Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6

Conflicts:

	drivers/message/fusion/mptlan.c
	drivers/net/sfc/ethtool.c
	net/mac80211/debugfs_sta.c

23 files changed, 384 insertions, 263 deletions

diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index fe34d74cfb19..2a117e286e54 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -718,6 +718,8 @@ got:
                         gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
                         free = ext4_free_blocks_after_init(sb, group, gdp);
                         gdp->bg_free_blocks_count = cpu_to_le16(free);
+                        gdp->bg_checksum = ext4_group_desc_csum(sbi, group,
+                                                                gdp);
                 }
                 spin_unlock(sb_bgl_lock(sbi, group));
 
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 8dbf6953845b..be21a5ae33cb 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2329,6 +2329,8 @@ static int ext4_da_writepage(struct page *page,
                         unlock_page(page);
                         return 0;
                 }
+                /* now mark the buffer_heads as dirty and uptodate */
+                block_commit_write(page, 0, PAGE_CACHE_SIZE);
         }
 
         if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode))
@@ -4580,9 +4582,10 @@ static int ext4_indirect_trans_blocks(struct inode *inode, int nrblocks,
 static int ext4_index_trans_blocks(struct inode *inode, int nrblocks, int chunk)
 {
         if (!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL))
-                return ext4_indirect_trans_blocks(inode, nrblocks, 0);
-        return ext4_ext_index_trans_blocks(inode, nrblocks, 0);
+                return ext4_indirect_trans_blocks(inode, nrblocks, chunk);
+        return ext4_ext_index_trans_blocks(inode, nrblocks, chunk);
 }
+
 /*
  * Account for index blocks, block groups bitmaps and block group
  * descriptor blocks if modify datablocks and index blocks
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index dfe17a134052..444ad998f72e 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -4441,6 +4441,7 @@ ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b,
                 else if (block >= (entry->start_blk + entry->count))
                         n = &(*n)->rb_right;
                 else {
+                        ext4_unlock_group(sb, group);
                         ext4_error(sb, __func__,
                             "Double free of blocks %d (%d %d)\n",
                             block, entry->start_blk, entry->count);
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 994859df010e..e4a241c65dbe 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1458,9 +1458,8 @@ static int ext4_fill_flex_info(struct super_block *sb)
 
         /* We allocate both existing and potentially added groups */
         flex_group_count = ((sbi->s_groups_count + groups_per_flex - 1) +
-                            ((sbi->s_es->s_reserved_gdt_blocks +1 ) <<
-                              EXT4_DESC_PER_BLOCK_BITS(sb))) /
-                           groups_per_flex;
+                        ((le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks) + 1) <<
+                              EXT4_DESC_PER_BLOCK_BITS(sb))) / groups_per_flex;
         sbi->s_flex_groups = kzalloc(flex_group_count *
                                      sizeof(struct flex_groups), GFP_KERNEL);
         if (sbi->s_flex_groups == NULL) {
@@ -2885,12 +2884,9 @@ int ext4_force_commit(struct super_block *sb)
 /*
  * Ext4 always journals updates to the superblock itself, so we don't
  * have to propagate any other updates to the superblock on disk at this
- * point.  Just start an async writeback to get the buffers on their way
- * to the disk.
- *
- * This implicitly triggers the writebehind on sync().
+ * point.  (We can probably nuke this function altogether, and remove
+ * any mention to sb->s_dirt in all of fs/ext4; eventual cleanup...)
  */
-
 static void ext4_write_super(struct super_block *sb)
 {
         if (mutex_trylock(&sb->s_lock) != 0)
@@ -2900,15 +2896,15 @@ static void ext4_write_super(struct super_block *sb)
 
 static int ext4_sync_fs(struct super_block *sb, int wait)
 {
-        tid_t target;
+        int ret = 0;
 
         trace_mark(ext4_sync_fs, "dev %s wait %d", sb->s_id, wait);
         sb->s_dirt = 0;
-        if (jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, &target)) {
-                if (wait)
-                        jbd2_log_wait_commit(EXT4_SB(sb)->s_journal, target);
-        }
-        return 0;
+        if (wait)
+                ret = ext4_force_commit(sb);
+        else
+                jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, NULL);
+        return ret;
 }
 
 /*
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index 1bd8d4acc6f2..61f32f3868cd 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -115,7 +115,7 @@ static int __try_to_free_cp_buf(struct journal_head *jh)
  */
 void __log_wait_for_space(journal_t *journal)
 {
-        int nblocks;
+        int nblocks, space_left;
         assert_spin_locked(&journal->j_state_lock);
 
         nblocks = jbd_space_needed(journal);
@@ -128,25 +128,42 @@ void __log_wait_for_space(journal_t *journal)
                 /*
                  * Test again, another process may have checkpointed while we
                  * were waiting for the checkpoint lock. If there are no
-                 * outstanding transactions there is nothing to checkpoint and
-                 * we can't make progress. Abort the journal in this case.
+                 * transactions ready to be checkpointed, try to recover
+                 * journal space by calling cleanup_journal_tail(), and if
+                 * that doesn't work, by waiting for the currently committing
+                 * transaction to complete.  If there is absolutely no way
+                 * to make progress, this is either a BUG or corrupted
+                 * filesystem, so abort the journal and leave a stack
+                 * trace for forensic evidence.
                  */
                 spin_lock(&journal->j_state_lock);
                 spin_lock(&journal->j_list_lock);
                 nblocks = jbd_space_needed(journal);
-                if (__log_space_left(journal) < nblocks) {
+                space_left = __log_space_left(journal);
+                if (space_left < nblocks) {
                         int chkpt = journal->j_checkpoint_transactions != NULL;
+                        tid_t tid = 0;
 
+                        if (journal->j_committing_transaction)
+                                tid = journal->j_committing_transaction->t_tid;
                         spin_unlock(&journal->j_list_lock);
                         spin_unlock(&journal->j_state_lock);
                         if (chkpt) {
                                 log_do_checkpoint(journal);
+                        } else if (cleanup_journal_tail(journal) == 0) {
+                                /* We were able to recover space; yay! */
+                                ;
+                        } else if (tid) {
+                                log_wait_commit(journal, tid);
                         } else {
-                                printk(KERN_ERR "%s: no transactions\n",
-                                       __func__);
+                                printk(KERN_ERR "%s: needed %d blocks and "
+                                       "only had %d space available\n",
+                                       __func__, nblocks, space_left);
+                                printk(KERN_ERR "%s: no way to get more "
+                                       "journal space\n", __func__);
+                                WARN_ON(1);
                                 journal_abort(journal, 0);
                         }
-
                         spin_lock(&journal->j_state_lock);
                 } else {
                         spin_unlock(&journal->j_list_lock);
diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
index 9203c3332f17..9497718fe920 100644
--- a/fs/jbd2/checkpoint.c
+++ b/fs/jbd2/checkpoint.c
@@ -116,7 +116,7 @@ static int __try_to_free_cp_buf(struct journal_head *jh)
  */
 void __jbd2_log_wait_for_space(journal_t *journal)
 {
-        int nblocks;
+        int nblocks, space_left;
         assert_spin_locked(&journal->j_state_lock);
 
         nblocks = jbd_space_needed(journal);
@@ -129,25 +129,43 @@ void __jbd2_log_wait_for_space(journal_t *journal)
                 /*
                  * Test again, another process may have checkpointed while we
                  * were waiting for the checkpoint lock. If there are no
-                 * outstanding transactions there is nothing to checkpoint and
-                 * we can't make progress. Abort the journal in this case.
+                 * transactions ready to be checkpointed, try to recover
+                 * journal space by calling cleanup_journal_tail(), and if
+                 * that doesn't work, by waiting for the currently committing
+                 * transaction to complete.  If there is absolutely no way
+                 * to make progress, this is either a BUG or corrupted
+                 * filesystem, so abort the journal and leave a stack
+                 * trace for forensic evidence.
                  */
                 spin_lock(&journal->j_state_lock);
                 spin_lock(&journal->j_list_lock);
                 nblocks = jbd_space_needed(journal);
-                if (__jbd2_log_space_left(journal) < nblocks) {
+                space_left = __jbd2_log_space_left(journal);
+                if (space_left < nblocks) {
                         int chkpt = journal->j_checkpoint_transactions != NULL;
+                        tid_t tid = 0;
 
+                        if (journal->j_committing_transaction)
+                                tid = journal->j_committing_transaction->t_tid;
                         spin_unlock(&journal->j_list_lock);
                         spin_unlock(&journal->j_state_lock);
                         if (chkpt) {
                                 jbd2_log_do_checkpoint(journal);
+                        } else if (jbd2_cleanup_journal_tail(journal) == 0) {
+                                /* We were able to recover space; yay! */
+                                ;
+                        } else if (tid) {
+                                jbd2_log_wait_commit(journal, tid);
                         } else {
-                                printk(KERN_ERR "%s: no transactions\n",
-                                       __func__);
+                                printk(KERN_ERR "%s: needed %d blocks and "
+                                       "only had %d space available\n",
+                                       __func__, nblocks, space_left);
+                                printk(KERN_ERR "%s: no way to get more "
+                                       "journal space in %s\n", __func__,
+                                       journal->j_devname);
+                                WARN_ON(1);
                                 jbd2_journal_abort(journal, 0);
                         }
-
                         spin_lock(&journal->j_state_lock);
                 } else {
                         spin_unlock(&journal->j_list_lock);
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 783de118de92..e70d657a19f8 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -1089,6 +1089,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
         if (!journal->j_wbuf) {
                 printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n",
                         __func__);
+                jbd2_stats_proc_exit(journal);
                 kfree(journal);
                 return NULL;
         }
@@ -1098,6 +1099,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
         if (err) {
                 printk(KERN_ERR "%s: Cannnot locate journal superblock\n",
                        __func__);
+                jbd2_stats_proc_exit(journal);
                 kfree(journal);
                 return NULL;
         }
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 848a03e83a42..4433c8f00163 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1875,11 +1875,11 @@ static int nfsd_buffered_readdir(struct file *file, filldir_t func,
                 return -ENOMEM;
 
         offset = *offsetp;
-        cdp->err = nfserr_eof; /* will be cleared on successful read */
 
         while (1) {
                 unsigned int reclen;
 
+                cdp->err = nfserr_eof; /* will be cleared on successful read */
                 buf.used = 0;
                 buf.full = 0;
 
@@ -1912,9 +1912,6 @@ static int nfsd_buffered_readdir(struct file *file, filldir_t func,
                         de = (struct buffered_dirent *)((char *)de + reclen);
                 }
                 offset = vfs_llseek(file, 0, SEEK_CUR);
-                cdp->err = nfserr_eof;
-                if (!buf.full)
-                        break;
         }
 
  done:
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index 7efe937a415f..e2570a3bc2b2 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -247,8 +247,8 @@ int ocfs2_update_inode_atime(struct inode *inode,
         mlog_entry_void();
 
         handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
-        if (handle == NULL) {
-                ret = -ENOMEM;
+        if (IS_ERR(handle)) {
+                ret = PTR_ERR(handle);
                 mlog_errno(ret);
                 goto out;
         }
@@ -312,8 +312,8 @@ static int ocfs2_simple_size_update(struct inode *inode,
         handle_t *handle = NULL;
 
         handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
-        if (handle == NULL) {
-                ret = -ENOMEM;
+        if (IS_ERR(handle)) {
+                ret = PTR_ERR(handle);
                 mlog_errno(ret);
                 goto out;
         }
@@ -1055,8 +1055,8 @@ static int __ocfs2_write_remove_suid(struct inode *inode,
                    (unsigned long long)OCFS2_I(inode)->ip_blkno, inode->i_mode);
 
         handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
-        if (handle == NULL) {
-                ret = -ENOMEM;
+        if (IS_ERR(handle)) {
+                ret = PTR_ERR(handle);
                 mlog_errno(ret);
                 goto out;
         }
@@ -1259,8 +1259,8 @@ static int __ocfs2_remove_inode_range(struct inode *inode,
         }
 
         handle = ocfs2_start_trans(osb, OCFS2_REMOVE_EXTENT_CREDITS);
-        if (handle == NULL) {
-                ret = -ENOMEM;
+        if (IS_ERR(handle)) {
+                ret = PTR_ERR(handle);
                 mlog_errno(ret);
                 goto out;
         }
@@ -1352,8 +1352,8 @@ static int ocfs2_zero_partial_clusters(struct inode *inode,
                 goto out;
 
         handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
-        if (handle == NULL) {
-                ret = -ENOMEM;
+        if (IS_ERR(handle)) {
+                ret = PTR_ERR(handle);
                 mlog_errno(ret);
                 goto out;
         }
@@ -1866,6 +1866,13 @@ relock:
                 written = generic_file_direct_write(iocb, iov, &nr_segs, *ppos,
                                                     ppos, count, ocount);
                 if (written < 0) {
+                        /*
+                         * direct write may have instantiated a few
+                         * blocks outside i_size. Trim these off again.
+                         * Don't need i_size_read because we hold i_mutex.
+                         */
+                        if (*ppos + count > inode->i_size)
+                                vmtruncate(inode, inode->i_size);
                         ret = written;
                         goto out_dio;
                 }
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 4903688f72a9..7aa00d511874 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -1106,6 +1106,12 @@ void ocfs2_clear_inode(struct inode *inode)
         oi->ip_last_trans = 0;
         oi->ip_dir_start_lookup = 0;
         oi->ip_blkno = 0ULL;
+
+        /*
+         * ip_jinode is used to track txns against this inode. We ensure that
+         * the journal is flushed before journal shutdown. Thus it is safe to
+         * have inodes get cleaned up after journal shutdown.
+         */
         jbd2_journal_release_jbd_inode(OCFS2_SB(inode->i_sb)->journal->j_journal,
                                        &oi->ip_jinode);
 
diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index 81e40677eecb..99fe9d584f3c 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -690,6 +690,7 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 
         /* Shutdown the kernel journal system */
         jbd2_journal_destroy(journal->j_journal);
+        journal->j_journal = NULL;
 
         OCFS2_I(inode)->ip_open_count--;
 
diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c
index 3dc18d67557c..eea1d24713ea 100644
--- a/fs/ocfs2/mmap.c
+++ b/fs/ocfs2/mmap.c
@@ -113,7 +113,11 @@ static int __ocfs2_page_mkwrite(struct inode *inode, struct buffer_head *di_bh,
          * ocfs2_write_begin_nolock().
          */
         if (!PageUptodate(page) || page->mapping != inode->i_mapping) {
-                ret = -EINVAL;
+                /*
+                 * the page has been umapped in ocfs2_data_downconvert_worker.
+                 * So return 0 here and let VFS retry.
+                 */
+                ret = 0;
                 goto out;
         }
 
diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 485a6aa0ad39..f4967e634ffd 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -378,8 +378,8 @@ static int ocfs2_mknod_locked(struct ocfs2_super *osb,
         }
 
         inode = new_inode(dir->i_sb);
-        if (IS_ERR(inode)) {
-                status = PTR_ERR(inode);
+        if (!inode) {
+                status = -ENOMEM;
                 mlog(ML_ERROR, "new_inode failed!\n");
                 goto leave;
         }
@@ -491,8 +491,10 @@ leave:
                         brelse(*new_fe_bh);
                         *new_fe_bh = NULL;
                 }
-                if (inode)
+                if (inode) {
+                        clear_nlink(inode);
                         iput(inode);
+                }
         }
 
         mlog_exit(status);
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h
index a21a465490c4..fef7ece32376 100644
--- a/fs/ocfs2/ocfs2.h
+++ b/fs/ocfs2/ocfs2.h
@@ -473,6 +473,9 @@ static inline int ocfs2_uses_extended_slot_map(struct ocfs2_super *osb)
                 (____gd)->bg_signature);                                \
 } while (0)
 
+#define OCFS2_IS_VALID_XATTR_BLOCK(ptr)                                 \
+        (!strcmp((ptr)->xb_signature, OCFS2_XATTR_BLOCK_SIGNATURE))
+
 static inline unsigned long ino_from_blkno(struct super_block *sb,
                                            u64 blkno)
 {
diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h
index f24ce3d3f956..5f180cf7abbd 100644
--- a/fs/ocfs2/ocfs2_fs.h
+++ b/fs/ocfs2/ocfs2_fs.h
@@ -742,12 +742,12 @@ struct ocfs2_group_desc
  */
 struct ocfs2_xattr_entry {
         __le32  xe_name_hash;    /* hash value of xattr prefix+suffix. */
-        __le16  xe_name_offset;  /* byte offset from the 1st etnry in the local
+        __le16  xe_name_offset;  /* byte offset from the 1st entry in the
                                     local xattr storage(inode, xattr block or
                                     xattr bucket). */
         __u8    xe_name_len;     /* xattr name len, does't include prefix. */
-        __u8    xe_type;         /* the low 7 bits indicates the name prefix's
-                                  * type and the highest 1 bits indicate whether
+        __u8    xe_type;         /* the low 7 bits indicate the name prefix
+                                  * type and the highest bit indicates whether
                                   * the EA is stored in the local storage. */
         __le64  xe_value_size;   /* real xattr value length. */
 };
@@ -766,9 +766,10 @@ struct ocfs2_xattr_header {
                                                    xattr. */
         __le16  xh_name_value_len;              /* total length of name/value
                                                    length in this bucket. */
-        __le16  xh_num_buckets;                 /* bucket nums in one extent
-                                                   record, only valid in the
-                                                   first bucket. */
+        __le16  xh_num_buckets;                 /* Number of xattr buckets
+                                                   in this extent record,
+                                                   only valid in the first
+                                                   bucket. */
         __le64  xh_csum;
         struct ocfs2_xattr_entry xh_entries[0]; /* xattr entry list. */
 };
@@ -776,8 +777,8 @@ struct ocfs2_xattr_header {
 /*
  * On disk structure for xattr value root.
  *
- * It is used when one extended attribute's size is larger, and we will save it
- * in an outside cluster. It will stored in a b-tree like file content.
+ * When an xattr's value is large enough, it is stored in an external
+ * b-tree like file data.  The xattr value root points to this structure.
  */
 struct ocfs2_xattr_value_root {
 /*00*/  __le32  xr_clusters;              /* clusters covered by xattr value. */
diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c
index 802c41492214..054e2efb0b7e 100644
--- a/fs/ocfs2/xattr.c
+++ b/fs/ocfs2/xattr.c
@@ -3,25 +3,20 @@
  *
  * xattr.c
  *
- * Copyright (C) 2008 Oracle.  All rights reserved.
+ * Copyright (C) 2004, 2008 Oracle.  All rights reserved.
  *
  * CREDITS:
- * Lots of code in this file is taken from ext3.
+ * Lots of code in this file is copy from linux/fs/ext3/xattr.c.
+ * Copyright (C) 2001-2003 Andreas Gruenbacher, <agruen@suse.de>
  *
  * 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
- * version 2 of the License, or (at your option) any later version.
+ * License version 2 as published by the Free Software Foundation.
  *
  * This program 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 this program; if not, write to the
- * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- * Boston, MA 021110-1307, USA.
  */
 
 #include <linux/capability.h>
@@ -83,7 +78,7 @@ struct xattr_handler *ocfs2_xattr_handlers[] = {
         NULL
 };
 
-static struct xattr_handler *ocfs2_xattr_handler_map[] = {
+static struct xattr_handler *ocfs2_xattr_handler_map[OCFS2_XATTR_MAX] = {
         [OCFS2_XATTR_INDEX_USER]        = &ocfs2_xattr_user_handler,
         [OCFS2_XATTR_INDEX_TRUSTED]     = &ocfs2_xattr_trusted_handler,
 };
@@ -116,6 +111,10 @@ static int ocfs2_xattr_bucket_get_name_value(struct inode *inode,
                                              int *block_off,
                                              int *new_offset);
 
+static int ocfs2_xattr_block_find(struct inode *inode,
+                                  int name_index,
+                                  const char *name,
+                                  struct ocfs2_xattr_search *xs);
 static int ocfs2_xattr_index_block_find(struct inode *inode,
                                         struct buffer_head *root_bh,
                                         int name_index,
@@ -137,6 +136,24 @@ static int ocfs2_xattr_set_entry_index_block(struct inode *inode,
 static int ocfs2_delete_xattr_index_block(struct inode *inode,
                                           struct buffer_head *xb_bh);
 
+static inline u16 ocfs2_xattr_buckets_per_cluster(struct ocfs2_super *osb)
+{
+        return (1 << osb->s_clustersize_bits) / OCFS2_XATTR_BUCKET_SIZE;
+}
+
+static inline u16 ocfs2_blocks_per_xattr_bucket(struct super_block *sb)
+{
+        return OCFS2_XATTR_BUCKET_SIZE / (1 << sb->s_blocksize_bits);
+}
+
+static inline u16 ocfs2_xattr_max_xe_in_bucket(struct super_block *sb)
+{
+        u16 len = sb->s_blocksize -
+                 offsetof(struct ocfs2_xattr_header, xh_entries);
+
+        return len / sizeof(struct ocfs2_xattr_entry);
+}
+
 static inline const char *ocfs2_xattr_prefix(int name_index)
 {
         struct xattr_handler *handler = NULL;
@@ -542,14 +559,12 @@ static int ocfs2_xattr_block_list(struct inode *inode,
                 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;
-        }
 
         xb = (struct ocfs2_xattr_block *)blk_bh->b_data;
+        if (!OCFS2_IS_VALID_XATTR_BLOCK(xb)) {
+                ret = -EIO;
+                goto cleanup;
+        }
 
         if (!(le16_to_cpu(xb->xb_flags) & OCFS2_XATTR_INDEXED)) {
                 struct ocfs2_xattr_header *header = &xb->xb_attrs.xb_header;
@@ -749,47 +764,25 @@ static int ocfs2_xattr_block_get(struct inode *inode,
                                  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;
         struct ocfs2_xattr_value_root *xv;
         size_t size;
         int ret = -ENODATA, name_offset, name_len, block_off, i;
 
-        if (!di->i_xattr_loc)
-                return ret;
-
         memset(&xs->bucket, 0, sizeof(xs->bucket));
 
-        ret = ocfs2_read_block(inode, le64_to_cpu(di->i_xattr_loc), &blk_bh);
-        if (ret < 0) {
+        ret = ocfs2_xattr_block_find(inode, name_index, name, xs);
+        if (ret) {
                 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;
-
-        if (!(le16_to_cpu(xb->xb_flags) & OCFS2_XATTR_INDEXED)) {
-                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);
-        } else
-                ret = ocfs2_xattr_index_block_find(inode, blk_bh,
-                                                   name_index,
-                                                   name, xs);
-
-        if (ret)
+        if (xs->not_found) {
+                ret = -ENODATA;
                 goto cleanup;
+        }
+
+        xb = (struct ocfs2_xattr_block *)xs->xattr_bh->b_data;
         size = le64_to_cpu(xs->here->xe_value_size);
         if (buffer) {
                 ret = -ERANGE;
@@ -828,7 +821,8 @@ cleanup:
                 brelse(xs->bucket.bhs[i]);
         memset(&xs->bucket, 0, sizeof(xs->bucket));
 
-        brelse(blk_bh);
+        brelse(xs->xattr_bh);
+        xs->xattr_bh = NULL;
         return ret;
 }
 
@@ -837,11 +831,11 @@ cleanup:
  * 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)
+static 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;
@@ -871,7 +865,7 @@ int ocfs2_xattr_get(struct inode *inode,
         down_read(&oi->ip_xattr_sem);
         ret = ocfs2_xattr_ibody_get(inode, name_index, name, buffer,
                                     buffer_size, &xis);
-        if (ret == -ENODATA)
+        if (ret == -ENODATA && di->i_xattr_loc)
                 ret = ocfs2_xattr_block_get(inode, name_index, name, buffer,
                                             buffer_size, &xbs);
         up_read(&oi->ip_xattr_sem);
@@ -1229,7 +1223,7 @@ static int ocfs2_xattr_set_entry(struct inode *inode,
 
         free = min_offs - ((void *)last - xs->base) - sizeof(__u32);
         if (free < 0)
-                return -EFAULT;
+                return -EIO;
 
         if (!xs->not_found) {
                 size_t size = 0;
@@ -1514,10 +1508,9 @@ static int ocfs2_xattr_free_block(struct inode *inode,
                 goto out;
         }
 
-        /*Verify the signature of xattr block*/
-        if (memcmp((void *)blk_bh->b_data, OCFS2_XATTR_BLOCK_SIGNATURE,
-                   strlen(OCFS2_XATTR_BLOCK_SIGNATURE))) {
-                ret = -EFAULT;
+        xb = (struct ocfs2_xattr_block *)blk_bh->b_data;
+        if (!OCFS2_IS_VALID_XATTR_BLOCK(xb)) {
+                ret = -EIO;
                 goto out;
         }
 
@@ -1527,7 +1520,6 @@ static int ocfs2_xattr_free_block(struct inode *inode,
                 goto out;
         }
 
-        xb = (struct ocfs2_xattr_block *)blk_bh->b_data;
         blk = le64_to_cpu(xb->xb_blkno);
         bit = le16_to_cpu(xb->xb_suballoc_bit);
         bg_blkno = ocfs2_which_suballoc_group(blk, bit);
@@ -1771,15 +1763,14 @@ static int ocfs2_xattr_block_find(struct inode *inode,
                 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;
+
+        xb = (struct ocfs2_xattr_block *)blk_bh->b_data;
+        if (!OCFS2_IS_VALID_XATTR_BLOCK(xb)) {
+                ret = -EIO;
+                goto cleanup;
         }
 
         xs->xattr_bh = blk_bh;
-        xb = (struct ocfs2_xattr_block *)blk_bh->b_data;
 
         if (!(le16_to_cpu(xb->xb_flags) & OCFS2_XATTR_INDEXED)) {
                 xs->header = &xb->xb_attrs.xb_header;
@@ -1806,52 +1797,6 @@ cleanup:
 }
 
 /*
- * When all the xattrs are deleted from index btree, the ocfs2_xattr_tree
- * will be erased and ocfs2_xattr_block will have its ocfs2_xattr_header
- * re-initialized.
- */
-static int ocfs2_restore_xattr_block(struct inode *inode,
-                                     struct ocfs2_xattr_search *xs)
-{
-        int ret;
-        handle_t *handle;
-        struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
-        struct ocfs2_xattr_block *xb =
-                (struct ocfs2_xattr_block *)xs->xattr_bh->b_data;
-        struct ocfs2_extent_list *el = &xb->xb_attrs.xb_root.xt_list;
-        u16 xb_flags = le16_to_cpu(xb->xb_flags);
-
-        BUG_ON(!(xb_flags & OCFS2_XATTR_INDEXED) ||
-                le16_to_cpu(el->l_next_free_rec) != 0);
-
-        handle = ocfs2_start_trans(osb, OCFS2_XATTR_BLOCK_UPDATE_CREDITS);
-        if (IS_ERR(handle)) {
-                ret = PTR_ERR(handle);
-                handle = NULL;
-                goto out;
-        }
-
-        ret = ocfs2_journal_access(handle, inode, xs->xattr_bh,
-                                   OCFS2_JOURNAL_ACCESS_WRITE);
-        if (ret < 0) {
-                mlog_errno(ret);
-                goto out_commit;
-        }
-
-        memset(&xb->xb_attrs, 0, inode->i_sb->s_blocksize -
-               offsetof(struct ocfs2_xattr_block, xb_attrs));
-
-        xb->xb_flags = cpu_to_le16(xb_flags & ~OCFS2_XATTR_INDEXED);
-
-        ocfs2_journal_dirty(handle, xs->xattr_bh);
-
-out_commit:
-        ocfs2_commit_trans(osb, handle);
-out:
-        return ret;
-}
-
-/*
  * ocfs2_xattr_block_set()
  *
  * Set, replace or remove an extended attribute into external block.
@@ -1961,8 +1906,6 @@ out:
         }
 
         ret = ocfs2_xattr_set_entry_index_block(inode, xi, xs);
-        if (!ret && xblk->xb_attrs.xb_root.xt_list.l_next_free_rec == 0)
-                ret = ocfs2_restore_xattr_block(inode, xs);
 
 end:
 
@@ -2398,7 +2341,8 @@ static int ocfs2_xattr_index_block_find(struct inode *inode,
         BUG_ON(p_blkno == 0 || num_clusters == 0 || first_hash > name_hash);
 
         mlog(0, "find xattr extent rec %u clusters from %llu, the first hash "
-             "in the rec is %u\n", num_clusters, p_blkno, first_hash);
+             "in the rec is %u\n", num_clusters, (unsigned long long)p_blkno,
+             first_hash);
 
         ret = ocfs2_xattr_bucket_find(inode, name_index, name, name_hash,
                                       p_blkno, first_hash, num_clusters, xs);
@@ -2422,7 +2366,7 @@ static int ocfs2_iterate_xattr_buckets(struct inode *inode,
         memset(&bucket, 0, sizeof(bucket));
 
         mlog(0, "iterating xattr buckets in %u clusters starting from %llu\n",
-             clusters, blkno);
+             clusters, (unsigned long long)blkno);
 
         for (i = 0; i < num_buckets; i++, blkno += blk_per_bucket) {
                 ret = ocfs2_read_blocks(inode, blkno, blk_per_bucket,
@@ -2440,7 +2384,8 @@ static int ocfs2_iterate_xattr_buckets(struct inode *inode,
                 if (i == 0)
                         num_buckets = le16_to_cpu(bucket.xh->xh_num_buckets);
 
-                mlog(0, "iterating xattr bucket %llu, first hash %u\n", blkno,
+                mlog(0, "iterating xattr bucket %llu, first hash %u\n",
+                     (unsigned long long)blkno,
                      le32_to_cpu(bucket.xh->xh_entries[0].xe_name_hash));
                 if (func) {
                         ret = func(inode, &bucket, para);
@@ -2776,7 +2721,8 @@ static int ocfs2_xattr_create_index_block(struct inode *inode,
          */
         blkno = ocfs2_clusters_to_blocks(inode->i_sb, bit_off);
 
-        mlog(0, "allocate 1 cluster from %llu to xattr block\n", blkno);
+        mlog(0, "allocate 1 cluster from %llu to xattr block\n",
+             (unsigned long long)blkno);
 
         xh_bh = sb_getblk(inode->i_sb, blkno);
         if (!xh_bh) {
@@ -2818,7 +2764,11 @@ static int ocfs2_xattr_create_index_block(struct inode *inode,
         if (data_bh)
                 ocfs2_journal_dirty(handle, data_bh);
 
-        ocfs2_xattr_update_xattr_search(inode, xs, xb_bh, xh_bh);
+        ret = ocfs2_xattr_update_xattr_search(inode, xs, xb_bh, xh_bh);
+        if (ret) {
+                mlog_errno(ret);
+                goto out_commit;
+        }
 
         /* Change from ocfs2_xattr_header to ocfs2_xattr_tree_root */
         memset(&xb->xb_attrs, 0, inode->i_sb->s_blocksize -
@@ -2941,8 +2891,8 @@ static int ocfs2_defrag_xattr_bucket(struct inode *inode,
 
         mlog(0, "adjust xattr bucket in %llu, count = %u, "
              "xh_free_start = %u, xh_name_value_len = %u.\n",
-             blkno, le16_to_cpu(xh->xh_count), xh_free_start,
-             le16_to_cpu(xh->xh_name_value_len));
+             (unsigned long long)blkno, le16_to_cpu(xh->xh_count),
+             xh_free_start, le16_to_cpu(xh->xh_name_value_len));
 
         /*
          * sort all the entries by their offset.
@@ -3058,7 +3008,7 @@ static int ocfs2_mv_xattr_bucket_cross_cluster(struct inode *inode,
         prev_blkno += (num_clusters - 1) * bpc + bpc / 2;
 
         mlog(0, "move half of xattrs in cluster %llu to %llu\n",
-             prev_blkno, new_blkno);
+             (unsigned long long)prev_blkno, (unsigned long long)new_blkno);
 
         /*
          * We need to update the 1st half of the new cluster and
@@ -3168,26 +3118,74 @@ static int ocfs2_read_xattr_bucket(struct inode *inode,
 }
 
 /*
- * Move half num of the xattrs in old bucket(blk) to new bucket(new_blk).
+ * Find the suitable pos when we divide a bucket into 2.
+ * We have to make sure the xattrs with the same hash value exist
+ * in the same bucket.
+ *
+ * If this ocfs2_xattr_header covers more than one hash value, find a
+ * place where the hash value changes.  Try to find the most even split.
+ * The most common case is that all entries have different hash values,
+ * and the first check we make will find a place to split.
+ */
+static int ocfs2_xattr_find_divide_pos(struct ocfs2_xattr_header *xh)
+{
+        struct ocfs2_xattr_entry *entries = xh->xh_entries;
+        int count = le16_to_cpu(xh->xh_count);
+        int delta, middle = count / 2;
+
+        /*
+         * We start at the middle.  Each step gets farther away in both
+         * directions.  We therefore hit the change in hash value
+         * nearest to the middle.  Note that this loop does not execute for
+         * count < 2.
+         */
+        for (delta = 0; delta < middle; delta++) {
+                /* Let's check delta earlier than middle */
+                if (cmp_xe(&entries[middle - delta - 1],
+                           &entries[middle - delta]))
+                        return middle - delta;
+
+                /* For even counts, don't walk off the end */
+                if ((middle + delta + 1) == count)
+                        continue;
+
+                /* Now try delta past middle */
+                if (cmp_xe(&entries[middle + delta],
+                           &entries[middle + delta + 1]))
+                        return middle + delta + 1;
+        }
+
+        /* Every entry had the same hash */
+        return count;
+}
+
+/*
+ * Move some xattrs in old bucket(blk) to new bucket(new_blk).
  * first_hash will record the 1st hash of the new bucket.
+ *
+ * Normally half of the xattrs will be moved.  But we have to make
+ * sure that the xattrs with the same hash value are stored in the
+ * same bucket. If all the xattrs in this bucket have the same hash
+ * value, the new bucket will be initialized as an empty one and the
+ * first_hash will be initialized as (hash_value+1).
  */
-static int ocfs2_half_xattr_bucket(struct inode *inode,
-                                   handle_t *handle,
-                                   u64 blk,
-                                   u64 new_blk,
-                                   u32 *first_hash,
-                                   int new_bucket_head)
+static int ocfs2_divide_xattr_bucket(struct inode *inode,
+                                    handle_t *handle,
+                                    u64 blk,
+                                    u64 new_blk,
+                                    u32 *first_hash,
+                                    int new_bucket_head)
 {
         int ret, i;
-        u16 count, start, len, name_value_len, xe_len, name_offset;
+        int count, start, len, name_value_len = 0, xe_len, name_offset = 0;
         u16 blk_per_bucket = ocfs2_blocks_per_xattr_bucket(inode->i_sb);
         struct buffer_head **s_bhs, **t_bhs = NULL;
         struct ocfs2_xattr_header *xh;
         struct ocfs2_xattr_entry *xe;
         int blocksize = inode->i_sb->s_blocksize;
 
-        mlog(0, "move half of xattrs from bucket %llu to %llu\n",
-             blk, new_blk);
+        mlog(0, "move some of xattrs from bucket %llu to %llu\n",
+             (unsigned long long)blk, (unsigned long long)new_blk);
 
         s_bhs = kcalloc(blk_per_bucket, sizeof(struct buffer_head *), GFP_NOFS);
         if (!s_bhs)
@@ -3220,21 +3218,44 @@ static int ocfs2_half_xattr_bucket(struct inode *inode,
 
         for (i = 0; i < blk_per_bucket; i++) {
                 ret = ocfs2_journal_access(handle, inode, t_bhs[i],
-                                           OCFS2_JOURNAL_ACCESS_CREATE);
+                                           new_bucket_head ?
+                                           OCFS2_JOURNAL_ACCESS_CREATE :
+                                           OCFS2_JOURNAL_ACCESS_WRITE);
                 if (ret) {
                         mlog_errno(ret);
                         goto out;
                 }
         }
 
+        xh = (struct ocfs2_xattr_header *)s_bhs[0]->b_data;
+        count = le16_to_cpu(xh->xh_count);
+        start = ocfs2_xattr_find_divide_pos(xh);
+
+        if (start == count) {
+                xe = &xh->xh_entries[start-1];
+
+                /*
+                 * initialized a new empty bucket here.
+                 * The hash value is set as one larger than
+                 * that of the last entry in the previous bucket.
+                 */
+                for (i = 0; i < blk_per_bucket; i++)
+                        memset(t_bhs[i]->b_data, 0, blocksize);
+
+                xh = (struct ocfs2_xattr_header *)t_bhs[0]->b_data;
+                xh->xh_free_start = cpu_to_le16(blocksize);
+                xh->xh_entries[0].xe_name_hash = xe->xe_name_hash;
+                le32_add_cpu(&xh->xh_entries[0].xe_name_hash, 1);
+
+                goto set_num_buckets;
+        }
+
         /* copy the whole bucket to the new first. */
         for (i = 0; i < blk_per_bucket; i++)
                 memcpy(t_bhs[i]->b_data, s_bhs[i]->b_data, blocksize);
 
         /* update the new bucket. */
         xh = (struct ocfs2_xattr_header *)t_bhs[0]->b_data;
-        count = le16_to_cpu(xh->xh_count);
-        start = count / 2;
 
         /*
          * Calculate the total name/value len and xh_free_start for
@@ -3291,6 +3312,7 @@ static int ocfs2_half_xattr_bucket(struct inode *inode,
                         xh->xh_free_start = xe->xe_name_offset;
         }
 
+set_num_buckets:
         /* set xh->xh_num_buckets for the new xh. */
         if (new_bucket_head)
                 xh->xh_num_buckets = cpu_to_le16(1);
@@ -3308,9 +3330,13 @@ static int ocfs2_half_xattr_bucket(struct inode *inode,
                 *first_hash = le32_to_cpu(xh->xh_entries[0].xe_name_hash);
 
         /*
-         * Now only update the 1st block of the old bucket.
-         * Please note that the entry has been sorted already above.
+         * Now only update the 1st block of the old bucket.  If we
+         * just added a new empty bucket, there is no need to modify
+         * it.
          */
+        if (start == count)
+                goto out;
+
         xh = (struct ocfs2_xattr_header *)s_bhs[0]->b_data;
         memset(&xh->xh_entries[start], 0,
                sizeof(struct ocfs2_xattr_entry) * (count - start));
@@ -3358,7 +3384,8 @@ static int ocfs2_cp_xattr_bucket(struct inode *inode,
         BUG_ON(s_blkno == t_blkno);
 
         mlog(0, "cp bucket %llu to %llu, target is %d\n",
-             s_blkno, t_blkno, t_is_new);
+             (unsigned long long)s_blkno, (unsigned long long)t_blkno,
+             t_is_new);
 
         s_bhs = kzalloc(sizeof(struct buffer_head *) * blk_per_bucket,
                         GFP_NOFS);
@@ -3382,6 +3409,8 @@ static int ocfs2_cp_xattr_bucket(struct inode *inode,
 
         for (i = 0; i < blk_per_bucket; i++) {
                 ret = ocfs2_journal_access(handle, inode, t_bhs[i],
+                                           t_is_new ?
+                                           OCFS2_JOURNAL_ACCESS_CREATE :
                                            OCFS2_JOURNAL_ACCESS_WRITE);
                 if (ret)
                         goto out;
@@ -3428,7 +3457,8 @@ static int ocfs2_cp_xattr_cluster(struct inode *inode,
         struct ocfs2_xattr_header *xh;
         u64 to_blk_start = to_blk;
 
-        mlog(0, "cp xattrs from cluster %llu to %llu\n", src_blk, to_blk);
+        mlog(0, "cp xattrs from cluster %llu to %llu\n",
+             (unsigned long long)src_blk, (unsigned long long)to_blk);
 
         /*
          * We need to update the new cluster and 1 more for the update of
@@ -3493,15 +3523,15 @@ out:
 }
 
 /*
- * Move half of the xattrs in this cluster to the new cluster.
+ * Move some xattrs in this cluster to the new cluster.
  * This function should only be called when bucket size == cluster size.
  * Otherwise ocfs2_mv_xattr_bucket_cross_cluster should be used instead.
  */
-static int ocfs2_half_xattr_cluster(struct inode *inode,
-                                    handle_t *handle,
-                                    u64 prev_blk,
-                                    u64 new_blk,
-                                    u32 *first_hash)
+static int ocfs2_divide_xattr_cluster(struct inode *inode,
+                                      handle_t *handle,
+                                      u64 prev_blk,
+                                      u64 new_blk,
+                                      u32 *first_hash)
 {
         u16 blk_per_bucket = ocfs2_blocks_per_xattr_bucket(inode->i_sb);
         int ret, credits = 2 * blk_per_bucket;
@@ -3515,8 +3545,8 @@ static int ocfs2_half_xattr_cluster(struct inode *inode,
         }
 
         /* Move half of the xattr in start_blk to the next bucket. */
-        return  ocfs2_half_xattr_bucket(inode, handle, prev_blk,
-                                        new_blk, first_hash, 1);
+        return  ocfs2_divide_xattr_bucket(inode, handle, prev_blk,
+                                          new_blk, first_hash, 1);
 }
 
 /*
@@ -3559,7 +3589,8 @@ static int ocfs2_adjust_xattr_cross_cluster(struct inode *inode,
         int bpc = ocfs2_clusters_to_blocks(inode->i_sb, 1);
 
         mlog(0, "adjust xattrs from cluster %llu len %u to %llu\n",
-             prev_blk, prev_clusters, new_blk);
+             (unsigned long long)prev_blk, prev_clusters,
+             (unsigned long long)new_blk);
 
         if (ocfs2_xattr_buckets_per_cluster(OCFS2_SB(inode->i_sb)) > 1)
                 ret = ocfs2_mv_xattr_bucket_cross_cluster(inode,
@@ -3578,9 +3609,9 @@ static int ocfs2_adjust_xattr_cross_cluster(struct inode *inode,
                                                      last_blk, new_blk,
                                                      v_start);
                 else {
-                        ret = ocfs2_half_xattr_cluster(inode, handle,
-                                                       last_blk, new_blk,
-                                                       v_start);
+                        ret = ocfs2_divide_xattr_cluster(inode, handle,
+                                                         last_blk, new_blk,
+                                                         v_start);
 
                         if ((*header_bh)->b_blocknr == last_blk && extend)
                                 *extend = 0;
@@ -3629,7 +3660,7 @@ static int ocfs2_add_new_xattr_cluster(struct inode *inode,
         mlog(0, "Add new xattr cluster for %llu, previous xattr hash = %u, "
              "previous xattr blkno = %llu\n",
              (unsigned long long)OCFS2_I(inode)->ip_blkno,
-             prev_cpos, prev_blkno);
+             prev_cpos, (unsigned long long)prev_blkno);
 
         ocfs2_init_xattr_tree_extent_tree(&et, inode, root_bh);
 
@@ -3716,7 +3747,7 @@ static int ocfs2_add_new_xattr_cluster(struct inode *inode,
                 }
         }
         mlog(0, "Insert %u clusters at block %llu for xattr at %u\n",
-             num_bits, block, v_start);
+             num_bits, (unsigned long long)block, v_start);
         ret = ocfs2_insert_extent(osb, handle, inode, &et, v_start, block,
                                   num_bits, 0, meta_ac);
         if (ret < 0) {
@@ -3761,7 +3792,7 @@ static int ocfs2_extend_xattr_bucket(struct inode *inode,
         u16 bucket = le16_to_cpu(first_xh->xh_num_buckets);
 
         mlog(0, "extend xattr bucket in %llu, xattr extend rec starting "
-             "from %llu, len = %u\n", start_blk,
+             "from %llu, len = %u\n", (unsigned long long)start_blk,
              (unsigned long long)first_bh->b_blocknr, num_clusters);
 
         BUG_ON(bucket >= num_buckets);
@@ -3797,8 +3828,8 @@ static int ocfs2_extend_xattr_bucket(struct inode *inode,
         }
 
         /* Move half of the xattr in start_blk to the next bucket. */
-        ret = ocfs2_half_xattr_bucket(inode, handle, start_blk,
-                                      start_blk + blk_per_bucket, NULL, 0);
+        ret = ocfs2_divide_xattr_bucket(inode, handle, start_blk,
+                                        start_blk + blk_per_bucket, NULL, 0);
 
         le16_add_cpu(&first_xh->xh_num_buckets, 1);
         ocfs2_journal_dirty(handle, first_bh);
@@ -4146,7 +4177,7 @@ static int ocfs2_xattr_value_update_size(struct inode *inode,
         handle_t *handle = NULL;
 
         handle = ocfs2_start_trans(osb, 1);
-        if (handle == NULL) {
+        if (IS_ERR(handle)) {
                 ret = -ENOMEM;
                 mlog_errno(ret);
                 goto out;
@@ -4313,7 +4344,7 @@ static int ocfs2_rm_xattr_cluster(struct inode *inode,
         }
 
         handle = ocfs2_start_trans(osb, OCFS2_REMOVE_EXTENT_CREDITS);
-        if (handle == NULL) {
+        if (IS_ERR(handle)) {
                 ret = -ENOMEM;
                 mlog_errno(ret);
                 goto out;
@@ -4489,11 +4520,21 @@ out:
         return ret;
 }
 
-/* check whether the xattr bucket is filled up with the same hash value. */
+/*
+ * check whether the xattr bucket is filled up with the same hash value.
+ * If we want to insert the xattr with the same hash, return -ENOSPC.
+ * If we want to insert a xattr with different hash value, go ahead
+ * and ocfs2_divide_xattr_bucket will handle this.
+ */
 static int ocfs2_check_xattr_bucket_collision(struct inode *inode,
-                                              struct ocfs2_xattr_bucket *bucket)
+                                              struct ocfs2_xattr_bucket *bucket,
+                                              const char *name)
 {
         struct ocfs2_xattr_header *xh = bucket->xh;
+        u32 name_hash = ocfs2_xattr_name_hash(inode, name, strlen(name));
+
+        if (name_hash != le32_to_cpu(xh->xh_entries[0].xe_name_hash))
+                return 0;
 
         if (xh->xh_entries[le16_to_cpu(xh->xh_count) - 1].xe_name_hash ==
             xh->xh_entries[0].xe_name_hash) {
@@ -4616,7 +4657,9 @@ try_again:
                  * one bucket's worth, so check it here whether we need to
                  * add a new bucket for the insert.
                  */
-                ret = ocfs2_check_xattr_bucket_collision(inode, &xs->bucket);
+                ret = ocfs2_check_xattr_bucket_collision(inode,
+                                                         &xs->bucket,
+                                                         xi->name);
                 if (ret) {
                         mlog_errno(ret);
                         goto out;
@@ -4727,14 +4770,11 @@ out:
 /*
  * 'trusted' attributes support
  */
-
-#define XATTR_TRUSTED_PREFIX "trusted."
-
 static size_t ocfs2_xattr_trusted_list(struct inode *inode, char *list,
                                        size_t list_size, const char *name,
                                        size_t name_len)
 {
-        const size_t prefix_len = sizeof(XATTR_TRUSTED_PREFIX) - 1;
+        const size_t prefix_len = XATTR_TRUSTED_PREFIX_LEN;
         const size_t total_len = prefix_len + name_len + 1;
 
         if (list && total_len <= list_size) {
@@ -4771,18 +4811,14 @@ struct xattr_handler ocfs2_xattr_trusted_handler = {
         .set    = ocfs2_xattr_trusted_set,
 };
 
-
 /*
  * 'user' attributes support
  */
-
-#define XATTR_USER_PREFIX "user."
-
 static size_t ocfs2_xattr_user_list(struct inode *inode, char *list,
                                     size_t list_size, const char *name,
                                     size_t name_len)
 {
-        const size_t prefix_len = sizeof(XATTR_USER_PREFIX) - 1;
+        const size_t prefix_len = XATTR_USER_PREFIX_LEN;
         const size_t total_len = prefix_len + name_len + 1;
         struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
 
diff --git a/fs/ocfs2/xattr.h b/fs/ocfs2/xattr.h
index c25c7c62a059..1d8314c7656d 100644
--- a/fs/ocfs2/xattr.h
+++ b/fs/ocfs2/xattr.h
@@ -3,24 +3,16 @@
  *
  * xattr.h
  *
- * Function prototypes
- *
- * Copyright (C) 2008 Oracle.  All rights reserved.
+ * Copyright (C) 2004, 2008 Oracle.  All rights reserved.
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public
- * License as published by the Free Software Foundation; either
- * version 2 of the License, or (at your option) any later version.
+ * License version 2 as published by the Free Software Foundation.
  *
  * This program 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 this program; if not, write to the
- * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- * Boston, MA 021110-1307, USA.
  */
 
 #ifndef OCFS2_XATTR_H
@@ -40,29 +32,11 @@ enum ocfs2_xattr_type {
 
 extern struct xattr_handler ocfs2_xattr_user_handler;
 extern struct xattr_handler ocfs2_xattr_trusted_handler;
-
-extern ssize_t ocfs2_listxattr(struct dentry *, char *, size_t);
-extern int ocfs2_xattr_get(struct inode *, int, const char *, void *, size_t);
-extern int ocfs2_xattr_set(struct inode *, int, const char *, const void *,
-                           size_t, int);
-extern int ocfs2_xattr_remove(struct inode *inode, struct buffer_head *di_bh);
 extern struct xattr_handler *ocfs2_xattr_handlers[];
 
-static inline u16 ocfs2_xattr_buckets_per_cluster(struct ocfs2_super *osb)
-{
-        return (1 << osb->s_clustersize_bits) / OCFS2_XATTR_BUCKET_SIZE;
-}
-
-static inline u16 ocfs2_blocks_per_xattr_bucket(struct super_block *sb)
-{
-        return OCFS2_XATTR_BUCKET_SIZE / (1 << sb->s_blocksize_bits);
-}
-
-static inline u16 ocfs2_xattr_max_xe_in_bucket(struct super_block *sb)
-{
-        u16 len = sb->s_blocksize -
-                 offsetof(struct ocfs2_xattr_header, xh_entries);
+ssize_t ocfs2_listxattr(struct dentry *, char *, size_t);
+int ocfs2_xattr_set(struct inode *, int, const char *, const void *,
+                    size_t, int);
+int ocfs2_xattr_remove(struct inode *, struct buffer_head *);
 
-        return len / sizeof(struct ocfs2_xattr_entry);
-}
 #endif /* OCFS2_XATTR_H */
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c
index 9e561a9cefca..a11a8390bf6c 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/xfs_da_btree.c
@@ -1566,11 +1566,14 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
         int nmap, error, w, count, c, got, i, mapi;
         xfs_trans_t *tp;
         xfs_mount_t *mp;
+        xfs_drfsbno_t   nblks;
 
         dp = args->dp;
         mp = dp->i_mount;
         w = args->whichfork;
         tp = args->trans;
+        nblks = dp->i_d.di_nblocks;
+
         /*
          * For new directories adjust the file offset and block count.
          */
@@ -1647,6 +1650,8 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
         }
         if (mapp != &map)
                 kmem_free(mapp);
+        /* account for newly allocated blocks in reserved blocks total */
+        args->total -= dp->i_d.di_nblocks - nblks;
         *new_blkno = (xfs_dablk_t)bno;
         return 0;
 }
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c
index 80e0dc51361c..1afb12278b8d 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/xfs_dir2.c
@@ -525,11 +525,13 @@ xfs_dir2_grow_inode(
         xfs_mount_t     *mp;
         int             nmap;           /* number of bmap entries */
         xfs_trans_t     *tp;
+        xfs_drfsbno_t   nblks;
 
         xfs_dir2_trace_args_s("grow_inode", args, space);
         dp = args->dp;
         tp = args->trans;
         mp = dp->i_mount;
+        nblks = dp->i_d.di_nblocks;
         /*
          * Set lowest possible block in the space requested.
          */
@@ -622,7 +624,11 @@ xfs_dir2_grow_inode(
          */
         if (mapp != &map)
                 kmem_free(mapp);
+
+        /* account for newly allocated blocks in reserved blocks total */
+        args->total -= dp->i_d.di_nblocks - nblks;
         *dbp = xfs_dir2_da_to_db(mp, (xfs_dablk_t)bno);
+
         /*
          * Update file's size if this is the data space and it grew.
          */
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index dbd9cef852ec..a391b955df01 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1414,7 +1414,7 @@ xfs_itruncate_start(
         mp = ip->i_mount;
 
         /* wait for the completion of any pending DIOs */
-        if (new_size < ip->i_size)
+        if (new_size == 0 || new_size < ip->i_size)
                 vn_iowait(ip);
 
         /*
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 0b02c6443551..3608a0f0a5f6 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -563,6 +563,11 @@ xfs_log_mount(
         }
 
         mp->m_log = xlog_alloc_log(mp, log_target, blk_offset, num_bblks);
+        if (!mp->m_log) {
+                cmn_err(CE_WARN, "XFS: Log allocation failed: No memory!");
+                error = ENOMEM;
+                goto out;
+        }
 
         /*
          * Initialize the AIL now we have a log.
@@ -601,6 +606,7 @@ xfs_log_mount(
         return 0;
 error:
         xfs_log_unmount_dealloc(mp);
+out:
         return error;
 }       /* xfs_log_mount */
 
@@ -1217,7 +1223,9 @@ xlog_alloc_log(xfs_mount_t	*mp,
         int                     i;
         int                     iclogsize;
 
-        log = (xlog_t *)kmem_zalloc(sizeof(xlog_t), KM_SLEEP);
+        log = kmem_zalloc(sizeof(xlog_t), KM_MAYFAIL);
+        if (!log)
+                return NULL;
 
         log->l_mp          = mp;
         log->l_targ        = log_target;
@@ -1249,6 +1257,8 @@ xlog_alloc_log(xfs_mount_t	*mp,
         xlog_get_iclog_buffer_size(mp, log);
 
         bp = xfs_buf_get_empty(log->l_iclog_size, mp->m_logdev_targp);
+        if (!bp)
+                goto out_free_log;
         XFS_BUF_SET_IODONE_FUNC(bp, xlog_iodone);
         XFS_BUF_SET_BDSTRAT_FUNC(bp, xlog_bdstrat_cb);
         XFS_BUF_SET_FSPRIVATE2(bp, (unsigned long)1);
@@ -1275,13 +1285,17 @@ xlog_alloc_log(xfs_mount_t	*mp,
         iclogsize = log->l_iclog_size;
         ASSERT(log->l_iclog_size >= 4096);
         for (i=0; i < log->l_iclog_bufs; i++) {
-                *iclogp = (xlog_in_core_t *)
-                          kmem_zalloc(sizeof(xlog_in_core_t), KM_SLEEP);
+                *iclogp = kmem_zalloc(sizeof(xlog_in_core_t), KM_MAYFAIL);
+                if (!*iclogp)
+                        goto out_free_iclog;
+
                 iclog = *iclogp;
                 iclog->ic_prev = prev_iclog;
                 prev_iclog = iclog;
 
                 bp = xfs_buf_get_noaddr(log->l_iclog_size, mp->m_logdev_targp);
+                if (!bp)
+                        goto out_free_iclog;
                 if (!XFS_BUF_CPSEMA(bp))
                         ASSERT(0);
                 XFS_BUF_SET_IODONE_FUNC(bp, xlog_iodone);
@@ -1323,6 +1337,25 @@ xlog_alloc_log(xfs_mount_t	*mp,
         log->l_iclog->ic_prev = prev_iclog;     /* re-write 1st prev ptr */
 
         return log;
+
+out_free_iclog:
+        for (iclog = log->l_iclog; iclog; iclog = prev_iclog) {
+                prev_iclog = iclog->ic_next;
+                if (iclog->ic_bp) {
+                        sv_destroy(&iclog->ic_force_wait);
+                        sv_destroy(&iclog->ic_write_wait);
+                        xfs_buf_free(iclog->ic_bp);
+                        xlog_trace_iclog_dealloc(iclog);
+                }
+                kmem_free(iclog);
+        }
+        spinlock_destroy(&log->l_icloglock);
+        spinlock_destroy(&log->l_grant_lock);
+        xlog_trace_loggrant_dealloc(log);
+        xfs_buf_free(log->l_xbuf);
+out_free_log:
+        kmem_free(log);
+        return NULL;
 }       /* xlog_alloc_log */
 
 
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 82d46ce69d5f..70e3ba32e6be 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -1419,7 +1419,13 @@ xlog_recover_add_to_trans(
                 return 0;
         item = trans->r_itemq;
         if (item == NULL) {
-                ASSERT(*(uint *)dp == XFS_TRANS_HEADER_MAGIC);
+                /* we need to catch log corruptions here */
+                if (*(uint *)dp != XFS_TRANS_HEADER_MAGIC) {
+                        xlog_warn("XFS: xlog_recover_add_to_trans: "
+                                  "bad header magic number");
+                        ASSERT(0);
+                        return XFS_ERROR(EIO);
+                }
                 if (len == sizeof(xfs_trans_header_t))
                         xlog_recover_add_item(&trans->r_itemq);
                 memcpy(&trans->r_theader, dp, len); /* d, s, l */
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index a4503f5e9497..15f5dd22fbb2 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -1245,6 +1245,9 @@ xfs_unmountfs(
 
         XFS_QM_DQPURGEALL(mp, XFS_QMOPT_QUOTALL | XFS_QMOPT_UMOUNTING);
 
+        if (mp->m_quotainfo)
+                XFS_QM_DONE(mp);
+
         /*
          * Flush out the log synchronously so that we know for sure
          * that nothing is pinned.  This is important because bflush()
@@ -1297,8 +1300,6 @@ xfs_unmountfs(
         xfs_errortag_clearall(mp, 0);
 #endif
         xfs_free_perag(mp);
-        if (mp->m_quotainfo)
-                XFS_QM_DONE(mp);
 }
 
 STATIC void