* No user xattr */
 };
 
 #define OCFS2_OSB_SOFT_RO	0x0001
@@ -218,6 +219,7 @@ struct ocfs2_super
 	u32 bitmap_cpg;
 	u8 *uuid;
 	char *uuid_str;
+	u32 uuid_hash;
 	u8 *vol_label;
 	u64 first_cluster_group_blkno;
 	u32 fs_generation;
diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h
index 1055ba0af9b..98e1f8bba0e 100644
--- a/fs/ocfs2/ocfs2_fs.h
+++ b/fs/ocfs2/ocfs2_fs.h
@@ -570,7 +570,7 @@ struct ocfs2_super_block {
 /*40*/	__le16 s_max_slots;		/* Max number of simultaneous mounts
 					   before tunefs required */
 	__le16 s_tunefs_flag;
-	__le32 s_reserved1;
+	__le32 s_uuid_hash;		/* hash value of uuid */
 	__le64 s_first_cluster_group;	/* Block offset of 1st cluster
 					 * group header */
 /*50*/	__u8  s_label[OCFS2_MAX_VOL_LABEL_LEN];	/* Label for mounting, etc. */
@@ -787,7 +787,11 @@ struct ocfs2_xattr_tree_root {
 /*10*/	struct ocfs2_extent_list xt_list; /* Extent record list */
 };
 
-#define OCFS2_XATTR_INDEXED 0x1
+#define OCFS2_XATTR_INDEXED	0x1
+#define OCFS2_HASH_SHIFT	5
+#define OCFS2_XATTR_ROUND	3
+#define OCFS2_XATTR_SIZE(size)	(((size) + OCFS2_XATTR_ROUND) & \
+				~(OCFS2_XATTR_ROUND))
 
 /*
  * On disk structure for xattr block.
diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c
index bb774d70d26..f1871ca8381 100644
--- a/fs/ocfs2/suballoc.c
+++ b/fs/ocfs2/suballoc.c
@@ -493,9 +493,9 @@ bail:
 	return status;
 }
 
-int ocfs2_reserve_new_metadata(struct ocfs2_super *osb,
-			       struct ocfs2_extent_list *root_el,
-			       struct ocfs2_alloc_context **ac)
+int ocfs2_reserve_new_metadata_blocks(struct ocfs2_super *osb,
+				      int blocks,
+				      struct ocfs2_alloc_context **ac)
 {
 	int status;
 	u32 slot;
@@ -507,7 +507,7 @@ int ocfs2_reserve_new_metadata(struct ocfs2_super *osb,
 		goto bail;
 	}
 
-	(*ac)->ac_bits_wanted = ocfs2_extend_meta_needed(root_el);
+	(*ac)->ac_bits_wanted = blocks;
 	(*ac)->ac_which = OCFS2_AC_USE_META;
 	slot = osb->slot_num;
 	(*ac)->ac_group_search = ocfs2_block_group_search;
@@ -532,6 +532,15 @@ bail:
 	return status;
 }
 
+int ocfs2_reserve_new_metadata(struct ocfs2_super *osb,
+			       struct ocfs2_extent_list *root_el,
+			       struct ocfs2_alloc_context **ac)
+{
+	return ocfs2_reserve_new_metadata_blocks(osb,
+					ocfs2_extend_meta_needed(root_el),
+					ac);
+}
+
 static int ocfs2_steal_inode_from_other_nodes(struct ocfs2_super *osb,
 					      struct ocfs2_alloc_context *ac)
 {
diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h
index 9e026c8afee..028fd633b44 100644
--- a/fs/ocfs2/suballoc.h
+++ b/fs/ocfs2/suballoc.h
@@ -67,6 +67,9 @@ static inline int ocfs2_alloc_context_bits_left(struct ocfs2_alloc_context *ac)
 int ocfs2_reserve_new_metadata(struct ocfs2_super *osb,
 			       struct ocfs2_extent_list *root_el,
 			       struct ocfs2_alloc_context **ac);
+int ocfs2_reserve_new_metadata_blocks(struct ocfs2_super *osb,
+				      int blocks,
+				      struct ocfs2_alloc_context **ac);
 int ocfs2_reserve_new_inode(struct ocfs2_super *osb,
 			    struct ocfs2_alloc_context **ac);
 int ocfs2_reserve_clusters(struct ocfs2_super *osb,
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index 9bdb3aeefe8..3b04f5d2e89 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -64,6 +64,7 @@
 #include "sysfile.h"
 #include "uptodate.h"
 #include "ver.h"
+#include "xattr.h"
 
 #include "buffer_head_io.h"
 
@@ -154,6 +155,8 @@ enum {
 	Opt_localalloc,
 	Opt_localflocks,
 	Opt_stack,
+	Opt_user_xattr,
+	Opt_nouser_xattr,
 	Opt_err,
 };
 
@@ -173,6 +176,8 @@ static const match_table_t tokens = {
 	{Opt_localalloc, "localalloc=%d"},
 	{Opt_localflocks, "localflocks"},
 	{Opt_stack, "cluster_stack=%s"},
+	{Opt_user_xattr, "user_xattr"},
+	{Opt_nouser_xattr, "nouser_xattr"},
 	{Opt_err, NULL}
 };
 
@@ -848,6 +853,12 @@ static int ocfs2_parse_options(struct super_block *sb,
 		case Opt_data_writeback:
 			mopt->mount_opt |= OCFS2_MOUNT_DATA_WRITEBACK;
 			break;
+		case Opt_user_xattr:
+			mopt->mount_opt &= ~OCFS2_MOUNT_NOUSERXATTR;
+			break;
+		case Opt_nouser_xattr:
+			mopt->mount_opt |= OCFS2_MOUNT_NOUSERXATTR;
+			break;
 		case Opt_atime_quantum:
 			if (match_int(&args[0], &option)) {
 				status = 0;
@@ -1135,6 +1146,7 @@ static void ocfs2_inode_init_once(void *data)
 	oi->ip_dir_start_lookup = 0;
 
 	init_rwsem(&oi->ip_alloc_sem);
+	init_rwsem(&oi->ip_xattr_sem);
 	mutex_init(&oi->ip_io_mutex);
 
 	oi->ip_blkno = 0ULL;
@@ -1378,6 +1390,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
 	sb->s_fs_info = osb;
 	sb->s_op = &ocfs2_sops;
 	sb->s_export_op = &ocfs2_export_ops;
+	sb->s_xattr = ocfs2_xattr_handlers;
 	sb->s_time_gran = 1;
 	sb->s_flags |= MS_NOATIME;
 	/* this is needed to support O_LARGEFILE */
@@ -1574,6 +1587,7 @@ static int ocfs2_initialize_super(struct super_block *sb,
 	osb->first_cluster_group_blkno =
 		le64_to_cpu(di->id2.i_super.s_first_cluster_group);
 	osb->fs_generation = le32_to_cpu(di->i_fs_generation);
+	osb->uuid_hash = le32_to_cpu(di->id2.i_super.s_uuid_hash);
 	mlog(0, "vol_label: %s\n", osb->vol_label);
 	mlog(0, "uuid: %s\n", osb->uuid_str);
 	mlog(0, "root_blkno=%llu, system_dir_blkno=%llu\n",
diff --git a/fs/ocfs2/symlink.c b/fs/ocfs2/symlink.c
index ba9dbb51d25..8c5879c7f84 100644
--- a/fs/ocfs2/symlink.c
+++ b/fs/ocfs2/symlink.c
@@ -50,6 +50,7 @@
 #include "inode.h"
 #include "journal.h"
 #include "symlink.h"
+#include "xattr.h"
 
 #include "buffer_head_io.h"
 
@@ -168,10 +169,18 @@ const struct inode_operations ocfs2_symlink_inode_operations = {
 	.follow_link	= ocfs2_follow_link,
 	.getattr	= ocfs2_getattr,
 	.setattr	= ocfs2_setattr,
+	.setxattr	= generic_setxattr,
+	.getxattr	= generic_getxattr,
+	.listxattr	= ocfs2_listxattr,
+	.removexattr	= generic_removexattr,
 };
 const struct inode_operations ocfs2_fast_symlink_inode_operations = {
 	.readlink	= ocfs2_readlink,
 	.follow_link	= ocfs2_follow_link,
 	.getattr	= ocfs2_getattr,
 	.setattr	= ocfs2_setattr,
+	.setxattr	= generic_setxattr,
+	.getxattr	= generic_getxattr,
+	.listxattr	= ocfs2_listxattr,
+	.removexattr	= generic_removexattr,
 };
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index 9604a4cd02b..67bebd9259e 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -5,6 +5,9 @@
  *
  * Copyright (C) 2008 Oracle.  All rights reserved.
  *
+ * CREDITS:
+ * Lots of code in this file is taken from ext3.
+ *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public
  * License as published by the Free Software Foundation; either
@@ -21,6 +24,19 @@
  * Boston, MA 021110-1307, USA.
  */
 
+#include <linux/capability.h>
+#include <linux/fs.h>
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/highmem.h>
+#include <linux/pagemap.h>
+#include <linux/uio.h>
+#include <linux/sched.h>
+#include <linux/splice.h>
+#include <linux/mount.h>
+#include <linux/writeback.h>
+#include <linux/falloc.h>
+
 #define MLOG_MASK_PREFIX ML_XATTR
 #include <cluster/masklog.h>
 
@@ -28,12 +44,119 @@
 #include "alloc.h"
 #include "dlmglue.h"
 #include "file.h"
+#include "symlink.h"
+#include "sysfile.h"
 #include "inode.h"
 #include "journal.h"
 #include "ocfs2_fs.h"
 #include "suballoc.h"
 #include "uptodate.h"
 #include "buffer_head_io.h"
+#include "xattr.h"
+
+
+struct ocfs2_xattr_def_value_root {
+	struct ocfs2_xattr_value_root	xv;
+	struct ocfs2_extent_rec		er;
+};
+
+#define OCFS2_XATTR_ROOT_SIZE	(sizeof(struct ocfs2_xattr_def_value_root))
+#define OCFS2_XATTR_INLINE_SIZE	80
+
+static struct ocfs2_xattr_def_value_root def_xv = {
+	.xv.xr_list.l_count = cpu_to_le16(1),
+};
+
+struct xattr_handler *ocfs2_xattr_handlers[] = {
+	&ocfs2_xattr_user_handler,
+	&ocfs2_xattr_trusted_handler,
+	NULL
+};
+
+static struct xattr_handler *ocfs2_xattr_handler_map[] = {
+	[OCFS2_XATTR_INDEX_USER]	= &ocfs2_xattr_user_handler,
+	[OCFS2_XATTR_INDEX_TRUSTED]	= &ocfs2_xattr_trusted_handler,
+};
+
+struct ocfs2_xattr_info {
+	int name_index;
+	const char *name;
+	const void *value;
+	size_t value_len;
+};
+
+struct ocfs2_xattr_search {
+	struct buffer_head *inode_bh;
+	/*
+	 * xattr_bh point to the block buffer head which has extended attribute
+	 * when extended attribute in inode, xattr_bh is equal to inode_bh.
+	 */
+	struct buffer_head *xattr_bh;
+	struct ocfs2_xattr_header *header;
+	void *base;
+	void *end;
+	struct ocfs2_xattr_entry *here;
+	int not_found;
+};
+
+static inline struct xattr_handler *ocfs2_xattr_handler(int name_index)
+{
+	struct xattr_handler *handler = NULL;
+
+	if (name_index > 0 && name_index < OCFS2_XATTR_MAX)
+		handler = ocfs2_xattr_handler_map[name_index];
+
+	return handler;
+}
+
+static inline u32 ocfs2_xattr_name_hash(struct inode *inode,
+					char *prefix,
+					int prefix_len,
+					char *name,
+					int name_len)
+{
+	/* Get hash value of uuid from super block */
+	u32 hash = OCFS2_SB(inode->i_sb)->uuid_hash;
+	int i;
+
+	/* hash extended attribute prefix */
+	for (i = 0; i < prefix_len; i++) {
+		hash = (hash << OCFS2_HASH_SHIFT) ^
+		       (hash >> (8*sizeof(hash) - OCFS2_HASH_SHIFT)) ^
+		       *prefix++;
+	}
+	/* hash extended attribute name */
+	for (i = 0; i < name_len; i++) {
+		hash = (hash << OCFS2_HASH_SHIFT) ^
+		       (hash >> (8*sizeof(hash) - OCFS2_HASH_SHIFT)) ^
+		       *name++;
+	}
+
+	return hash;
+}
+
+/*
+ * ocfs2_xattr_hash_entry()
+ *
+ * Compute the hash of an extended attribute.
+ */
+static void ocfs2_xattr_hash_entry(struct inode *inode,
+				   struct ocfs2_xattr_header *header,
+				   struct ocfs2_xattr_entry *entry)
+{
+	u32 hash = 0;
+	struct xattr_handler *handler =
+			ocfs2_xattr_handler(ocfs2_xattr_get_type(entry));
+	char *prefix = handler->prefix;
+	char *name = (char *)header + le16_to_cpu(entry->xe_name_offset);
+	int prefix_len = strlen(handler->prefix);
+
+	hash = ocfs2_xattr_name_hash(inode, prefix, prefix_len, name,
+				     entry->xe_name_len);
+	entry->xe_name_hash = cpu_to_le32(hash);
+
+	return;
+}
 
 static int ocfs2_xattr_extend_allocation(struct inode *inode,
 					 u32 clusters_to_add,
@@ -303,3 +426,1500 @@ static int ocfs2_xattr_value_truncate(struct inode *inode,
 
 	return ret;
 }
+
+static int ocfs2_xattr_list_entries(struct inode *inode,
+				    struct ocfs2_xattr_header *header,
+				    char *buffer, size_t buffer_size)
+{
+	size_t rest = buffer_size;
+	int i;
+
+	for (i = 0 ; i < le16_to_cpu(header->xh_count); i++) {
+		struct ocfs2_xattr_entry *entry = &header->xh_entries[i];
+		struct xattr_handler *handler =
+			ocfs2_xattr_handler(ocfs2_xattr_get_type(entry));
+
+		if (handler) {
+			size_t size = handler->list(inode, buffer, rest,
+					((char *)header +
+					le16_to_cpu(entry->xe_name_offset)),
+					entry->xe_name_len);
+			if (buffer) {
+				if (size > rest)
+					return -ERANGE;
+				buffer += size;
+			}
+			rest -= size;
+		}
+	}
+
+	return buffer_size - rest;
+}
+
+static int ocfs2_xattr_ibody_list(struct inode *inode,
+				  struct ocfs2_dinode *di,
+				  char *buffer,
+				  size_t buffer_size)
+{
+	struct ocfs2_xattr_header *header = NULL;
+	struct ocfs2_inode_info *oi = OCFS2_I(inode);
+	int ret = 0;
+
+	if (!(oi->ip_dyn_features & OCFS2_INLINE_XATTR_FL))
+		return ret;
+
+	header = (struct ocfs2_xattr_header *)
+		 ((void *)di + inode->i_sb->s_blocksize -
+		 le16_to_cpu(di->i_xattr_inline_size));
+
+	ret = ocfs2_xattr_list_entries(inode, header, buffer, buffer_size);
+
+	return ret;
+}
+
+static int ocfs2_xattr_block_list(struct inode *inode,
+				  struct ocfs2_dinode *di,
+				  char *buffer,
+				  size_t buffer_size)
+{
+	struct buffer_head *blk_bh = NULL;
+	struct ocfs2_xattr_header *header = NULL;
+	int ret = 0;
+
+	if (!di->i_xattr_loc)
+		return ret;
+
+	ret = ocfs2_read_block(OCFS2_SB(inode->i_sb),
+			       le64_to_cpu(di->i_xattr_loc),
+			       &blk_bh, OCFS2_BH_CACHED, inode);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return ret;
+	}
+	/*Verify the signature of xattr block*/
+	if (memcmp((void *)blk_bh->b_data, OCFS2_XATTR_BLOCK_SIGNATURE,
+		   strlen(OCFS2_XATTR_BLOCK_SIGNATURE))) {
+		ret = -EFAULT;
+		goto cleanup;
+	}
+
+	header = &((struct ocfs2_xattr_block *)blk_bh->b_data)->
+		 xb_attrs.xb_header;
+
+	ret = ocfs2_xattr_list_entries(inode, header, buffer, buffer_size);
+cleanup:
+	brelse(blk_bh);
+
+	return ret;
+}
+
+ssize_t ocfs2_listxattr(struct dentry *dentry,
+			char *buffer,
+			size_t size)
+{
+	int ret = 0, i_ret = 0, b_ret = 0;
+	struct buffer_head *di_bh = NULL;
+	struct ocfs2_dinode *di = NULL;
+	struct ocfs2_inode_info *oi = OCFS2_I(dentry->d_inode);
+
+	if (!(oi->ip_dyn_features & OCFS2_HAS_XATTR_FL))
+		return ret;
+
+	ret = ocfs2_inode_lock(dentry->d_inode, &di_bh, 0);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return ret;
+	}
+
+	di = (struct ocfs2_dinode *)di_bh->b_data;
+
+	down_read(&oi->ip_xattr_sem);
+	i_ret = ocfs2_xattr_ibody_list(dentry->d_inode, di, buffer, size);
+	if (i_ret < 0)
+		b_ret = 0;
+	else {
+		if (buffer) {
+			buffer += i_ret;
+			size -= i_ret;
+		}
+		b_ret = ocfs2_xattr_block_list(dentry->d_inode, di,
+					       buffer, size);
+		if (b_ret < 0)
+			i_ret = 0;
+	}
+	up_read(&oi->ip_xattr_sem);
+	ocfs2_inode_unlock(dentry->d_inode, 0);
+
+	brelse(di_bh);
+
+	return i_ret + b_ret;
+}
+
+static int ocfs2_xattr_find_entry(int name_index,
+				  const char *name,
+				  struct ocfs2_xattr_search *xs)
+{
+	struct ocfs2_xattr_entry *entry;
+	size_t name_len;
+	int i, cmp = 1;
+
+	if (name == NULL)
+		return -EINVAL;
+
+	name_len = strlen(name);
+	entry = xs->here;
+	for (i = 0; i < le16_to_cpu(xs->header->xh_count); i++) {
+		cmp = name_index - ocfs2_xattr_get_type(entry);
+		if (!cmp)
+			cmp = name_len - entry->xe_name_len;
+		if (!cmp)
+			cmp = memcmp(name, (xs->base +
+				     le16_to_cpu(entry->xe_name_offset)),
+				     name_len);
+		if (cmp == 0)
+			break;
+		entry += 1;
+	}
+	xs->here = entry;
+
+	return cmp ? -ENODATA : 0;
+}
+
+static int ocfs2_xattr_get_value_outside(struct inode *inode,
+					 struct ocfs2_xattr_search *xs,
+					 void *buffer,
+					 size_t len)
+{
+	u32 cpos, p_cluster, num_clusters, bpc, clusters;
+	u64 blkno;
+	int i, ret = 0;
+	size_t cplen, blocksize;
+	struct buffer_head *bh = NULL;
+	struct ocfs2_xattr_value_root *xv;
+	struct ocfs2_extent_list *el;
+
+	xv = (struct ocfs2_xattr_value_root *)
+		(xs->base + le16_to_cpu(xs->here->xe_name_offset) +
+		OCFS2_XATTR_SIZE(xs->here->xe_name_len));
+	el = &xv->xr_list;
+	clusters = le32_to_cpu(xv->xr_clusters);
+	bpc = ocfs2_clusters_to_blocks(inode->i_sb, 1);
+	blocksize = inode->i_sb->s_blocksize;
+
+	cpos = 0;
+	while (cpos < clusters) {
+		ret = ocfs2_xattr_get_clusters(inode, cpos, &p_cluster,
+					       &num_clusters, el);
+		if (ret) {
+			mlog_errno(ret);
+			goto out;
+		}
+
+		blkno = ocfs2_clusters_to_blocks(inode->i_sb, p_cluster);
+		/* Copy ocfs2_xattr_value */
+		for (i = 0; i < num_clusters * bpc; i++, blkno++) {
+			ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), blkno,
+					       &bh, OCFS2_BH_CACHED, inode);
+			if (ret) {
+				mlog_errno(ret);
+				goto out;
+			}
+
+			cplen = len >= blocksize ? blocksize : len;
+			memcpy(buffer, bh->b_data, cplen);
+			len -= cplen;
+			buffer += cplen;
+
+			brelse(bh);
+			bh = NULL;
+			if (len == 0)
+				break;
+		}
+		cpos += num_clusters;
+	}
+out:
+	return ret;
+}
+
+static int ocfs2_xattr_ibody_get(struct inode *inode,
+				 int name_index,
+				 const char *name,
+				 void *buffer,
+				 size_t buffer_size,
+				 struct ocfs2_xattr_search *xs)
+{
+	struct ocfs2_inode_info *oi = OCFS2_I(inode);
+	struct ocfs2_dinode *di = (struct ocfs2_dinode *)xs->inode_bh->b_data;
+	size_t size;
+	int ret = 0;
+
+	if (!(oi->ip_dyn_features & OCFS2_INLINE_XATTR_FL))
+		return -ENODATA;
+
+	xs->end = (void *)di + inode->i_sb->s_blocksize;
+	xs->header = (struct ocfs2_xattr_header *)
+			(xs->end - le16_to_cpu(di->i_xattr_inline_size));
+	xs->base = (void *)xs->header;
+	xs->here = xs->header->xh_entries;
+
+	ret = ocfs2_xattr_find_entry(name_index, name, xs);
+	if (ret)
+		return ret;
+	size = le64_to_cpu(xs->here->xe_value_size);
+	if (buffer) {
+		if (size > buffer_size)
+			return -ERANGE;
+		if (ocfs2_xattr_is_local(xs->here)) {
+			memcpy(buffer, (void *)xs->base +
+			       le16_to_cpu(xs->here->xe_name_offset) +
+			       OCFS2_XATTR_SIZE(xs->here->xe_name_len), size);
+		} else {
+			ret = ocfs2_xattr_get_value_outside(inode, xs,
+							    buffer, size);
+			if (ret < 0) {
+				mlog_errno(ret);
+				return ret;
+			}
+		}
+	}
+
+	return size;
+}
+
+static int ocfs2_xattr_block_get(struct inode *inode,
+				 int name_index,
+				 const char *name,
+				 void *buffer,
+				 size_t buffer_size,
+				 struct ocfs2_xattr_search *xs)
+{
+	struct ocfs2_dinode *di = (struct ocfs2_dinode *)xs->inode_bh->b_data;
+	struct buffer_head *blk_bh = NULL;
+	struct ocfs2_xattr_block *xb;
+	size_t size;
+	int ret = -ENODATA;
+
+	if (!di->i_xattr_loc)
+		return ret;
+
+	ret = ocfs2_read_block(OCFS2_SB(inode->i_sb),
+			       le64_to_cpu(di->i_xattr_loc),
+			       &blk_bh, OCFS2_BH_CACHED, inode);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return ret;
+	}
+	/*Verify the signature of xattr block*/
+	if (memcmp((void *)blk_bh->b_data, OCFS2_XATTR_BLOCK_SIGNATURE,
+		   strlen(OCFS2_XATTR_BLOCK_SIGNATURE))) {
+		ret = -EFAULT;
+		goto cleanup;
+	}
+
+	xs->xattr_bh = blk_bh;
+	xb = (struct ocfs2_xattr_block *)blk_bh->b_data;
+	xs->header = &xb->xb_attrs.xb_header;
+	xs->base = (void *)xs->header;
+	xs->end = (void *)(blk_bh->b_data) + blk_bh->b_size;
+	xs->here = xs->header->xh_entries;
+
+	ret = ocfs2_xattr_find_entry(name_index, name, xs);
+	if (ret)
+		goto cleanup;
+	size = le64_to_cpu(xs->here->xe_value_size);
+	if (buffer) {
+		ret = -ERANGE;
+		if (size > buffer_size)
+			goto cleanup;
+		if (ocfs2_xattr_is_local(xs->here)) {
+			memcpy(buffer, (void *)xs->base +
+			       le16_to_cpu(xs->here->xe_name_offset) +
+			       OCFS2_XATTR_SIZE(xs->here->xe_name_len), size);
+		} else {
+			ret = ocfs2_xattr_get_value_outside(inode, xs,
+							    buffer, size);
+			if (ret < 0) {
+				mlog_errno(ret);
+				goto cleanup;
+			}
+		}
+	}
+	ret = size;
+cleanup:
+	brelse(blk_bh);
+
+	return ret;
+}
+
+/* ocfs2_xattr_get()
+ *
+ * Copy an extended attribute into the buffer provided.
+ * Buffer is NULL to compute the size of buffer required.
+ */
+int ocfs2_xattr_get(struct inode *inode,
+		    int name_index,
+		    const char *name,
+		    void *buffer,
+		    size_t buffer_size)
+{
+	int ret;
+	struct ocfs2_dinode *di = NULL;
+	struct buffer_head *di_bh = NULL;
+	struct ocfs2_inode_info *oi = OCFS2_I(inode);
+	struct ocfs2_xattr_search xis = {
+		.not_found = -ENODATA,
+	};
+	struct ocfs2_xattr_search xbs = {
+		.not_found = -ENODATA,
+	};
+
+	if (!(oi->ip_dyn_features & OCFS2_HAS_XATTR_FL))
+		ret = -ENODATA;
+
+	ret = ocfs2_inode_lock(inode, &di_bh, 0);
+	if (ret < 0) {
+		mlog_errno(ret);
+		return ret;
+	}
+	xis.inode_bh = xbs.inode_bh = di_bh;
+	di = (struct ocfs2_dinode *)di_bh->b_data;
+
+	down_read(&oi->ip_xattr_sem);
+	ret = ocfs2_xattr_ibody_get(inode, name_index, name, buffer,
+				    buffer_size, &xis);
+	if (ret == -ENODATA)
+		ret = ocfs2_xattr_block_get(inode, name_index, name, buffer,
+					    buffer_size, &xbs);
+	up_read(&oi->ip_xattr_sem);
+	ocfs2_inode_unlock(inode, 0);
+
+	brelse(di_bh);
+
+	return ret;
+}
+
+static int __ocfs2_xattr_set_value_outside(struct inode *inode,
+					   struct ocfs2_xattr_value_root *xv,
+					   const void *value,
+					   int value_len)
+{
+	int ret = 0, i, cp_len, credits;
+	u16 blocksize = inode->i_sb->s_blocksize;
+	u32 p_cluster, num_clusters;
+	u32 cpos = 0, bpc = ocfs2_clusters_to_blocks(inode->i_sb, 1);
+	u32 clusters = ocfs2_clusters_for_bytes(inode->i_sb, value_len);
+	u64 blkno;
+	struct buffer_head *bh = NULL;
+	handle_t *handle;
+
+	BUG_ON(clusters > le32_to_cpu(xv->xr_clusters));
+
+	credits = clusters * bpc;
+	handle = ocfs2_start_trans(OCFS2_SB(inode->i_sb), credits);
+	if (IS_ERR(handle)) {
+		ret = PTR_ERR(handle);
+		mlog_errno(ret);
+		goto out;
+	}
+
+	while (cpos < clusters) {
+		ret = ocfs2_xattr_get_clusters(inode, cpos, &p_cluster,
+					       &num_clusters, &xv->xr_list);
+		if (ret) {
+			mlog_errno(ret);
+			goto out_commit;
+		}
+
+		blkno = ocfs2_clusters_to_blocks(inode->i_sb, p_cluster);
+
+		for (i = 0; i < num_clusters * bpc; i++, blkno++) {
+			ret = ocfs2_read_block(OCFS2_SB(inode->i_sb), blkno,
+					       &bh, OCFS2_BH_CACHED, inode);
+			if (ret) {
+				mlog_errno(ret);
+				goto out_commit;
+			}
+
+			ret = ocfs2_journal_access(handle,
+						   inode,
+						   bh,
+						   OCFS2_JOURNAL_ACCESS_WRITE);
+			if (ret < 0) {
+				mlog_errno(ret);
+				goto out_commit;
+			}
+
+			cp_len = value_len > blocksize ? blocksize : value_len;
+			memcpy(bh->b_data, value, cp_len);
+			value_len -= cp_len;
+			value += cp_len;
+			if (cp_len < blocksize)
+				memset(bh->b_data + cp_len, 0,
+				       blocksize - cp_len);
+
+			ret = ocfs2_journal_dirty(handle, bh);
+			if (ret < 0) {
+				mlog_errno(ret);
+				goto out_commit;
+			}
+			brelse(bh);
+			bh = NULL;
+
+			/*
+			 * XXX: do we need to empty all the following
+			 * blocks in this cluster?
+			 */
+			if (!value_len)
+				break;
+		}
+		cpos += num_clusters;
+	}
+out_commit:
+	ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
+out:
+	brelse(bh);
+
+	return ret;
+}
+
+static int ocfs2_xattr_cleanup(struct inode *inode,
+			       struct ocfs2_xattr_info *xi,
+			       struct ocfs2_xattr_search *xs,
+			       size_t offs)
+{
+	handle_t *handle = NULL;
+	int ret = 0;
+	size_t name_len = strlen(xi->name);
+	void *val = xs->base + offs;
+	size_t size = OCFS2_XATTR_SIZE(name_len) + OCFS2_XATTR_ROOT_SIZE;
+
+	handle = ocfs2_start_trans((OCFS2_SB(inode->i_sb)),
+				   OCFS2_XATTR_BLOCK_UPDATE_CREDITS);
+	if (IS_ERR(handle)) {
+		ret = PTR_ERR(handle);
+		mlog_errno(ret);
+		goto out;
+	}
+	ret = ocfs2_journal_access(handle, inode, xs->xattr_bh,
+				   OCFS2_JOURNAL_ACCESS_WRITE);
+	if (ret) {
+		mlog_errno(ret);
+		goto out_commit;
+	}
+	/* Decrease xattr count */
+	le16_add_cpu(&xs->header->xh_count, -1);
+	/* Remove the xattr entry and tree root which has already be set*/
+	memset((void *)xs->here, 0, sizeof(struct ocfs2_xattr_entry));
+	memset(val, 0, size);
+
+	ret = ocfs2_journal_dirty(handle, xs->xattr_bh);
+	if (ret < 0)
+		mlog_errno(ret);
+out_commit:
+	ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
+out:
+	return ret;
+}
+
+static int ocfs2_xattr_update_entry(struct inode *inode,
+				    struct ocfs2_xattr_info *xi,
+				    struct ocfs2_xattr_search *xs,
+				    size_t offs)
+{
+	handle_t *handle = NULL;
+	int ret = 0;
+
+	handle = ocfs2_start_trans((OCFS2_SB(inode->i_sb)),
+				   OCFS2_XATTR_BLOCK_UPDATE_CREDITS);
+	if (IS_ERR(handle)) {
+		ret = PTR_ERR(handle);
+		mlog_errno(ret);
+		goto out;
+	}
+	ret = ocfs2_journal_access(handle, inode, xs->xattr_bh,
+				   OCFS2_JOURNAL_ACCESS_WRITE);
+	if (ret) {
+		mlog_errno(ret);
+		goto out_commit;
+	}
+
+	xs->here->xe_name_offset = cpu_to_le16(offs);
+	xs->here->xe_value_size = cpu_to_le64(xi->value_len);
+	if (xi->value_len <= OCFS2_XATTR_INLINE_SIZE)
+		ocfs2_xattr_set_local(xs->here, 1);
+	else
+		ocfs2_xattr_set_local(xs->here, 0);
+	ocfs2_xattr_hash_entry(inode, xs->header, xs->here);
+
+	ret = ocfs2_journal_dirty(handle, xs->xattr_bh);
+	if (ret < 0)
+		mlog_errno(ret);
+out_commit:
+	ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
+out:
+	return ret;
+}
+
+/*
+ * ocfs2_xattr_set_value_outside()
+ *
+ * Set large size value in B tree.
+ */
+static int ocfs2_xattr_set_value_outside(struct inode *inode,
+					 struct ocfs2_xattr_info *xi,
+					 struct ocfs2_xattr_search *xs,
+					 size_t offs)</