Merge branch 'master' into next

Conflicts: security/keys/internal.h security/keys/process_keys.c security/keys/request_key.c Fixed conflicts above by using the non 'tsk' versions. Signed-off-by: James Morris <jmorris@namei.org>
author: James Morris 2008-11-14 11:29:12 +1100
committer: James Morris 2008-11-14 11:29:12 +1100
commit: 2b828925652340277a889cbc11b2d0637f7cdaf7 (patch)
tree: 32fcb3d3e466fc419fad2d3717956a5b5ad3d35a /fs
parent: 3a3b7ce9336952ea7b9564d976d068a238976c9d (diff)
parent: 58e20d8d344b0ee083febb18c2b021d2427e56ca (diff)
45 files changed, 1155 insertions, 548 deletions
diff --git a/fs/Makefile b/fs/Makefile
index 2168c902d5ca..d9f8afe6f0c4 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -81,8 +81,6 @@ obj-$(CONFIG_HUGETLBFS)		+= hugetlbfs/
 obj-$(CONFIG_CODA_FS)		+= coda/
 obj-$(CONFIG_MINIX_FS)		+= minix/
 obj-$(CONFIG_FAT_FS)		+= fat/
-obj-$(CONFIG_MSDOS_FS)		+= msdos/
-obj-$(CONFIG_VFAT_FS)		+= vfat/
 obj-$(CONFIG_BFS_FS)		+= bfs/
 obj-$(CONFIG_ISO9660_FS)	+= isofs/
 obj-$(CONFIG_HFSPLUS_FS)	+= hfsplus/ # Before hfs to find wrapped HFS+
diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c
index ec16255d27dd..63b7c7afe8df 100644
--- a/fs/autofs4/dev-ioctl.c
+++ b/fs/autofs4/dev-ioctl.c
@@ -128,9 +128,10 @@ static inline void free_dev_ioctl(struct autofs_dev_ioctl *param)
  */
 static int validate_dev_ioctl(int cmd, struct autofs_dev_ioctl *param)
 {
-	int err = -EINVAL;
+	int err;
 
-	if (check_dev_ioctl_version(cmd, param)) {
+	err = check_dev_ioctl_version(cmd, param);
+	if (err) {
 		AUTOFS_WARN("invalid device control module version "
 		     "supplied for cmd(0x%08x)", cmd);
 		goto out;
diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index cde2f8e8935a..4b6fb3f628c0 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -56,12 +56,23 @@ static int autofs4_mount_busy(struct vfsmount *mnt, struct dentry *dentry)
 	mntget(mnt);
 	dget(dentry);
 
-	if (!autofs4_follow_mount(&mnt, &dentry))
+	if (!follow_down(&mnt, &dentry))
 		goto done;
 
-	/* This is an autofs submount, we can't expire it */
-	if (is_autofs4_dentry(dentry))
-		goto done;
+	if (is_autofs4_dentry(dentry)) {
+		struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+
+		/* This is an autofs submount, we can't expire it */
+		if (sbi->type == AUTOFS_TYPE_INDIRECT)
+			goto done;
+
+		/*
+		 * Otherwise it's an offset mount and we need to check
+		 * if we can umount its mount, if there is one.
+		 */
+		if (!d_mountpoint(dentry))
+			goto done;
+	}
 
 	/* Update the expiry counter if fs is busy */
 	if (!may_umount_tree(mnt)) {
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 88a776fa0ef6..db831efbdbbd 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -986,7 +986,6 @@ static int __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part);
 static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 {
 	struct gendisk *disk;
-	struct hd_struct *part = NULL;
 	int ret;
 	int partno;
 	int perm = 0;
@@ -1004,24 +1003,25 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 		return ret;
 	}
 
-	ret = -ENXIO;
-
 	lock_kernel();
 
+	ret = -ENXIO;
 	disk = get_gendisk(bdev->bd_dev, &partno);
 	if (!disk)
 		goto out_unlock_kernel;
-	part = disk_get_part(disk, partno);
-	if (!part)
-		goto out_unlock_kernel;
 
 	mutex_lock_nested(&bdev->bd_mutex, for_part);
 	if (!bdev->bd_openers) {
 		bdev->bd_disk = disk;
-		bdev->bd_part = part;
 		bdev->bd_contains = bdev;
 		if (!partno) {
 			struct backing_dev_info *bdi;
+
+			ret = -ENXIO;
+			bdev->bd_part = disk_get_part(disk, partno);
+			if (!bdev->bd_part)
+				goto out_clear;
+
 			if (disk->fops->open) {
 				ret = disk->fops->open(bdev, mode);
 				if (ret)
@@ -1049,18 +1049,17 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 			bdev->bd_contains = whole;
 			bdev->bd_inode->i_data.backing_dev_info =
 			   whole->bd_inode->i_data.backing_dev_info;
+			bdev->bd_part = disk_get_part(disk, partno);
 			if (!(disk->flags & GENHD_FL_UP) ||
-			    !part || !part->nr_sects) {
+			    !bdev->bd_part || !bdev->bd_part->nr_sects) {
 				ret = -ENXIO;
 				goto out_clear;
 			}
-			bd_set_size(bdev, (loff_t)part->nr_sects << 9);
+			bd_set_size(bdev, (loff_t)bdev->bd_part->nr_sects << 9);
 		}
 	} else {
-		disk_put_part(part);
 		put_disk(disk);
 		module_put(disk->fops->owner);
-		part = NULL;
 		disk = NULL;
 		if (bdev->bd_contains == bdev) {
 			if (bdev->bd_disk->fops->open) {
@@ -1080,6 +1079,7 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 	return 0;
 
  out_clear:
+	disk_put_part(bdev->bd_part);
 	bdev->bd_disk = NULL;
 	bdev->bd_part = NULL;
 	bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info;
@@ -1091,7 +1091,6 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
  out_unlock_kernel:
 	unlock_kernel();
 
-	disk_put_part(part);
 	if (disk)
 		module_put(disk->fops->owner);
 	put_disk(disk);
diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index d910501de6d2..8d86b7960f0d 100644
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -812,7 +812,7 @@ int dlm_release_lockspace(void *lockspace, int force)
 	error = release_lockspace(ls, force);
 	if (!error)
 		ls_count--;
-	else if (!ls_count)
+	if (!ls_count)
 		threads_stop();
 	mutex_unlock(&ls_lock);
 
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index e5717a4fae67..f6c94f232ec1 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -2375,12 +2375,9 @@ int ext3_force_commit(struct super_block *sb)
 /*
  * Ext3 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/ext3; eventual cleanup...)
  */
-
 static void ext3_write_super (struct super_block * sb)
 {
 	if (mutex_trylock(&sb->s_lock) != 0)
@@ -2390,13 +2387,12 @@ static void ext3_write_super (struct super_block * sb)
 
 static int ext3_sync_fs(struct super_block *sb, int wait)
 {
-	tid_t target;
-
 	sb->s_dirt = 0;
-	if (journal_start_commit(EXT3_SB(sb)->s_journal, &target)) {
-		if (wait)
-			log_wait_commit(EXT3_SB(sb)->s_journal, target);
-	}
+	if (wait)
+		ext3_force_commit(sb);
+	else
+		journal_start_commit(EXT3_SB(sb)->s_journal, NULL);
+
 	return 0;
 }
 
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index c8ea50ed023a..08cac9fcace2 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/fat/Makefile b/fs/fat/Makefile
index bfb5f06cf2c8..e06190322c1c 100644
--- a/fs/fat/Makefile
+++ b/fs/fat/Makefile
@@ -3,5 +3,9 @@
 #
 
 obj-$(CONFIG_FAT_FS) += fat.o
+obj-$(CONFIG_VFAT_FS) += vfat.o
+obj-$(CONFIG_MSDOS_FS) += msdos.o
 
-fat-objs := cache.o dir.o fatent.o file.o inode.o misc.o
+fat-y := cache.o dir.o fatent.o file.o inode.o misc.o
+vfat-y := namei_vfat.o
+msdos-y := namei_msdos.o
diff --git a/fs/fat/cache.c b/fs/fat/cache.c
index 3222f51c41cf..b42602298087 100644
--- a/fs/fat/cache.c
+++ b/fs/fat/cache.c
@@ -9,8 +9,8 @@
  */
 
 #include <linux/fs.h>
-#include <linux/msdos_fs.h>
 #include <linux/buffer_head.h>
+#include "fat.h"
 
 /* this must be > 0. */
 #define FAT_MAX_CACHE	8
@@ -293,10 +293,12 @@ static int fat_bmap_cluster(struct inode *inode, int cluster)
 }
 
 int fat_bmap(struct inode *inode, sector_t sector, sector_t *phys,
-	     unsigned long *mapped_blocks)
+	     unsigned long *mapped_blocks, int create)
 {
 	struct super_block *sb = inode->i_sb;
 	struct msdos_sb_info *sbi = MSDOS_SB(sb);
+	const unsigned long blocksize = sb->s_blocksize;
+	const unsigned char blocksize_bits = sb->s_blocksize_bits;
 	sector_t last_block;
 	int cluster, offset;
 
@@ -309,10 +311,21 @@ int fat_bmap(struct inode *inode, sector_t sector, sector_t *phys,
 		}
 		return 0;
 	}
-	last_block = (MSDOS_I(inode)->mmu_private + (sb->s_blocksize - 1))
-		>> sb->s_blocksize_bits;
-	if (sector >= last_block)
-		return 0;
+
+	last_block = (i_size_read(inode) + (blocksize - 1)) >> blocksize_bits;
+	if (sector >= last_block) {
+		if (!create)
+			return 0;
+
+		/*
+		 * ->mmu_private can access on only allocation path.
+		 * (caller must hold ->i_mutex)
+		 */
+		last_block = (MSDOS_I(inode)->mmu_private + (blocksize - 1))
+			>> blocksize_bits;
+		if (sector >= last_block)
+			return 0;
+	}
 
 	cluster = sector >> (sbi->cluster_bits - sb->s_blocksize_bits);
 	offset  = sector & (sbi->sec_per_clus - 1);
diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index bae1c3292522..67e058357098 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -16,11 +16,11 @@
 #include <linux/module.h>
 #include <linux/slab.h>
 #include <linux/time.h>
-#include <linux/msdos_fs.h>
 #include <linux/smp_lock.h>
 #include <linux/buffer_head.h>
 #include <linux/compat.h>
 #include <asm/uaccess.h>
+#include "fat.h"
 
 static inline loff_t fat_make_i_pos(struct super_block *sb,
 				    struct buffer_head *bh,
@@ -77,7 +77,7 @@ next:
 
 	*bh = NULL;
 	iblock = *pos >> sb->s_blocksize_bits;
-	err = fat_bmap(dir, iblock, &phys, &mapped_blocks);
+	err = fat_bmap(dir, iblock, &phys, &mapped_blocks, 0);
 	if (err || !phys)
 		return -1;	/* beyond EOF or error */
 
@@ -86,7 +86,7 @@ next:
 	*bh = sb_bread(sb, phys);
 	if (*bh == NULL) {
 		printk(KERN_ERR "FAT: Directory bread(block %llu) failed\n",
-		       (unsigned long long)phys);
+		       (llu)phys);
 		/* skip this block */
 		*pos = (iblock + 1) << sb->s_blocksize_bits;
 		goto next;
@@ -373,9 +373,10 @@ parse_record:
 		if (de->attr == ATTR_EXT) {
 			int status = fat_parse_long(inode, &cpos, &bh, &de,
 						    &unicode, &nr_slots);
-			if (status < 0)
-				return status;
-			else if (status == PARSE_INVALID)
+			if (status < 0) {
+				err = status;
+				goto end_of_dir;
+			} else if (status == PARSE_INVALID)
 				continue;
 			else if (status == PARSE_NOT_LONGNAME)
 				goto parse_record;
@@ -832,6 +833,7 @@ static long fat_compat_dir_ioctl(struct file *filp, unsigned cmd,
 #endif /* CONFIG_COMPAT */
 
 const struct file_operations fat_dir_operations = {
+	.llseek		= generic_file_llseek,
 	.read		= generic_read_dir,
 	.readdir	= fat_readdir,
 	.ioctl		= fat_dir_ioctl,
@@ -1089,6 +1091,7 @@ int fat_alloc_new_dir(struct inode *dir, struct timespec *ts)
 	struct msdos_dir_entry *de;
 	sector_t blknr;
 	__le16 date, time;
+	u8 time_cs;
 	int err, cluster;
 
 	err = fat_alloc_clusters(dir, &cluster, 1);
@@ -1102,7 +1105,7 @@ int fat_alloc_new_dir(struct inode *dir, struct timespec *ts)
 		goto error_free;
 	}
 
-	fat_date_unix2dos(ts->tv_sec, &time, &date, sbi->options.tz_utc);
+	fat_time_unix2fat(sbi, ts, &time, &date, &time_cs);
 
 	de = (struct msdos_dir_entry *)bhs[0]->b_data;
 	/* filling the new directory slots ("." and ".." entries) */
@@ -1112,13 +1115,14 @@ int fat_alloc_new_dir(struct inode *dir, struct timespec *ts)
 	de[0].lcase = de[1].lcase = 0;
 	de[0].time = de[1].time = time;
 	de[0].date = de[1].date = date;
-	de[0].ctime_cs = de[1].ctime_cs = 0;
 	if (sbi->options.isvfat) {
 		/* extra timestamps */
 		de[0].ctime = de[1].ctime = time;
+		de[0].ctime_cs = de[1].ctime_cs = time_cs;
 		de[0].adate = de[0].cdate = de[1].adate = de[1].cdate = date;
 	} else {
 		de[0].ctime = de[1].ctime = 0;
+		de[0].ctime_cs = de[1].ctime_cs = 0;
 		de[0].adate = de[0].cdate = de[1].adate = de[1].cdate = 0;
 	}
 	de[0].start = cpu_to_le16(cluster);
diff --git a/fs/fat/fat.h b/fs/fat/fat.h
new file mode 100644
index 000000000000..ea440d65819c
--- /dev/null
+++ b/fs/fat/fat.h
@@ -0,0 +1,329 @@
+#ifndef _FAT_H
+#define _FAT_H
+
+#include <linux/buffer_head.h>
+#include <linux/string.h>
+#include <linux/nls.h>
+#include <linux/fs.h>
+#include <linux/mutex.h>
+#include <linux/msdos_fs.h>
+
+/*
+ * vfat shortname flags
+ */
+#define VFAT_SFN_DISPLAY_LOWER	0x0001 /* convert to lowercase for display */
+#define VFAT_SFN_DISPLAY_WIN95	0x0002 /* emulate win95 rule for display */
+#define VFAT_SFN_DISPLAY_WINNT	0x0004 /* emulate winnt rule for display */
+#define VFAT_SFN_CREATE_WIN95	0x0100 /* emulate win95 rule for create */
+#define VFAT_SFN_CREATE_WINNT	0x0200 /* emulate winnt rule for create */
+
+struct fat_mount_options {
+	uid_t fs_uid;
+	gid_t fs_gid;
+	unsigned short fs_fmask;
+	unsigned short fs_dmask;
+	unsigned short codepage;  /* Codepage for shortname conversions */
+	char *iocharset;          /* Charset used for filename input/display */
+	unsigned short shortname; /* flags for shortname display/create rule */
+	unsigned char name_check; /* r = relaxed, n = normal, s = strict */
+	unsigned short allow_utime;/* permission for setting the [am]time */
+	unsigned quiet:1,         /* set = fake successful chmods and chowns */
+		 showexec:1,      /* set = only set x bit for com/exe/bat */
+		 sys_immutable:1, /* set = system files are immutable */
+		 dotsOK:1,        /* set = hidden and system files are named '.filename' */
+		 isvfat:1,        /* 0=no vfat long filename support, 1=vfat support */
+		 utf8:1,	  /* Use of UTF-8 character set (Default) */
+		 unicode_xlate:1, /* create escape sequences for unhandled Unicode */
+		 numtail:1,       /* Does first alias have a numeric '~1' type tail? */
+		 flush:1,	  /* write things quickly */
+		 nocase:1,	  /* Does this need case conversion? 0=need case conversion*/
+		 usefree:1,	  /* Use free_clusters for FAT32 */
+		 tz_utc:1,	  /* Filesystem timestamps are in UTC */
+		 rodir:1;	  /* allow ATTR_RO for directory */
+};
+
+#define FAT_HASH_BITS	8
+#define FAT_HASH_SIZE	(1UL << FAT_HASH_BITS)
+
+/*
+ * MS-DOS file system in-core superblock data
+ */
+struct msdos_sb_info {
+	unsigned short sec_per_clus; /* sectors/cluster */
+	unsigned short cluster_bits; /* log2(cluster_size) */
+	unsigned int cluster_size;   /* cluster size */
+	unsigned char fats,fat_bits; /* number of FATs, FAT bits (12 or 16) */
+	unsigned short fat_start;
+	unsigned long fat_length;    /* FAT start & length (sec.) */
+	unsigned long dir_start;
+	unsigned short dir_entries;  /* root dir start & entries */
+	unsigned long data_start;    /* first data sector */
+	unsigned long max_cluster;   /* maximum cluster number */
+	unsigned long root_cluster;  /* first cluster of the root directory */
+	unsigned long fsinfo_sector; /* sector number of FAT32 fsinfo */
+	struct mutex fat_lock;
+	unsigned int prev_free;      /* previously allocated cluster number */
+	unsigned int free_clusters;  /* -1 if undefined */
+	unsigned int free_clus_valid; /* is free_clusters valid? */
+	struct fat_mount_options options;
+	struct nls_table *nls_disk;  /* Codepage used on disk */
+	struct nls_table *nls_io;    /* Charset used for input and display */
+	const void *dir_ops;		     /* Opaque; default directory operations */
+	int dir_per_block;	     /* dir entries per block */
+	int dir_per_block_bits;	     /* log2(dir_per_block) */
+
+	int fatent_shift;
+	struct fatent_operations *fatent_ops;
+
+	spinlock_t inode_hash_lock;
+	struct hlist_head inode_hashtable[FAT_HASH_SIZE];
+};
+
+#define FAT_CACHE_VALID	0	/* special case for valid cache */
+
+/*
+ * MS-DOS file system inode data in memory
+ */
+struct msdos_inode_info {
+	spinlock_t cache_lru_lock;
+	struct list_head cache_lru;
+	int nr_caches;
+	/* for avoiding the race between fat_free() and fat_get_cluster() */
+	unsigned int cache_valid_id;
+
+	/* NOTE: mmu_private is 64bits, so must hold ->i_mutex to access */
+	loff_t mmu_private;	/* physically allocated size */
+
+	int i_start;		/* first cluster or 0 */
+	int i_logstart;		/* logical first cluster */
+	int i_attrs;		/* unused attribute bits */
+	loff_t i_pos;		/* on-disk position of directory entry or 0 */
+	struct hlist_node i_fat_hash;	/* hash by i_location */
+	struct inode vfs_inode;
+};
+
+struct fat_slot_info {
+	loff_t i_pos;		/* on-disk position of directory entry */
+	loff_t slot_off;	/* offset for slot or de start */
+	int nr_slots;		/* number of slots + 1(de) in filename */
+	struct msdos_dir_entry *de;
+	struct buffer_head *bh;
+};
+
+static inline struct msdos_sb_info *MSDOS_SB(struct super_block *sb)
+{
+	return sb->s_fs_info;
+}
+
+static inline struct msdos_inode_info *MSDOS_I(struct inode *inode)
+{
+	return container_of(inode, struct msdos_inode_info, vfs_inode);
+}
+
+/*
+ * If ->i_mode can't hold S_IWUGO (i.e. ATTR_RO), we use ->i_attrs to
+ * save ATTR_RO instead of ->i_mode.
+ *
+ * If it's directory and !sbi->options.rodir, ATTR_RO isn't read-only
+ * bit, it's just used as flag for app.
+ */
+static inline int fat_mode_can_hold_ro(struct inode *inode)
+{
+	struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb);
+	mode_t mask;
+
+	if (S_ISDIR(inode->i_mode)) {
+		if (!sbi->options.rodir)
+			return 0;
+		mask = ~sbi->options.fs_dmask;
+	} else
+		mask = ~sbi->options.fs_fmask;
+
+	if (!(mask & S_IWUGO))
+		return 0;
+	return 1;
+}
+
+/* Convert attribute bits and a mask to the UNIX mode. */
+static inline mode_t fat_make_mode(struct msdos_sb_info *sbi,
+				   u8 attrs, mode_t mode)
+{
+	if (attrs & ATTR_RO && !((attrs & ATTR_DIR) && !sbi->options.rodir))
+		mode &= ~S_IWUGO;
+
+	if (attrs & ATTR_DIR)
+		return (mode & ~sbi->options.fs_dmask) | S_IFDIR;
+	else
+		return (mode & ~sbi->options.fs_fmask) | S_IFREG;
+}
+
+/* Return the FAT attribute byte for this inode */
+static inline u8 fat_make_attrs(struct inode *inode)
+{
+	u8 attrs = MSDOS_I(inode)->i_attrs;
+	if (S_ISDIR(inode->i_mode))
+		attrs |= ATTR_DIR;
+	if (fat_mode_can_hold_ro(inode) && !(inode->i_mode & S_IWUGO))
+		attrs |= ATTR_RO;
+	return attrs;
+}
+
+static inline void fat_save_attrs(struct inode *inode, u8 attrs)
+{
+	if (fat_mode_can_hold_ro(inode))
+		MSDOS_I(inode)->i_attrs = attrs & ATTR_UNUSED;
+	else
+		MSDOS_I(inode)->i_attrs = attrs & (ATTR_UNUSED | ATTR_RO);
+}
+
+static inline unsigned char fat_checksum(const __u8 *name)
+{
+	unsigned char s = name[0];
+	s = (s<<7) + (s>>1) + name[1];	s = (s<<7) + (s>>1) + name[2];
+	s = (s<<7) + (s>>1) + name[3];	s = (s<<7) + (s>>1) + name[4];
+	s = (s<<7) + (s>>1) + name[5];	s = (s<<7) + (s>>1) + name[6];
+	s = (s<<7) + (s>>1) + name[7];	s = (s<<7) + (s>>1) + name[8];
+	s = (s<<7) + (s>>1) + name[9];	s = (s<<7) + (s>>1) + name[10];
+	return s;
+}
+
+static inline sector_t fat_clus_to_blknr(struct msdos_sb_info *sbi, int clus)
+{
+	return ((sector_t)clus - FAT_START_ENT) * sbi->sec_per_clus
+		+ sbi->data_start;
+}
+
+static inline void fat16_towchar(wchar_t *dst, const __u8 *src, size_t len)
+{
+#ifdef __BIG_ENDIAN
+	while (len--) {
+		*dst++ = src[0] | (src[1] << 8);
+		src += 2;
+	}
+#else
+	memcpy(dst, src, len * 2);
+#endif
+}
+
+static inline void fatwchar_to16(__u8 *dst, const wchar_t *src, size_t len)
+{
+#ifdef __BIG_ENDIAN
+	while (len--) {
+		dst[0] = *src & 0x00FF;
+		dst[1] = (*src & 0xFF00) >> 8;
+		dst += 2;
+		src++;
+	}
+#else
+	memcpy(dst, src, len * 2);
+#endif
+}
+
+/* fat/cache.c */
+extern void fat_cache_inval_inode(struct inode *inode);
+extern int fat_get_cluster(struct inode *inode, int cluster,
+			   int *fclus, int *dclus);
+extern int fat_bmap(struct inode *inode, sector_t sector, sector_t *phys,
+		    unsigned long *mapped_blocks, int create);
+
+/* fat/dir.c */
+extern const struct file_operations fat_dir_operations;
+extern int fat_search_long(struct inode *inode, const unsigned char *name,
+			   int name_len, struct fat_slot_info *sinfo);
+extern int fat_dir_empty(struct inode *dir);
+extern int fat_subdirs(struct inode *dir);
+extern int fat_scan(struct inode *dir, const unsigned char *name,
+		    struct fat_slot_info *sinfo);
+extern int fat_get_dotdot_entry(struct inode *dir, struct buffer_head **bh,
+				struct msdos_dir_entry **de, loff_t *i_pos);
+extern int fat_alloc_new_dir(struct inode *dir, struct timespec *ts);
+extern int fat_add_entries(struct inode *dir, void *slots, int nr_slots,
+			   struct fat_slot_info *sinfo);
+extern int fat_remove_entries(struct inode *dir, struct fat_slot_info *sinfo);
+
+/* fat/fatent.c */
+struct fat_entry {
+	int entry;
+	union {
+		u8 *ent12_p[2];
+		__le16 *ent16_p;
+		__le32 *ent32_p;
+	} u;
+	int nr_bhs;
+	struct buffer_head *bhs[2];
+};
+
+static inline void fatent_init(struct fat_entry *fatent)
+{
+	fatent->nr_bhs = 0;
+	fatent->entry = 0;
+	fatent->u.ent32_p = NULL;
+	fatent->bhs[0] = fatent->bhs[1] = NULL;
+}
+
+static inline void fatent_set_entry(struct fat_entry *fatent, int entry)
+{
+	fatent->entry = entry;
+	fatent->u.ent32_p = NULL;
+}
+
+static inline void fatent_brelse(struct fat_entry *fatent)
+{
+	int i;
+	fatent->u.ent32_p = NULL;
+	for (i = 0; i < fatent->nr_bhs; i++)
+		brelse(fatent->bhs[i]);
+	fatent->nr_bhs = 0;
+	fatent->bhs[0] = fatent->bhs[1] = NULL;
+}
+
+extern void fat_ent_access_init(struct super_block *sb);
+extern int fat_ent_read(struct inode *inode, struct fat_entry *fatent,
+			int entry);
+extern int fat_ent_write(struct inode *inode, struct fat_entry *fatent,
+			 int new, int wait);
+extern int fat_alloc_clusters(struct inode *inode, int *cluster,
+			      int nr_cluster);
+extern int fat_free_clusters(struct inode *inode, int cluster);
+extern int fat_count_free_clusters(struct super_block *sb);
+
+/* fat/file.c */
+extern int fat_generic_ioctl(struct inode *inode, struct file *filp,
+			     unsigned int cmd, unsigned long arg);
+extern const struct file_operations fat_file_operations;
+extern const struct inode_operations fat_file_inode_operations;
+extern int fat_setattr(struct dentry * dentry, struct iattr * attr);
+extern void fat_truncate(struct inode *inode);
+extern int fat_getattr(struct vfsmount *mnt, struct dentry *dentry,
+		       struct kstat *stat);
+
+/* fat/inode.c */
+extern void fat_attach(struct inode *inode, loff_t i_pos);
+extern void fat_detach(struct inode *inode);
+extern struct inode *fat_iget(struct super_block *sb, loff_t i_pos);
+extern struct inode *fat_build_inode(struct super_block *sb,
+			struct msdos_dir_entry *de, loff_t i_pos);
+extern int fat_sync_inode(struct inode *inode);
+extern int fat_fill_super(struct super_block *sb, void *data, int silent,
+			const struct inode_operations *fs_dir_inode_ops, int isvfat);
+
+extern int fat_flush_inodes(struct super_block *sb, struct inode *i1,
+		            struct inode *i2);
+/* fat/misc.c */
+extern void fat_fs_panic(struct super_block *s, const char *fmt, ...)
+	__attribute__ ((format (printf, 2, 3))) __cold;
+extern void fat_clusters_flush(struct super_block *sb);
+extern int fat_chain_add(struct inode *inode, int new_dclus, int nr_cluster);
+extern void fat_time_fat2unix(struct msdos_sb_info *sbi, struct timespec *ts,
+			      __le16 __time, __le16 __date, u8 time_cs);
+extern void fat_time_unix2fat(struct msdos_sb_info *sbi, struct timespec *ts,
+			      __le16 *time, __le16 *date, u8 *time_cs);
+extern int fat_sync_bhs(struct buffer_head **bhs, int nr_bhs);
+
+int fat_cache_init(void);
+void fat_cache_destroy(void);
+
+/* helper for printk */
+typedef unsigned long long	llu;
+
+#endif /* !_FAT_H */
diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c
index fb98b3d847ed..da6eea47872f 100644
--- a/fs/fat/fatent.c
+++ b/fs/fat/fatent.c
@@ -7,6 +7,7 @@
 #include <linux/fs.h>
 #include <linux/msdos_fs.h>
 #include <linux/blkdev.h>
+#include "fat.h"
 
 struct fatent_operations {
 	void (*ent_blocknr)(struct super_block *, int, int *, sector_t *);
@@ -92,8 +93,7 @@ static int fat12_ent_bread(struct super_block *sb, struct fat_entry *fatent,
 err_brelse:
 	brelse(bhs[0]);
 err:
-	printk(KERN_ERR "FAT: FAT read failed (blocknr %llu)\n",
-	       (unsigned long long)blocknr);
+	printk(KERN_ERR "FAT: FAT read failed (blocknr %llu)\n", (llu)blocknr);
 	return -EIO;
 }
 
@@ -106,7 +106,7 @@ static int fat_ent_bread(struct super_block *sb, struct fat_entry *fatent,
 	fatent->bhs[0] = sb_bread(sb, blocknr);
 	if (!fatent->bhs[0]) {
 		printk(KERN_ERR "FAT: FAT read failed (blocknr %llu)\n",
-		       (unsigned long long)blocknr);
+		       (llu)blocknr);
 		return -EIO;
 	}
 	fatent->nr_bhs = 1;
@@ -316,10 +316,20 @@ static inline int fat_ent_update_ptr(struct super_block *sb,
 	/* Is this fatent's blocks including this entry? */
 	if (!fatent->nr_bhs || bhs[0]->b_blocknr != blocknr)
 		return 0;
-	/* Does this entry need the next block? */
-	if (sbi->fat_bits == 12 && (offset + 1) >= sb->s_blocksize) {
-		if (fatent->nr_bhs != 2 || bhs[1]->b_blocknr != (blocknr + 1))
-			return 0;
+	if (sbi->fat_bits == 12) {
+		if ((offset + 1) < sb->s_blocksize) {
+			/* This entry is on bhs[0]. */
+			if (fatent->nr_bhs == 2) {
+				brelse(bhs[1]);
+				fatent->nr_bhs = 1;
+			}
+		} else {
+			/* This entry needs the next block. */
+			if (fatent->nr_bhs != 2)
+				return 0;
+			if (bhs[1]->b_blocknr != (blocknr + 1))
+				return 0;
+		}
 	}
 	ops->ent_set_ptr(fatent, offset);
 	return 1;
diff --git a/fs/fat/file.c b/fs/fat/file.c
index 81e203288340..0a7f4a9918b3 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -10,13 +10,13 @@
 #include <linux/module.h>
 #include <linux/mount.h>
 #include <linux/time.h>
-#include <linux/msdos_fs.h>
 #include <linux/buffer_head.h>
 #include <linux/writeback.h>
 #include <linux/backing-dev.h>
 #include <linux/blkdev.h>
 #include <linux/fsnotify.h>
 #include <linux/security.h>
+#include "fat.h"
 
 int fat_generic_ioctl(struct inode *inode, struct file *filp,
 		      unsigned int cmd, unsigned long arg)
@@ -29,10 +29,9 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp,
 	{
 		u32 attr;
 
-		if (inode->i_ino == MSDOS_ROOT_INO)
-			attr = ATTR_DIR;
-		else
-			attr = fat_attr(inode);
+		mutex_lock(&inode->i_mutex);
+		attr = fat_make_attrs(inode);
+		mutex_unlock(&inode->i_mutex);
 
 		return put_user(attr, user_attr);
 	}
@@ -62,20 +61,16 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp,
 		/* Merge in ATTR_VOLUME and ATTR_DIR */
 		attr |= (MSDOS_I(inode)->i_attrs & ATTR_VOLUME) |
 			(is_dir ? ATTR_DIR : 0);
-		oldattr = fat_attr(inode);
+		oldattr = fat_make_attrs(inode);
 
 		/* Equivalent to a chmod() */
 		ia.ia_valid = ATTR_MODE | ATTR_CTIME;
 		ia.ia_ctime = current_fs_time(inode->i_sb);
-		if (is_dir) {
-			ia.ia_mode = MSDOS_MKMODE(attr,
-				S_IRWXUGO & ~sbi->options.fs_dmask)
-				| S_IFDIR;
-		} else {
-			ia.ia_mode = MSDOS_MKMODE(attr,
-				(S_IRUGO | S_IWUGO | (inode->i_mode & S_IXUGO))
-				& ~sbi->options.fs_fmask)
-				| S_IFREG;
+		if (is_dir)
+			ia.ia_mode = fat_make_mode(sbi, attr, S_IRWXUGO);
+		else {
+			ia.ia_mode = fat_make_mode(sbi, attr,
+				S_IRUGO | S_IWUGO | (inode->i_mode & S_IXUGO));
 		}
 
 		/* The root directory has no attributes */
@@ -115,7 +110,7 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp,
 				inode->i_flags &= S_IMMUTABLE;
 		}
 
-		MSDOS_I(inode)->i_attrs = attr & ATTR_UNUSED;
+		fat_save_attrs(inode, attr);
 		mark_inode_dirty(inode);
 up:
 		mnt_drop_write(filp->f_path.mnt);
@@ -274,7 +269,7 @@ static int fat_sanitize_mode(const struct msdos_sb_info *sbi,
 
 	/*
 	 * Note, the basic check is already done by a caller of
-	 * (attr->ia_mode & ~MSDOS_VALID_MODE)
+	 * (attr->ia_mode & ~FAT_VALID_MODE)
 	 */
 
 	if (S_ISREG(inode->i_mode))
@@ -287,11 +282,18 @@ static int fat_sanitize_mode(const struct msdos_sb_info *sbi,
 	/*
 	 * Of the r and x bits, all (subject to umask) must be present. Of the
 	 * w bits, either all (subject to umask) or none must be present.
+	 *
+	 * If fat_mode_can_hold_ro(inode) is false, can't change w bits.
 	 */
 	if ((perm & (S_IRUGO | S_IXUGO)) != (inode->i_mode & (S_IRUGO|S_IXUGO)))
 		return -EPERM;
-	if ((perm & S_IWUGO) && ((perm & S_IWUGO) != (S_IWUGO & ~mask)))
-		return -EPERM;
+	if (fat_mode_can_hold_ro(inode)) {
+		if ((perm & S_IWUGO) && ((perm & S_IWUGO) != (S_IWUGO & ~mask)))
+			return -EPERM;
+	} else {
+		if ((perm & S_IWUGO) != (S_IWUGO & ~mask))
+			return -EPERM;
+	}
 
 	*mode_ptr &= S_IFMT | perm;
 
@@ -314,13 +316,15 @@ static int fat_allow_set_time(struct msdos_sb_info *sbi, struct inode *inode)
 }
 
 #define TIMES_SET_FLAGS	(ATTR_MTIME_SET | ATTR_ATIME_SET | ATTR_TIMES_SET)
+/* valid file mode bits */
+#define FAT_VALID_MODE	(S_IFREG | S_IFDIR | S_IRWXUGO)
 
 int fat_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct msdos_sb_info *sbi = MSDOS_SB(dentry->d_sb);
 	struct inode *inode = dentry->d_inode;
-	int error = 0;
 	unsigned int ia_valid;
+	int error;
 
 	/*
 	 * Expand the file. Since inode_setattr() updates ->i_size
@@ -356,7 +360,7 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr)
 	    ((attr->ia_valid & ATTR_GID) &&
 	     (attr->ia_gid != sbi->options.fs_gid)) ||
 	    ((attr->ia_valid & ATTR_MODE) &&
-	     (attr->ia_mode & ~MSDOS_VALID_MODE)))
+	     (attr->ia_mode & ~FAT_VALID_MODE)))
 		error = -EPERM;
 
 	if (error) {
@@ -374,7 +378,8 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr)
 			attr->ia_valid &= ~ATTR_MODE;
 	}
 
-	error = inode_setattr(inode, attr);
+	if (attr->ia_valid)
+		error = inode_setattr(inode, attr);
 out:
 	return error;
 }
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index cf621acd9e9a..d937aaf77374 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -16,7 +16,6 @@
 #include <linux/slab.h>
 #include <linux/smp_lock.h>
 #include <linux/seq_file.h>
-#include <linux/msdos_fs.h>
 #include <linux/pagemap.h>
 #include <linux/mpage.h>
 #include <linux/buffer_head.h>
@@ -27,7 +26,9 @@
 #include <linux/uio.h>
 #include <linux/writeback.h>
 #include <linux/log2.h>
+#include <linux/hash.h>
 #include <asm/unaligned.h>
+#include "fat.h"
 
 #ifndef CONFIG_FAT_DEFAULT_IOCHARSET
 /* if user don't select VFAT, this is undefined. */
@@ -63,7 +64,7 @@ static inline int __fat_get_block(struct inode *inode, sector_t iblock,
 	sector_t phys;
 	int err, offset;
 
-	err = fat_bmap(inode, iblock, &phys, &mapped_blocks);
+	err = fat_bmap(inode, iblock, &phys, &mapped_blocks, create);
 	if (err)
 		return err;
 	if (phys) {
@@ -93,7 +94,7 @@ static inline int __fat_get_block(struct inode *inode, sector_t iblock,
 	*max_blocks = min(mapped_blocks, *max_blocks);
 	MSDOS_I(inode)->mmu_private += *max_blocks << sb->s_blocksize_bits;
 
-	err = fat_bmap(inode, iblock, &phys, &mapped_blocks);
+	err = fat_bmap(inode, iblock, &phys, &mapped_blocks, create);
 	if (err)
 		return err;
 
@@ -198,7 +199,14 @@ static ssize_t fat_direct_IO(int rw, struct kiocb *iocb,
 
 static sector_t _fat_bmap(struct address_space *mapping, sector_t block)
 {
-	return generic_block_bmap(mapping, block, fat_get_block);
+	sector_t blocknr;
+
+	/* fat_get_cluster() assumes the requested blocknr isn't truncated. */
+	mutex_lock(&mapping->host->i_mutex);
+	blocknr = generic_block_bmap(mapping, block, fat_get_block);
+	mutex_unlock(&mapping->host->i_mutex);
+
+	return blocknr;
 }
 
 static const struct address_space_operations fat_aops = {
@@ -247,25 +255,21 @@ static void fat_hash_init(struct super_block *sb)
 		INIT_HLIST_HEAD(&sbi->inode_hashtable[i]);
 }
 
-static inline unsigned long fat_hash(struct super_block *sb, loff_t i_pos)
+static inline unsigned long fat_hash(loff_t i_pos)
 {
-	unsigned long tmp = (unsigned long)i_pos | (unsigned long) sb;
-	tmp = tmp + (tmp >> FAT_HASH_BITS) + (tmp >> FAT_HASH_BITS * 2);
-	return tmp & FAT_HASH_MASK;
+	return hash_32(i_pos, FAT_HASH_BITS);
 }
 
 void fat_attach(struct inode *inode, loff_t i_pos)
 {
-	struct super_block *sb = inode->i_sb;
-	struct msdos_sb_info *sbi = MSDOS_SB(sb);
+	struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb);
+	struct hlist_head *head = sbi->inode_hashtable + fat_hash(i_pos);
 
 	spin_lock(&sbi->inode_hash_lock);
 	MSDOS_I(inode)->i_pos = i_pos;
-	hlist_add_head(&MSDOS_I(inode)->i_fat_hash,
-			sbi->inode_hashtable + fat_hash(sb, i_pos));
+	hlist_add_head(&MSDOS_I(inode)->i_fat_hash, head);
 	spin_unlock(&sbi->inode_hash_lock);
 }
-
 EXPORT_SYMBOL_GPL(fat_attach);
 
 void fat_detach(struct inode *inode)
@@ -276,13 +280,12 @@ void fat_detach(struct inode *inode)
 	hlist_del_init(&MSDOS_I(inode)->i_fat_hash);
 	spin_unlock(&sbi->inode_hash_lock);
 }
-
 EXPORT_SYMBOL_GPL(fat_detach);
 
 struct inode *fat_iget(struct super_block *sb, loff_t i_pos)
 {
 	struct msdos_sb_info *sbi = MSDOS_SB(sb);
-	struct hlist_head *head = sbi->inode_hashtable + fat_hash(sb, i_pos);
+	struct hlist_head *head = sbi->inode_hashtable + fat_hash(i_pos);
 	struct hlist_node *_p;
 	struct msdos_inode_info *i;
 	struct inode *inode = NULL;
@@ -341,8 +344,7 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 
 	if ((de->attr & ATTR_DIR) && !IS_FREE(de->name)) {
 		inode->i_generation &= ~1;
-		inode->i_mode = MSDOS_MKMODE(de->attr,
-			S_IRWXUGO & ~sbi->options.fs_dmask) | S_IFDIR;
+		inode->i_mode = fat_make_mode(sbi, de->attr, S_IRWXUGO);
 		inode->i_op = sbi->dir_ops;
 		inode->i_fop = &fat_dir_operations;
 
@@ -359,10 +361,9 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 		inode->i_nlink = fat_subdirs(inode);
 	} else { /* not a directory */
 		inode->i_generation |= 1;
-		inode->i_mode = MSDOS_MKMODE(de->attr,
-		    ((sbi->options.showexec && !is_exec(de->name + 8))
-			? S_IRUGO|S_IWUGO : S_IRWXUGO)
-		    & ~sbi->options.fs_fmask) | S_IFREG;
+		inode->i_mode = fat_make_mode(sbi, de->attr,
+			((sbi->options.showexec && !is_exec(de->name + 8))
+			 ? S_IRUGO|S_IWUGO : S_IRWXUGO));
 		MSDOS_I(inode)->i_start = le16_to_cpu(de->start);
 		if (sbi->fat_bits == 32)
 			MSDOS_I(inode)->i_start |= (le16_to_cpu(de->starthi) << 16);
@@ -378,25 +379,16 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 		if (sbi->options.sys_immutable)
 			inode->i_flags |= S_IMMUTABLE;
 	}
-	MSDOS_I(inode)->i_attrs = de->attr & ATTR_UNUSED;
+	fat_save_attrs(inode, de->attr);
+
 	inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1))
 			   & ~((loff_t)sbi->cluster_size - 1)) >> 9;
-	inode->i_mtime.tv_sec =
-		date_dos2unix(le16_to_cpu(de->time), le16_to_cpu(de->date),
-			      sbi->options.tz_utc);
-	inode->i_mtime.tv_nsec = 0;
+
+	fat_time_fat2unix(sbi, &inode->i_mtime, de->time, de->date, 0);
 	if (sbi->options.isvfat) {
-		int secs = de->ctime_cs / 100;
-		int csecs = de->ctime_cs % 100;
-		inode->i_ctime.tv_sec  =
-			date_dos2unix(le16_to_cpu(de->ctime),
-				      le16_to_cpu(de->cdate),
-				      sbi->options.tz_utc) + secs;
-		inode->i_ctime.tv_nsec = csecs * 10000000;
-		inode->i_atime.tv_sec =
-			date_dos2unix(0, le16_to_cpu(de->adate),
-				      sbi->options.tz_utc);
-		inode->i_atime.tv_nsec = 0;
+		fat_time_fat2unix(sbi, &inode->i_ctime, de->ctime,
+				  de->cdate, de->ctime_cs);
+		fat_time_fat2unix(sbi, &inode->i_atime, 0, de->adate, 0);
 	} else
 		inode->i_ctime = inode->i_atime = inode->i_mtime;
 
@@ -443,13 +435,8 @@ static void fat_delete_inode(struct inode *inode)
 
 static void fat_clear_inode(struct inode *inode)
 {
-	struct super_block *sb = inode->i_sb;
-	struct msdos_sb_info *sbi = MSDOS_SB(sb);
-
-	spin_lock(&sbi->inode_hash_lock);
 	fat_cache_inval_inode(inode);
-	hlist_del_init(&MSDOS_I(inode)->i_fat_hash);
-	spin_unlock(&sbi->inode_hash_lock);
+	fat_detach(inode);
 }
 
 static void fat_write_super(struct super_block *sb)
@@ -555,6 +542,20 @@ static int fat_statfs(struct dentry *dentry, struct kstatfs *buf)
 	return 0;
 }
 
+static inline loff_t fat_i_pos_read(struct msdos_sb_info *sbi,
+				    struct inode *inode)
+{
+	loff_t i_pos;
+#if BITS_PER_LONG == 32
+	spin_lock(&sbi->inode_hash_lock);
+#endif
+	i_pos = MSDOS_I(inode)->i_pos;
+#if BITS_PER_LONG == 32
+	spin_unlock(&sbi->inode_hash_lock);
+#endif
+	return i_pos;
+}
+
 static int fat_write_inode(struct inode *inode, int wait)
 {
 	struct super_block *sb = inode->i_sb;
@@ -564,9 +565,12 @@ static int fat_write_inode(struct inode *inode, int wait)
 	loff_t i_pos;
 	int err;
 
+	if (inode->i_ino == MSDOS_ROOT_INO)
+		return 0;
+
 retry:
-	i_pos = MSDOS_I(inode)->i_pos;
-	if (inode->i_ino == MSDOS_ROOT_INO || !i_pos)
+	i_pos = fat_i_pos_read(sbi, inode);
+	if (!i_pos)
 		return 0;
 
 	bh = sb_bread(sb, i_pos >> sbi->dir_per_block_bits);
@@ -588,19 +592,17 @@ retry:
 		raw_entry->size = 0;
 	else
 		raw_entry->size = cpu_to_le32(inode->i_size);
-	raw_entry->attr = fat_attr(inode);
+	raw_entry->attr = fat_make_attrs(inode);
 	raw_entry->start = cpu_to_le16(MSDOS_I(inode)->i_logstart);
 	raw_entry->starthi = cpu_to_le16(MSDOS_I(inode)->i_logstart >> 16);
-	fat_date_unix2dos(inode->i_mtime.tv_sec, &raw_entry->time,
-			  &raw_entry->date, sbi->options.tz_utc);
+	fat_time_unix2fat(sbi, &inode->i_mtime, &raw_entry->time,
+			  &raw_entry->date, NULL);
 	if (sbi->options.isvfat) {
 		__le16 atime;
-		fat_date_unix2dos(inode->i_ctime.tv_sec, &raw_entry->ctime,
-				  &raw_entry->cdate, sbi->options.tz_utc);
-		fat_date_unix2dos(inode->i_atime.tv_sec, &atime,
-				  &raw_entry->adate, sbi->options.tz_utc);
-		raw_entry->ctime_cs = (inode->i_ctime.tv_sec & 1) * 100 +
-			inode->i_ctime.tv_nsec / 10000000;
+		fat_time_unix2fat(sbi, &inode->i_ctime, &raw_entry->ctime,
+				  &raw_entry->cdate, &raw_entry->ctime_cs);
+		fat_time_unix2fat(sbi, &inode->i_atime, &atime,
+				  &raw_entry->adate, NULL);
 	}
 	spin_unlock(&sbi->inode_hash_lock);
 	mark_buffer_dirty(bh);
@@ -819,8 +821,10 @@ static int fat_show_options(struct seq_file *m, struct vfsmount *mnt)
 			seq_puts(m, ",uni_xlate");
 		if (!opts->numtail)
 			seq_puts(m, ",nonumtail");
+		if (opts->rodir)
+			seq_puts(m, ",rodir");
 	}
-	if (sbi->options.flush)
+	if (opts->flush)
 		seq_puts(m, ",flush");
 	if (opts->tz_utc)
 		seq_puts(m, ",tz=UTC");
@@ -836,7 +840,7 @@ enum {
 	Opt_charset, Opt_shortname_lower, Opt_shortname_win95,
 	Opt_shortname_winnt, Opt_shortname_mixed, Opt_utf8_no, Opt_utf8_yes,
 	Opt_uni_xl_no, Opt_uni_xl_yes, Opt_nonumtail_no, Opt_nonumtail_yes,
-	Opt_obsolate, Opt_flush, Opt_tz_utc, Opt_err,
+	Opt_obsolate, Opt_flush, Opt_tz_utc, Opt_rodir, Opt_err,
 };
 
 static const match_table_t fat_tokens = {
@@ -908,6 +912,7 @@ static const match_table_t vfat_tokens = {
 	{Opt_nonumtail_yes, "nonumtail=yes"},
 	{Opt_nonumtail_yes, "nonumtail=true"},
 	{Opt_nonumtail_yes, "nonumtail"},
+	{Opt_rodir, "rodir"},
 	{Opt_err, NULL}
 };
 
@@ -927,10 +932,13 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug,
 	opts->allow_utime = -1;
 	opts->codepage = fat_default_codepage;
 	opts->iocharset = fat_default_iocharset;
-	if (is_vfat)
+	if (is_vfat) {
 		opts->shortname = VFAT_SFN_DISPLAY_LOWER|VFAT_SFN_CREATE_WIN95;
-	else
+		opts->rodir = 0;
+	} else {
 		opts->shortname = 0;
+		opts->rodir = 1;
+	}
 	opts->name_check = 'n';
 	opts->quiet = opts->showexec = opts->sys_immutable = opts->dotsOK =  0;
 	opts->utf8 = opts->unicode_xlate = 0;
@@ -1081,6 +1089,9 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug,
 		case Opt_nonumtail_yes:		/* empty or 1 or yes or true */
 			opts->numtail = 0;	/* negated option */
 			break;
+		case Opt_rodir:
+			opts->rodir = 1;
+			break;
 
 		/* obsolete mount options */
 		case Opt_obsolate:
@@ -1126,7 +1137,7 @@ static int fat_read_root(struct inode *inode)
 	inode->i_gid = sbi->options.fs_gid;
 	inode->i_version++;
 	inode->i_generation = 0;
-	inode->i_mode = (S_IRWXUGO & ~sbi->options.fs_dmask) | S_IFDIR;
+	inode->i_mode = fat_make_mode(sbi, ATTR_DIR, S_IRWXUGO);
 	inode->i_op = sbi->dir_ops;
 	inode->i_fop = &fat_dir_operations;
 	if (sbi->fat_bits == 32) {
@@ -1143,7 +1154,7 @@ static int fat_read_root(struct inode *inode)
 	MSDOS_I(inode)->i_logstart = 0;
 	MSDOS_I(inode)->mmu_private = inode->i_size;
 
-	MSDOS_I(inode)->i_attrs = ATTR_NONE;
+	fat_save_attrs(inode, ATTR_DIR);
 	inode->i_mtime.tv_sec = inode->i_atime.tv_sec = inode->i_ctime.tv_sec = 0;
 	inode->i_mtime.tv_nsec = inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec = 0;
 	inode->i_nlink = fat_subdirs(inode)+2;
diff --git a/fs/fat/misc.c b/fs/fat/misc.c
index 79fb98ad36d4..ac39ebcc1496 100644
--- a/fs/fat/misc.c
+++ b/fs/fat/misc.c
@@ -8,8 +8,8 @@
 
 #include <linux/module.h>
 #include <linux/fs.h>
-#include <linux/msdos_fs.h>
 #include <linux/buffer_head.h>
+#include "fat.h"
 
 /*
  * fat_fs_panic reports a severe file system problem and sets the file system
@@ -124,8 +124,9 @@ int fat_chain_add(struct inode *inode, int new_dclus, int nr_cluster)
 			mark_inode_dirty(inode);
 	}
 	if (new_fclus != (inode->i_blocks >> (sbi->cluster_bits - 9))) {
-		fat_fs_panic(sb, "clusters badly computed (%d != %lu)",
-			new_fclus, inode->i_blocks >> (sbi->cluster_bits - 9));
+		fat_fs_panic(sb, "clusters badly computed (%d != %llu)",
+			     new_fclus,
+			     (llu)(inode->i_blocks >> (sbi->cluster_bits - 9)));
 		fat_cache_inval_inode(inode);
 	}
 	inode->i_blocks += nr_cluster << (sbi->cluster_bits - 9);
@@ -135,65 +136,131 @@ int fat_chain_add(struct inode *inode, int new_dclus, int nr_cluster)
 
 extern struct timezone sys_tz;
 
+/*
+ * The epoch of FAT timestamp is 1980.
+ *     :  bits :     value
+ * date:  0 -  4: day	(1 -  31)
+ * date:  5 -  8: month	(1 -  12)
+ * date:  9 - 15: year	(0 - 127) from 1980
+ * time:  0 -  4: sec	(0 -  29) 2sec counts
+ * time:  5 - 10: min	(0 -  59)
+ * time: 11 - 15: hour	(0 -  23)
+ */
+#define SECS_PER_MIN	60
+#define SECS_PER_HOUR	(60 * 60)
+#define SECS_PER_DAY	(SECS_PER_HOUR * 24)
+#define UNIX_SECS_1980	315532800L
+#if BITS_PER_LONG == 64
+#define UNIX_SECS_2108	4354819200L
+#endif
+/* days between 1.1.70 and 1.1.80 (2 leap days) */
+#define DAYS_DELTA	(365 * 10 + 2)
+/* 120 (2100 - 1980) isn't leap year */
+#define YEAR_2100	120
+#define IS_LEAP_YEAR(y)	(!((y) & 3) && (y) != YEAR_2100)
+
 /* Linear day numbers of the respective 1sts in non-leap years. */
-static int day_n[] = {
-   /* Jan  Feb  Mar  Apr   May  Jun  Jul  Aug  Sep  Oct  Nov  Dec */
-	0,  31,  59,  90,  120, 151, 181, 212, 243, 273, 304, 334, 0, 0, 0, 0
+static time_t days_in_year[] = {
+	/* Jan  Feb  Mar  Apr  May  Jun  Jul  Aug  Sep  Oct  Nov  Dec */
+	0,   0,  31,  59,  90, 120, 151, 181, 212, 243, 273, 304, 334, 0, 0, 0,
 };
 
-/* Convert a MS-DOS time/date pair to a UNIX date (seconds since 1 1 70). */
-int date_dos2unix(unsigned short time, unsigned short date, int tz_utc)
+/* Convert a FAT time/date pair to a UNIX date (seconds since 1 1 70). */
+void fat_time_fat2unix(struct msdos_sb_info *sbi, struct timespec *ts,
+		       __le16 __time, __le16 __date, u8 time_cs)
 {
-	int month, year, secs;
+	u16 time = le16_to_cpu(__time), date = le16_to_cpu(__date);
+	time_t second, day, leap_day, month, year;
 
-	/*
-	 * first subtract and mask after that... Otherwise, if
-	 * date == 0, bad things happen
-	 */
-	month = ((date >> 5) - 1) & 15;
-	year = date >> 9;
-	secs = (time & 31)*2+60*((time >> 5) & 63)+(time >> 11)*3600+86400*
-	    ((date & 31)-1+day_n[month]+(year/4)+year*365-((year & 3) == 0 &&
-	    month < 2 ? 1 : 0)+3653);
-			/* days since 1.1.70 plus 80's leap day */
-	if (!tz_utc)
-		secs += sys_tz.tz_minuteswest*60;
-	return secs;
+	year  = date >> 9;
+	month = max(1, (date >> 5) & 0xf);
+	day   = max(1, date & 0x1f) - 1;
+
+	leap_day = (year + 3) / 4;
+	if (year > YEAR_2100)		/* 2100 isn't leap year */
+		leap_day--;
+	if (IS_LEAP_YEAR(year) && month > 2)
+		leap_day++;
+
+	second =  (time & 0x1f) << 1;
+	second += ((time >> 5) & 0x3f) * SECS_PER_MIN;
+	second += (time >> 11) * SECS_PER_HOUR;
+	second += (year * 365 + leap_day
+		   + days_in_year[month] + day
+		   + DAYS_DELTA) * SECS_PER_DAY;
+
+	if (!sbi->options.tz_utc)
+		second += sys_tz.tz_minuteswest * SECS_PER_MIN;
+
+	if (time_cs) {
+		ts->tv_sec = second + (time_cs / 100);
+		ts->tv_nsec = (time_cs % 100) * 10000000;
+	} else {
+		ts->tv_sec = second;
+		ts->tv_nsec = 0;
+	}
 }
 
-/* Convert linear UNIX date to a MS-DOS time/date pair. */
-void fat_date_unix2dos(int unix_date, __le16 *time, __le16 *date, int tz_utc)
+/* Convert linear UNIX date to a FAT time/date pair. */
+void fat_time_unix2fat(struct msdos_sb_info *sbi, struct timespec *ts,
+		       __le16 *time, __le16 *date, u8 *time_cs)
 {
-	int day, year, nl_day, month;
+	time_t second = ts->tv_sec;
+	time_t day, leap_day, month, year;
 
-	if (!tz_utc)
-		unix_date -= sys_tz.tz_minuteswest*60;
+	if (!sbi->options.tz_utc)
+		second -= sys_tz.tz_minuteswest * SECS_PER_MIN;
 
 	/* Jan 1 GMT 00:00:00 1980. But what about another time zone? */
-	if (unix_date < 315532800)
-		unix_date = 315532800;
-
-	*time = cpu_to_le16((unix_date % 60)/2+(((unix_date/60) % 60) << 5)+
-	    (((unix_date/3600) % 24) << 11));
-	day = unix_date/86400-3652;
-	year = day/365;
-	if ((year+3)/4+365*year > day)
+	if (second < UNIX_SECS_1980) {
+		*time = 0;
+		*date = cpu_to_le16((0 << 9) | (1 << 5) | 1);
+		if (time_cs)
+			*time_cs = 0;
+		return;
+	}
+#if BITS_PER_LONG == 64
+	if (second >= UNIX_SECS_2108) {
+		*time = cpu_to_le16((23 << 11) | (59 << 5) | 29);
+		*date = cpu_to_le16((127 << 9) | (12 << 5) | 31);
+		if (time_cs)
+			*time_cs = 199;
+		return;
+	}
+#endif
+
+	day = second / SECS_PER_DAY - DAYS_DELTA;
+	year = day / 365;
+	leap_day = (year + 3) / 4;
+	if (year > YEAR_2100)		/* 2100 isn't leap year */
+		leap_day--;
+	if (year * 365 + leap_day > day)
 		year--;
-	day -= (year+3)/4+365*year;
-	if (day == 59 && !(year & 3)) {
-		nl_day = day;
+	leap_day = (year + 3) / 4;
+	if (year > YEAR_2100)		/* 2100 isn't leap year */
+		leap_day--;
+	day -= year * 365 + leap_day;
+
+	if (IS_LEAP_YEAR(year) && day == days_in_year[3]) {
 		month = 2;
 	} else {
-		nl_day = (year & 3) || day <= 59 ? day : day-1;
-		for (month = 0; month < 12; month++) {
-			if (day_n[month] > nl_day)
+		if (IS_LEAP_YEAR(year) && day > days_in_year[3])
+			day--;
+		for (month = 1; month < 12; month++) {
+			if (days_in_year[month + 1] > day)
 				break;
 		}
 	}
-	*date = cpu_to_le16(nl_day-day_n[month-1]+1+(month << 5)+(year << 9));
-}
+	day -= days_in_year[month];
 
-EXPORT_SYMBOL_GPL(fat_date_unix2dos);
+	*time = cpu_to_le16(((second / SECS_PER_HOUR) % 24) << 11
+			    | ((second / SECS_PER_MIN) % 60) << 5
+			    | (second % SECS_PER_MIN) >> 1);
+	*date = cpu_to_le16((year << 9) | (month << 5) | (day + 1));
+	if (time_cs)
+		*time_cs = (ts->tv_sec & 1) * 100 + ts->tv_nsec / 10000000;
+}
+EXPORT_SYMBOL_GPL(fat_time_unix2fat);
 
 int fat_sync_bhs(struct buffer_head **bhs, int nr_bhs)
 {
diff --git a/fs/msdos/namei.c b/fs/fat/namei_msdos.c
index e844b9809d27..7ba03a4acbe0 100644
--- a/fs/msdos/namei.c
+++ b/fs/fat/namei_msdos.c
@@ -9,8 +9,8 @@
 #include <linux/module.h>
 #include <linux/time.h>
 #include <linux/buffer_head.h>
-#include <linux/msdos_fs.h>
 #include <linux/smp_lock.h>
+#include "fat.h"
 
 /* Characters that are undesirable in an MS-DOS file name */
 static unsigned char bad_chars[] = "*?<>|\"";
@@ -203,33 +203,37 @@ static struct dentry *msdos_lookup(struct inode *dir, struct dentry *dentry,
 {
 	struct super_block *sb = dir->i_sb;
 	struct fat_slot_info sinfo;
-	struct inode *inode = NULL;
-	int res;
-
-	dentry->d_op = &msdos_dentry_operations;
+	struct inode *inode;
+	int err;
 
 	lock_super(sb);
-	res = msdos_find(dir, dentry->d_name.name, dentry->d_name.len, &sinfo);
-	if (res == -ENOENT)
-		goto add;
-	if (res < 0)
-		goto out;
+
+	err = msdos_find(dir, dentry->d_name.name, dentry->d_name.len, &sinfo);
+	if (err) {
+		if (err == -ENOENT) {
+			inode = NULL;
+			goto out;
+		}
+		goto error;
+	}
+
 	inode = fat_build_inode(sb, sinfo.de, sinfo.i_pos);
 	brelse(sinfo.bh);
 	if (IS_ERR(inode)) {
-		res = PTR_ERR(inode);
-		goto out;
+		err = PTR_ERR(inode);
+		goto error;
 	}
-add:
-	res = 0;
+out:
+	unlock_super(sb);
+	dentry->d_op = &msdos_dentry_operations;
 	dentry = d_splice_alias(inode, dentry);
 	if (dentry)
 		dentry->d_op = &msdos_dentry_operations;
-out:
+	return dentry;
+
+error:
 	unlock_super(sb);
-	if (!res)
-		return dentry;
-	return ERR_PTR(res);
+	return ERR_PTR(err);
 }
 
 /***** Creates a directory entry (name is already formatted). */
@@ -247,7 +251,7 @@ static int msdos_add_entry(struct inode *dir, const unsigned char *name,
 	if (is_hid)
 		de.attr |= ATTR_HIDDEN;
 	de.lcase = 0;
-	fat_date_unix2dos(ts->tv_sec, &time, &date, sbi->options.tz_utc);
+	fat_time_unix2fat(sbi, ts, &time, &date, NULL);
 	de.cdate = de.adate = 0;
 	de.ctime = 0;
 	de.ctime_cs = 0;
diff --git a/fs/vfat/namei.c b/fs/fat/namei_vfat.c
index 155c10b4adbd..bf326d4356a3 100644
--- a/fs/vfat/namei.c
+++ b/fs/fat/namei_vfat.c
@@ -16,36 +16,75 @@
  */
 
 #include <linux/module.h>
-
 #include <linux/jiffies.h>
-#include <linux/msdos_fs.h>
 #include <linux/ctype.h>
 #include <linux/slab.h>
 #include <linux/smp_lock.h>
 #include <linux/buffer_head.h>
 #include <linux/namei.h>
+#include "fat.h"
 
-static int vfat_revalidate(struct dentry *dentry, struct nameidata *nd)
+/*
+ * If new entry was created in the parent, it could create the 8.3
+ * alias (the shortname of logname).  So, the parent may have the
+ * negative-dentry which matches the created 8.3 alias.
+ *
+ * If it happened, the negative dentry isn't actually negative
+ * anymore.  So, drop it.
+ */
+static int vfat_revalidate_shortname(struct dentry *dentry)
 {
 	int ret = 1;
-
-	if (!dentry->d_inode &&
-	    nd && !(nd->flags & LOOKUP_CONTINUE) && (nd->flags & LOOKUP_CREATE))
-		/*
-		 * negative dentry is dropped, in order to make sure
-		 * to use the name which a user desires if this is
-		 * create path.
-		 */
+	spin_lock(&dentry->d_lock);
+	if (dentry->d_time != dentry->d_parent->d_inode->i_version)
 		ret = 0;
-	else {
-		spin_lock(&dentry->d_lock);
-		if (dentry->d_time != dentry->d_parent->d_inode->i_version)
-			ret = 0;
-		spin_unlock(&dentry->d_lock);
-	}
+	spin_unlock(&dentry->d_lock);
 	return ret;
 }
 
+static int vfat_revalidate(struct dentry *dentry, struct nameidata *nd)
+{
+	/* This is not negative dentry. Always valid. */
+	if (dentry->d_inode)
+		return 1;
+	return vfat_revalidate_shortname(dentry);
+}
+
+static int vfat_revalidate_ci(struct dentry *dentry, struct nameidata *nd)
+{
+	/*
+	 * This is not negative dentry. Always valid.
+	 *
+	 * Note, rename() to existing directory entry will have ->d_inode,
+	 * and will use existing name which isn't specified name by user.
+	 *
+	 * We may be able to drop this positive dentry here. But dropping
+	 * positive dentry isn't good idea. So it's unsupported like
+	 * rename("filename", "FILENAME") for now.
+	 */
+	if (dentry->d_inode)
+		return 1;
+
+	/*
+	 * This may be nfsd (or something), anyway, we can't see the
+	 * intent of this. So, since this can be for creation, drop it.
+	 */
+	if (!nd)
+		return 0;
+
+	/*
+	 * Drop the negative dentry, in order to make sure to use the
+	 * case sensitive name which is specified by user if this is
+	 * for creation.
+	 */
+	if (!(nd->flags & (LOOKUP_CONTINUE | LOOKUP_PARENT))) {
+		if (nd->flags & LOOKUP_CREATE)
+			return 0;
+	}
+
+	return vfat_revalidate_shortname(dentry);
+}
+
 /* returns the length of a struct qstr, ignoring trailing dots */
 static unsigned int vfat_striptail_len(struct qstr *qstr)
 {
@@ -127,25 +166,16 @@ static int vfat_cmp(struct dentry *dentry, struct qstr *a, struct qstr *b)
 	return 1;
 }
 
-static struct dentry_operations vfat_dentry_ops[4] = {
-	{
-		.d_hash		= vfat_hashi,
-		.d_compare	= vfat_cmpi,
-	},
-	{
-		.d_revalidate	= vfat_revalidate,
-		.d_hash		= vfat_hashi,
-		.d_compare	= vfat_cmpi,
-	},
-	{
-		.d_hash		= vfat_hash,
-		.d_compare	= vfat_cmp,
-	},
-	{
-		.d_revalidate	= vfat_revalidate,
-		.d_hash		= vfat_hash,
-		.d_compare	= vfat_cmp,
-	}
+static struct dentry_operations vfat_ci_dentry_ops = {
+	.d_revalidate	= vfat_revalidate_ci,
+	.d_hash		= vfat_hashi,
+	.d_compare	= vfat_cmpi,
+};
+
+static struct dentry_operations vfat_dentry_ops = {
+	.d_revalidate	= vfat_revalidate,
+	.d_hash		= vfat_hash,
+	.d_compare	= vfat_cmp,
 };
 
 /* Characters that are undesirable in an MS-DOS file name */
@@ -569,6 +599,7 @@ static int vfat_build_slots(struct inode *dir, const unsigned char *name,
 	unsigned char msdos_name[MSDOS_NAME];
 	wchar_t *uname;
 	__le16 time, date;
+	u8 time_cs;
 	int err, ulen, usize, i;
 	loff_t offset;
 
@@ -621,10 +652,10 @@ shortname:
 	memcpy(de->name, msdos_name, MSDOS_NAME);
 	de->attr = is_dir ? ATTR_DIR : ATTR_ARCH;
 	de->lcase = lcase;
-	fat_date_unix2dos(ts->tv_sec, &time, &date, sbi->options.tz_utc);
+	fat_time_unix2fat(sbi, ts, &time, &date, &time_cs);
 	de->time = de->ctime = time;
 	de->date = de->cdate = de->adate = date;
-	de->ctime_cs = 0;
+	de->ctime_cs = time_cs;
 	de->start = cpu_to_le16(cluster);
 	de->starthi = cpu_to_le16(cluster >> 16);
 	de->size = 0;
@@ -683,46 +714,58 @@ static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry,
 {
 	struct super_block *sb = dir->i_sb;
 	struct fat_slot_info sinfo;
-	struct inode *inode = NULL;
+	struct inode *inode;
 	struct dentry *alias;
-	int err, table;
+	int err;
 
 	lock_super(sb);
-	table = (MSDOS_SB(sb)->options.name_check == 's') ? 2 : 0;
-	dentry->d_op = &vfat_dentry_ops[table];
 
 	err = vfat_find(dir, &dentry->d_name, &sinfo);
 	if (err) {
-		table++;
+		if (err == -ENOENT) {
+			inode = NULL;
+			goto out;
+		}
 		goto error;
 	}
+
 	inode = fat_build_inode(sb, sinfo.de, sinfo.i_pos);
 	brelse(sinfo.bh);
 	if (IS_ERR(inode)) {
-		unlock_super(sb);
-		return ERR_CAST(inode);
+		err = PTR_ERR(inode);
+		goto error;
 	}
-	alias = d_find_alias(inode);
-	if (alias) {
-		if (d_invalidate(alias) == 0)
-			dput(alias);
-		else {
-			iput(inode);
-			unlock_super(sb);
-			return alias;
-		}
 
+	alias = d_find_alias(inode);
+	if (alias && !(alias->d_flags & DCACHE_DISCONNECTED)) {
+		/*
+		 * This inode has non DCACHE_DISCONNECTED dentry. This
+		 * means, the user did ->lookup() by an another name
+		 * (longname vs 8.3 alias of it) in past.
+		 *
+		 * Switch to new one for reason of locality if possible.
+		 */
+		BUG_ON(d_unhashed(alias));
+		if (!S_ISDIR(inode->i_mode))
+			d_move(alias, dentry);
+		iput(inode);
+		unlock_super(sb);
+		return alias;
 	}
-error:
+out:
 	unlock_super(sb);
-	dentry->d_op = &vfat_dentry_ops[table];
+	dentry->d_op = sb->s_root->d_op;
 	dentry->d_time = dentry->d_parent->d_inode->i_version;
 	dentry = d_splice_alias(inode, dentry);
 	if (dentry) {
-		dentry->d_op = &vfat_dentry_ops[table];
+		dentry->d_op = sb->s_root->d_op;
 		dentry->d_time = dentry->d_parent->d_inode->i_version;
 	}
 	return dentry;
+
+error:
+	unlock_super(sb);
+	return ERR_PTR(err);
 }
 
 static int vfat_create(struct inode *dir, struct dentry *dentry, int mode,
@@ -1014,9 +1057,9 @@ static int vfat_fill_super(struct super_block *sb, void *data, int silent)
 		return res;
 
 	if (MSDOS_SB(sb)->options.name_check != 's')
-		sb->s_root->d_op = &vfat_dentry_ops[0];
+		sb->s_root->d_op = &vfat_ci_dentry_ops;
 	else
-		sb->s_root->d_op = &vfat_dentry_ops[2];
+		sb->s_root->d_op = &vfat_dentry_ops;
 
 	return 0;
 }
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/jffs2/background.c b/fs/jffs2/background.c
index 8adebd3e43c6..3cceef4ad2b7 100644
--- a/fs/jffs2/background.c
+++ b/fs/jffs2/background.c
@@ -85,15 +85,15 @@ static int jffs2_garbage_collect_thread(void *_c)
 	for (;;) {
 		allow_signal(SIGHUP);
 	again:
+		spin_lock(&c->erase_completion_lock);
 		if (!jffs2_thread_should_wake(c)) {
 			set_current_state (TASK_INTERRUPTIBLE);
+			spin_unlock(&c->erase_completion_lock);
 			D1(printk(KERN_DEBUG "jffs2_garbage_collect_thread sleeping...\n"));
-			/* Yes, there's a race here; we checked jffs2_thread_should_wake()
-			   before setting current->state to TASK_INTERRUPTIBLE. But it doesn't
-			   matter - We don't care if we miss a wakeup, because the GC thread
-			   is only an optimisation anyway. */
 			schedule();
-		}
+		} else
+			spin_unlock(&c->erase_completion_lock);
+			
 
 		/* This thread is purely an optimisation. But if it runs when
 		   other things could be running, it actually makes things a
diff --git a/fs/jffs2/compr_lzo.c b/fs/jffs2/compr_lzo.c
index 47b045797e42..90cb60d09787 100644
--- a/fs/jffs2/compr_lzo.c
+++ b/fs/jffs2/compr_lzo.c
@@ -19,7 +19,7 @@
 
 static void *lzo_mem;
 static void *lzo_compress_buf;
-static DEFINE_MUTEX(deflate_mutex);
+static DEFINE_MUTEX(deflate_mutex);	/* for lzo_mem and lzo_compress_buf */
 
 static void free_workspace(void)
 {
@@ -49,18 +49,21 @@ static int jffs2_lzo_compress(unsigned char *data_in, unsigned char *cpage_out,
 
 	mutex_lock(&deflate_mutex);
 	ret = lzo1x_1_compress(data_in, *sourcelen, lzo_compress_buf, &compress_size, lzo_mem);
-	mutex_unlock(&deflate_mutex);
-
 	if (ret != LZO_E_OK)
-		return -1;
+		goto fail;
 
 	if (compress_size > *dstlen)
-		return -1;
+		goto fail;
 
 	memcpy(cpage_out, lzo_compress_buf, compress_size);
-	*dstlen = compress_size;
+	mutex_unlock(&deflate_mutex);
 
+	*dstlen = compress_size;
 	return 0;
+
+ fail:
+	mutex_unlock(&deflate_mutex);
+	return -1;
 }
 
 static int jffs2_lzo_decompress(unsigned char *data_in, unsigned char *cpage_out,
diff --git a/fs/jffs2/nodemgmt.c b/fs/jffs2/nodemgmt.c
index 0875b60b4bf7..21a052915aa9 100644
--- a/fs/jffs2/nodemgmt.c
+++ b/fs/jffs2/nodemgmt.c
@@ -261,9 +261,11 @@ static int jffs2_find_nextblock(struct jffs2_sb_info *c)
 
 	jffs2_sum_reset_collected(c->summary); /* reset collected summary */
 
+#ifdef CONFIG_JFFS2_FS_WRITEBUFFER
 	/* adjust write buffer offset, else we get a non contiguous write bug */
 	if (!(c->wbuf_ofs % c->sector_size) && !c->wbuf_len)
 		c->wbuf_ofs = 0xffffffff;
+#endif
 
 	D1(printk(KERN_DEBUG "jffs2_find_nextblock(): new nextblock = 0x%08x\n", c->nextblock->offset));
 
diff --git a/fs/msdos/Makefile b/fs/msdos/Makefile
deleted file mode 100644
index ea67646fcb95..000000000000
--- a/fs/msdos/Makefile
+++ /dev/null
@@ -1,7 +0,0 @@
-#
-# Makefile for the Linux msdos filesystem routines.
-#
-
-obj-$(CONFIG_MSDOS_FS) += msdos.o
-
-msdos-y := namei.o
diff --git a/fs/namespace.c b/fs/namespace.c
index d8bc2c4704a5..1c09cab8f7cf 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1815,8 +1815,8 @@ static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts)
 		while (!list_empty(&graveyard)) {
 			m = list_first_entry(&graveyard, struct vfsmount,
 						mnt_expire);
-			touch_mnt_namespace(mnt->mnt_ns);
-			umount_tree(mnt, 1, umounts);
+			touch_mnt_namespace(m->mnt_ns);
+			umount_tree(m, 1, umounts);
 		}
 	}
 }
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index b59ec5a6ed24..d1c5f787b365 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1876,11 +1876,11 @@ 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;
 
@@ -1913,9 +1913,6 @@ 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 f95f3654ee68..2545e7402efe 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/vfat/Makefile b/fs/vfat/Makefile
deleted file mode 100644
index 40f2798a4f08..000000000000
--- a/fs/vfat/Makefile
+++ /dev/null
@@ -1,7 +0,0 @@
-#
-# Makefile for the linux vfat-filesystem routines.
-#
-
-obj-$(CONFIG_VFAT_FS) += vfat.o
-
-vfat-y := namei.o
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
author	James Morris	2008-11-14 11:29:12 +1100
committer	James Morris	2008-11-14 11:29:12 +1100
commit	2b828925652340277a889cbc11b2d0637f7cdaf7 (patch)
tree	32fcb3d3e466fc419fad2d3717956a5b5ad3d35a /fs
parent	3a3b7ce9336952ea7b9564d976d068a238976c9d (diff)
parent	58e20d8d344b0ee083febb18c2b021d2427e56ca (diff)