Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (44 commits)
  ext4: fix trimming starting with block 0 with small blocksize
  ext4: revert buggy trim overflow patch
  ext4: don't pass entire map to check_eofblocks_fl
  ext4: fix memory leak in ext4_free_branches
  ext4: remove ext4_mb_return_to_preallocation()
  ext4: flush the i_completed_io_list during ext4_truncate
  ext4: add error checking to calls to ext4_handle_dirty_metadata()
  ext4: fix trimming of a single group
  ext4: fix uninitialized variable in ext4_register_li_request
  ext4: dynamically allocate the jbd2_inode in ext4_inode_info as necessary
  ext4: drop i_state_flags on architectures with 64-bit longs
  ext4: reorder ext4_inode_info structure elements to remove unneeded padding
  ext4: drop ec_type from the ext4_ext_cache structure
  ext4: use ext4_lblk_t instead of sector_t for logical blocks
  ext4: replace i_delalloc_reserved_flag with EXT4_STATE_DELALLOC_RESERVED
  ext4: fix 32bit overflow in ext4_ext_find_goal()
  ext4: add more error checks to ext4_mkdir()
  ext4: ext4_ext_migrate should use NULL not 0
  ext4: Use ext4_error_file() to print the pathname to the corrupted inode
  ext4: use IS_ERR() to check for errors in ext4_error_file
  ...

22 files changed, 543 insertions, 376 deletions

diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index f84700be3274..c2e4dce984d2 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -199,14 +199,6 @@ bad_block:	ext2_error(inode->i_sb, "ext2_xattr_get",
 			goto found;
 		entry = next;
 	}
-	/* Check the remaining name entries */
-	while (!IS_LAST_ENTRY(entry)) {
-		struct ext2_xattr_entry *next =
-			EXT2_XATTR_NEXT(entry);
-		if ((char *)next >= end)
-			goto bad_block;
-		entry = next;
-	}
 	if (ext2_xattr_cache_insert(bh))
 		ea_idebug(inode, "cache insert failed");
 	error = -ENODATA;
@@ -355,7 +347,7 @@ static void ext2_xattr_update_super_block(struct super_block *sb)
 /*
  * ext2_xattr_set()
  *
- * Create, replace or remove an extended attribute for this inode. Buffer
+ * Create, replace or remove an extended attribute for this inode.  Value
  * is NULL to remove an existing extended attribute, and non-NULL to
  * either replace an existing extended attribute, or create a new extended
  * attribute. The flags XATTR_REPLACE and XATTR_CREATE
diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c
index e69dc6dfaa89..32e6cc23bd9a 100644
--- a/fs/ext3/xattr.c
+++ b/fs/ext3/xattr.c
@@ -925,7 +925,7 @@ ext3_xattr_ibody_set(handle_t *handle, struct inode *inode,
 /*
  * ext3_xattr_set_handle()
  *
- * Create, replace or remove an extended attribute for this inode. Buffer
+ * Create, replace or remove an extended attribute for this inode.  Value
  * is NULL to remove an existing extended attribute, and non-NULL to
  * either replace an existing extended attribute, or create a new extended
  * attribute. The flags XATTR_REPLACE and XATTR_CREATE
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index 14c3af26c671..adf96b822781 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -592,7 +592,8 @@ ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode,
 	 * Account for the allocated meta blocks.  We will never
 	 * fail EDQUOT for metdata, but we do account for it.
 	 */
-	if (!(*errp) && EXT4_I(inode)->i_delalloc_reserved_flag) {
+	if (!(*errp) &&
+	    ext4_test_inode_state(inode, EXT4_STATE_DELALLOC_RESERVED)) {
 		spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
 		EXT4_I(inode)->i_allocated_meta_blocks += ar.len;
 		spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index ece76fb6a40c..164c56092e58 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -60,9 +60,13 @@ static unsigned char get_dtype(struct super_block *sb, int filetype)
 	return (ext4_filetype_table[filetype]);
 }
 
-
+/*
+ * Return 0 if the directory entry is OK, and 1 if there is a problem
+ *
+ * Note: this is the opposite of what ext2 and ext3 historically returned...
+ */
 int __ext4_check_dir_entry(const char *function, unsigned int line,
-			   struct inode *dir,
+			   struct inode *dir, struct file *filp,
 			   struct ext4_dir_entry_2 *de,
 			   struct buffer_head *bh,
 			   unsigned int offset)
@@ -71,26 +75,37 @@ int __ext4_check_dir_entry(const char *function, unsigned int line,
 	const int rlen = ext4_rec_len_from_disk(de->rec_len,
 						dir->i_sb->s_blocksize);
 
-	if (rlen < EXT4_DIR_REC_LEN(1))
+	if (unlikely(rlen < EXT4_DIR_REC_LEN(1)))
 		error_msg = "rec_len is smaller than minimal";
-	else if (rlen % 4 != 0)
+	else if (unlikely(rlen % 4 != 0))
 		error_msg = "rec_len % 4 != 0";
-	else if (rlen < EXT4_DIR_REC_LEN(de->name_len))
+	else if (unlikely(rlen < EXT4_DIR_REC_LEN(de->name_len)))
 		error_msg = "rec_len is too small for name_len";
-	else if (((char *) de - bh->b_data) + rlen > dir->i_sb->s_blocksize)
+	else if (unlikely(((char *) de - bh->b_data) + rlen >
+			  dir->i_sb->s_blocksize))
 		error_msg = "directory entry across blocks";
-	else if (le32_to_cpu(de->inode) >
-			le32_to_cpu(EXT4_SB(dir->i_sb)->s_es->s_inodes_count))
+	else if (unlikely(le32_to_cpu(de->inode) >
+			le32_to_cpu(EXT4_SB(dir->i_sb)->s_es->s_inodes_count)))
 		error_msg = "inode out of bounds";
+	else
+		return 0;
 
-	if (error_msg != NULL)
-		ext4_error_inode(dir, function, line, bh->b_blocknr,
-			"bad entry in directory: %s - "
-			"offset=%u(%u), inode=%u, rec_len=%d, name_len=%d",
-			error_msg, (unsigned) (offset%bh->b_size), offset,
-			le32_to_cpu(de->inode),
-			rlen, de->name_len);
-	return error_msg == NULL ? 1 : 0;
+	if (filp)
+		ext4_error_file(filp, function, line, bh ? bh->b_blocknr : 0,
+				"bad entry in directory: %s - offset=%u(%u), "
+				"inode=%u, rec_len=%d, name_len=%d",
+				error_msg, (unsigned) (offset%bh->b_size),
+				offset, le32_to_cpu(de->inode),
+				rlen, de->name_len);
+	else
+		ext4_error_inode(dir, function, line, bh ? bh->b_blocknr : 0,
+				"bad entry in directory: %s - offset=%u(%u), "
+				"inode=%u, rec_len=%d, name_len=%d",
+				error_msg, (unsigned) (offset%bh->b_size),
+				offset, le32_to_cpu(de->inode),
+				rlen, de->name_len);
+
+	return 1;
 }
 
 static int ext4_readdir(struct file *filp,
@@ -152,8 +167,9 @@ static int ext4_readdir(struct file *filp,
 		 */
 		if (!bh) {
 			if (!dir_has_error) {
-				EXT4_ERROR_INODE(inode, "directory "
-					   "contains a hole at offset %Lu",
+				EXT4_ERROR_FILE(filp, 0,
+						"directory contains a "
+						"hole at offset %llu",
 					   (unsigned long long) filp->f_pos);
 				dir_has_error = 1;
 			}
@@ -194,8 +210,8 @@ revalidate:
 		while (!error && filp->f_pos < inode->i_size
 		       && offset < sb->s_blocksize) {
 			de = (struct ext4_dir_entry_2 *) (bh->b_data + offset);
-			if (!ext4_check_dir_entry(inode, de,
-						  bh, offset)) {
+			if (ext4_check_dir_entry(inode, filp, de,
+						 bh, offset)) {
 				/*
 				 * On error, skip the f_pos to the next block
 				 */
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 94ce3d7a1c4b..bab2387fba43 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -62,8 +62,8 @@
 #define EXT4_ERROR_INODE_BLOCK(inode, block, fmt, a...)			\
 	ext4_error_inode((inode), __func__, __LINE__, (block), (fmt), ## a)
 
-#define EXT4_ERROR_FILE(file, fmt, a...)	\
-	ext4_error_file(__func__, __LINE__, (file), (fmt), ## a)
+#define EXT4_ERROR_FILE(file, block, fmt, a...)				\
+	ext4_error_file((file), __func__, __LINE__, (block), (fmt), ## a)
 
 /* data type for block offset of block group */
 typedef int ext4_grpblk_t;
@@ -561,22 +561,6 @@ struct ext4_new_group_data {
 #define EXT4_IOC32_SETVERSION_OLD	FS_IOC32_SETVERSION
 #endif
 
-
-/*
- *  Mount options
- */
-struct ext4_mount_options {
-	unsigned long s_mount_opt;
-	uid_t s_resuid;
-	gid_t s_resgid;
-	unsigned long s_commit_interval;
-	u32 s_min_batch_time, s_max_batch_time;
-#ifdef CONFIG_QUOTA
-	int s_jquota_fmt;
-	char *s_qf_names[MAXQUOTAS];
-#endif
-};
-
 /* Max physical block we can addres w/o extents */
 #define EXT4_MAX_BLOCK_FILE_PHYS	0xFFFFFFFF
 
@@ -709,6 +693,8 @@ do {									       \
 	if (EXT4_FITS_IN_INODE(raw_inode, EXT4_I(inode), xtime ## _extra))     \
 		ext4_decode_extra_time(&(inode)->xtime,			       \
 				       raw_inode->xtime ## _extra);	       \
+	else								       \
+		(inode)->xtime.tv_nsec = 0;				       \
 } while (0)
 
 #define EXT4_EINODE_GET_XTIME(xtime, einode, raw_inode)			       \
@@ -719,6 +705,8 @@ do {									       \
 	if (EXT4_FITS_IN_INODE(raw_inode, einode, xtime ## _extra))	       \
 		ext4_decode_extra_time(&(einode)->xtime,		       \
 				       raw_inode->xtime ## _extra);	       \
+	else								       \
+		(einode)->xtime.tv_nsec = 0;				       \
 } while (0)
 
 #define i_disk_version osd1.linux1.l_i_version
@@ -750,12 +738,13 @@ do {									       \
 
 /*
  * storage for cached extent
+ * If ec_len == 0, then the cache is invalid.
+ * If ec_start == 0, then the cache represents a gap (null mapping)
  */
 struct ext4_ext_cache {
 	ext4_fsblk_t	ec_start;
 	ext4_lblk_t	ec_block;
 	__u32		ec_len; /* must be 32bit to return holes */
-	__u32		ec_type;
 };
 
 /*
@@ -774,10 +763,12 @@ struct ext4_inode_info {
 	 * near to their parent directory's inode.
 	 */
 	ext4_group_t	i_block_group;
+	ext4_lblk_t	i_dir_start_lookup;
+#if (BITS_PER_LONG < 64)
 	unsigned long	i_state_flags;		/* Dynamic state flags */
+#endif
 	unsigned long	i_flags;
 
-	ext4_lblk_t		i_dir_start_lookup;
 #ifdef CONFIG_EXT4_FS_XATTR
 	/*
 	 * Extended attributes can be read independently of the main file
@@ -820,7 +811,7 @@ struct ext4_inode_info {
 	 */
 	struct rw_semaphore i_data_sem;
 	struct inode vfs_inode;
-	struct jbd2_inode jinode;
+	struct jbd2_inode *jinode;
 
 	struct ext4_ext_cache i_cached_extent;
 	/*
@@ -840,14 +831,12 @@ struct ext4_inode_info {
 	unsigned int i_reserved_data_blocks;
 	unsigned int i_reserved_meta_blocks;
 	unsigned int i_allocated_meta_blocks;
-	unsigned short i_delalloc_reserved_flag;
-	sector_t i_da_metadata_calc_last_lblock;
+	ext4_lblk_t i_da_metadata_calc_last_lblock;
 	int i_da_metadata_calc_len;
 
 	/* on-disk additional length */
 	__u16 i_extra_isize;
 
-	spinlock_t i_block_reservation_lock;
 #ifdef CONFIG_QUOTA
 	/* quota space reservation, managed internally by quota code */
 	qsize_t i_reserved_quota;
@@ -856,9 +845,11 @@ struct ext4_inode_info {
 	/* completed IOs that might need unwritten extents handling */
 	struct list_head i_completed_io_list;
 	spinlock_t i_completed_io_lock;
+	atomic_t i_ioend_count;	/* Number of outstanding io_end structs */
 	/* current io_end structure for async DIO write*/
 	ext4_io_end_t *cur_aio_dio;
-	atomic_t i_ioend_count;	/* Number of outstanding io_end structs */
+
+	spinlock_t i_block_reservation_lock;
 
 	/*
 	 * Transactions that contain inode's metadata needed to complete
@@ -917,11 +908,20 @@ struct ext4_inode_info {
 #define EXT4_MOUNT_DISCARD		0x40000000 /* Issue DISCARD requests */
 #define EXT4_MOUNT_INIT_INODE_TABLE	0x80000000 /* Initialize uninitialized itables */
 
-#define clear_opt(o, opt)		o &= ~EXT4_MOUNT_##opt
-#define set_opt(o, opt)			o |= EXT4_MOUNT_##opt
+#define clear_opt(sb, opt)		EXT4_SB(sb)->s_mount_opt &= \
+						~EXT4_MOUNT_##opt
+#define set_opt(sb, opt)		EXT4_SB(sb)->s_mount_opt |= \
+						EXT4_MOUNT_##opt
 #define test_opt(sb, opt)		(EXT4_SB(sb)->s_mount_opt & \
 					 EXT4_MOUNT_##opt)
 
+#define clear_opt2(sb, opt)		EXT4_SB(sb)->s_mount_opt2 &= \
+						~EXT4_MOUNT2_##opt
+#define set_opt2(sb, opt)		EXT4_SB(sb)->s_mount_opt2 |= \
+						EXT4_MOUNT2_##opt
+#define test_opt2(sb, opt)		(EXT4_SB(sb)->s_mount_opt2 & \
+					 EXT4_MOUNT2_##opt)
+
 #define ext4_set_bit			ext2_set_bit
 #define ext4_set_bit_atomic		ext2_set_bit_atomic
 #define ext4_clear_bit			ext2_clear_bit
@@ -1087,6 +1087,7 @@ struct ext4_sb_info {
 	struct ext4_super_block *s_es;	/* Pointer to the super block in the buffer */
 	struct buffer_head **s_group_desc;
 	unsigned int s_mount_opt;
+	unsigned int s_mount_opt2;
 	unsigned int s_mount_flags;
 	ext4_fsblk_t s_sb_block;
 	uid_t s_resuid;
@@ -1237,24 +1238,39 @@ enum {
 	EXT4_STATE_EXT_MIGRATE,		/* Inode is migrating */
 	EXT4_STATE_DIO_UNWRITTEN,	/* need convert on dio done*/
 	EXT4_STATE_NEWENTRY,		/* File just added to dir */
+	EXT4_STATE_DELALLOC_RESERVED,	/* blks already reserved for delalloc */
 };
 
-#define EXT4_INODE_BIT_FNS(name, field)					\
+#define EXT4_INODE_BIT_FNS(name, field, offset)				\
 static inline int ext4_test_inode_##name(struct inode *inode, int bit)	\
 {									\
-	return test_bit(bit, &EXT4_I(inode)->i_##field);		\
+	return test_bit(bit + (offset), &EXT4_I(inode)->i_##field);	\
 }									\
 static inline void ext4_set_inode_##name(struct inode *inode, int bit)	\
 {									\
-	set_bit(bit, &EXT4_I(inode)->i_##field);			\
+	set_bit(bit + (offset), &EXT4_I(inode)->i_##field);		\
 }									\
 static inline void ext4_clear_inode_##name(struct inode *inode, int bit) \
 {									\
-	clear_bit(bit, &EXT4_I(inode)->i_##field);			\
+	clear_bit(bit + (offset), &EXT4_I(inode)->i_##field);		\
 }
 
-EXT4_INODE_BIT_FNS(flag, flags)
-EXT4_INODE_BIT_FNS(state, state_flags)
+EXT4_INODE_BIT_FNS(flag, flags, 0)
+#if (BITS_PER_LONG < 64)
+EXT4_INODE_BIT_FNS(state, state_flags, 0)
+
+static inline void ext4_clear_state_flags(struct ext4_inode_info *ei)
+{
+	(ei)->i_state_flags = 0;
+}
+#else
+EXT4_INODE_BIT_FNS(state, flags, 32)
+
+static inline void ext4_clear_state_flags(struct ext4_inode_info *ei)
+{
+	/* We depend on the fact that callers will set i_flags */
+}
+#endif
 #else
 /* Assume that user mode programs are passing in an ext4fs superblock, not
  * a kernel struct super_block.  This will allow us to call the feature-test
@@ -1642,10 +1658,12 @@ extern unsigned ext4_init_block_bitmap(struct super_block *sb,
 
 /* dir.c */
 extern int __ext4_check_dir_entry(const char *, unsigned int, struct inode *,
+				  struct file *,
 				  struct ext4_dir_entry_2 *,
 				  struct buffer_head *, unsigned int);
-#define ext4_check_dir_entry(dir, de, bh, offset) \
-	__ext4_check_dir_entry(__func__, __LINE__, (dir), (de), (bh), (offset))
+#define ext4_check_dir_entry(dir, filp, de, bh, offset)			\
+	unlikely(__ext4_check_dir_entry(__func__, __LINE__, (dir), (filp), \
+					(de), (bh), (offset)))
 extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 				    __u32 minor_hash,
 				    struct ext4_dir_entry_2 *dirent);
@@ -1653,6 +1671,7 @@ extern void ext4_htree_free_dir_info(struct dir_private_info *p);
 
 /* fsync.c */
 extern int ext4_sync_file(struct file *, int);
+extern int ext4_flush_completed_IO(struct inode *);
 
 /* hash.c */
 extern int ext4fs_dirhash(const char *name, int len, struct
@@ -1752,8 +1771,8 @@ extern void ext4_error_inode(struct inode *, const char *, unsigned int,
 			     ext4_fsblk_t, const char *, ...)
 	__attribute__ ((format (printf, 5, 6)));
 extern void ext4_error_file(struct file *, const char *, unsigned int,
-			    const char *, ...)
-	__attribute__ ((format (printf, 4, 5)));
+			    ext4_fsblk_t, const char *, ...)
+	__attribute__ ((format (printf, 5, 6)));
 extern void __ext4_std_error(struct super_block *, const char *,
 			     unsigned int, int);
 extern void __ext4_abort(struct super_block *, const char *, unsigned int,
diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h
index 28ce70fd9cd0..2e29abb30f76 100644
--- a/fs/ext4/ext4_extents.h
+++ b/fs/ext4/ext4_extents.h
@@ -119,10 +119,6 @@ struct ext4_ext_path {
  * structure for external API
  */
 
-#define EXT4_EXT_CACHE_NO	0
-#define EXT4_EXT_CACHE_GAP	1
-#define EXT4_EXT_CACHE_EXTENT	2
-
 /*
  * to be called by ext4_ext_walk_space()
  * negative retcode - error
@@ -197,7 +193,7 @@ static inline unsigned short ext_depth(struct inode *inode)
 static inline void
 ext4_ext_invalidate_cache(struct inode *inode)
 {
-	EXT4_I(inode)->i_cached_extent.ec_type = EXT4_EXT_CACHE_NO;
+	EXT4_I(inode)->i_cached_extent.ec_len = 0;
 }
 
 static inline void ext4_ext_mark_uninitialized(struct ext4_extent *ext)
@@ -278,7 +274,7 @@ static inline void ext4_idx_store_pblock(struct ext4_extent_idx *ix,
 }
 
 extern int ext4_ext_calc_metadata_amount(struct inode *inode,
-					 sector_t lblocks);
+					 ext4_lblk_t lblocks);
 extern int ext4_extent_tree_init(handle_t *, struct inode *);
 extern int ext4_ext_calc_credits_for_single_extent(struct inode *inode,
 						   int num,
diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h
index b0bd792c58c5..d8b992e658c1 100644
--- a/fs/ext4/ext4_jbd2.h
+++ b/fs/ext4/ext4_jbd2.h
@@ -253,7 +253,7 @@ static inline int ext4_journal_force_commit(journal_t *journal)
 static inline int ext4_jbd2_file_inode(handle_t *handle, struct inode *inode)
 {
 	if (ext4_handle_valid(handle))
-		return jbd2_journal_file_inode(handle, &EXT4_I(inode)->jinode);
+		return jbd2_journal_file_inode(handle, EXT4_I(inode)->jinode);
 	return 0;
 }
 
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 0554c48cb1fd..e910720e8bb8 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -117,11 +117,33 @@ static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode,
 		struct ext4_extent *ex;
 		depth = path->p_depth;
 
-		/* try to predict block placement */
+		/*
+		 * Try to predict block placement assuming that we are
+		 * filling in a file which will eventually be
+		 * non-sparse --- i.e., in the case of libbfd writing
+		 * an ELF object sections out-of-order but in a way
+		 * the eventually results in a contiguous object or
+		 * executable file, or some database extending a table
+		 * space file.  However, this is actually somewhat
+		 * non-ideal if we are writing a sparse file such as
+		 * qemu or KVM writing a raw image file that is going
+		 * to stay fairly sparse, since it will end up
+		 * fragmenting the file system's free space.  Maybe we
+		 * should have some hueristics or some way to allow
+		 * userspace to pass a hint to file system,
+		 * especiially if the latter case turns out to be
+		 * common.
+		 */
 		ex = path[depth].p_ext;
-		if (ex)
-			return (ext4_ext_pblock(ex) +
-				(block - le32_to_cpu(ex->ee_block)));
+		if (ex) {
+			ext4_fsblk_t ext_pblk = ext4_ext_pblock(ex);
+			ext4_lblk_t ext_block = le32_to_cpu(ex->ee_block);
+
+			if (block > ext_block)
+				return ext_pblk + (block - ext_block);
+			else
+				return ext_pblk - (ext_block - block);
+		}
 
 		/* it looks like index is empty;
 		 * try to find starting block from index itself */
@@ -244,7 +266,7 @@ static inline int ext4_ext_space_root_idx(struct inode *inode, int check)
  * to allocate @blocks
  * Worse case is one block per extent
  */
-int ext4_ext_calc_metadata_amount(struct inode *inode, sector_t lblock)
+int ext4_ext_calc_metadata_amount(struct inode *inode, ext4_lblk_t lblock)
 {
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	int idxs, num = 0;
@@ -1872,12 +1894,10 @@ static int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block,
 			cbex.ec_block = start;
 			cbex.ec_len = end - start;
 			cbex.ec_start = 0;
-			cbex.ec_type = EXT4_EXT_CACHE_GAP;
 		} else {
 			cbex.ec_block = le32_to_cpu(ex->ee_block);
 			cbex.ec_len = ext4_ext_get_actual_len(ex);
 			cbex.ec_start = ext4_ext_pblock(ex);
-			cbex.ec_type = EXT4_EXT_CACHE_EXTENT;
 		}
 
 		if (unlikely(cbex.ec_len == 0)) {
@@ -1917,13 +1937,12 @@ static int ext4_ext_walk_space(struct inode *inode, ext4_lblk_t block,
 
 static void
 ext4_ext_put_in_cache(struct inode *inode, ext4_lblk_t block,
-			__u32 len, ext4_fsblk_t start, int type)
+			__u32 len, ext4_fsblk_t start)
 {
 	struct ext4_ext_cache *cex;
 	BUG_ON(len == 0);
 	spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
 	cex = &EXT4_I(inode)->i_cached_extent;
-	cex->ec_type = type;
 	cex->ec_block = block;
 	cex->ec_len = len;
 	cex->ec_start = start;
@@ -1976,15 +1995,18 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path,
 	}
 
 	ext_debug(" -> %u:%lu\n", lblock, len);
-	ext4_ext_put_in_cache(inode, lblock, len, 0, EXT4_EXT_CACHE_GAP);
+	ext4_ext_put_in_cache(inode, lblock, len, 0);
 }
 
+/*
+ * Return 0 if cache is invalid; 1 if the cache is valid
+ */
 static int
 ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block,
 			struct ext4_extent *ex)
 {
 	struct ext4_ext_cache *cex;
-	int ret = EXT4_EXT_CACHE_NO;
+	int ret = 0;
 
 	/*
 	 * We borrow i_block_reservation_lock to protect i_cached_extent
@@ -1993,11 +2015,9 @@ ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block,
 	cex = &EXT4_I(inode)->i_cached_extent;
 
 	/* has cache valid data? */
-	if (cex->ec_type == EXT4_EXT_CACHE_NO)
+	if (cex->ec_len == 0)
 		goto errout;
 
-	BUG_ON(cex->ec_type != EXT4_EXT_CACHE_GAP &&
-			cex->ec_type != EXT4_EXT_CACHE_EXTENT);
 	if (in_range(block, cex->ec_block, cex->ec_len)) {
 		ex->ee_block = cpu_to_le32(cex->ec_block);
 		ext4_ext_store_pblock(ex, cex->ec_start);
@@ -2005,7 +2025,7 @@ ext4_ext_in_cache(struct inode *inode, ext4_lblk_t block,
 		ext_debug("%u cached by %u:%u:%llu\n",
 				block,
 				cex->ec_block, cex->ec_len, cex->ec_start);
-		ret = cex->ec_type;
+		ret = 1;
 	}
 errout:
 	spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
@@ -3082,7 +3102,7 @@ static void unmap_underlying_metadata_blocks(struct block_device *bdev,
  * Handle EOFBLOCKS_FL flag, clearing it if necessary
  */
 static int check_eofblocks_fl(handle_t *handle, struct inode *inode,
-			      struct ext4_map_blocks *map,
+			      ext4_lblk_t lblk,
 			      struct ext4_ext_path *path,
 			      unsigned int len)
 {
@@ -3112,7 +3132,7 @@ static int check_eofblocks_fl(handle_t *handle, struct inode *inode,
 	 * this turns out to be false, we can bail out from this
 	 * function immediately.
 	 */
-	if (map->m_lblk + len < le32_to_cpu(last_ex->ee_block) +
+	if (lblk + len < le32_to_cpu(last_ex->ee_block) +
 	    ext4_ext_get_actual_len(last_ex))
 		return 0;
 	/*
@@ -3168,8 +3188,8 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
 							path);
 		if (ret >= 0) {
 			ext4_update_inode_fsync_trans(handle, inode, 1);
-			err = check_eofblocks_fl(handle, inode, map, path,
-						 map->m_len);
+			err = check_eofblocks_fl(handle, inode, map->m_lblk,
+						 path, map->m_len);
 		} else
 			err = ret;
 		goto out2;
@@ -3199,7 +3219,8 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
 	ret = ext4_ext_convert_to_initialized(handle, inode, map, path);
 	if (ret >= 0) {
 		ext4_update_inode_fsync_trans(handle, inode, 1);
-		err = check_eofblocks_fl(handle, inode, map, path, map->m_len);
+		err = check_eofblocks_fl(handle, inode, map->m_lblk, path,
+					 map->m_len);
 		if (err < 0)
 			goto out2;
 	}
@@ -3276,7 +3297,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 	struct ext4_extent_header *eh;
 	struct ext4_extent newex, *ex;
 	ext4_fsblk_t newblock;
-	int err = 0, depth, ret, cache_type;
+	int err = 0, depth, ret;
 	unsigned int allocated = 0;
 	struct ext4_allocation_request ar;
 	ext4_io_end_t *io = EXT4_I(inode)->cur_aio_dio;
@@ -3285,9 +3306,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 		  map->m_lblk, map->m_len, inode->i_ino);
 
 	/* check in cache */
-	cache_type = ext4_ext_in_cache(inode, map->m_lblk, &newex);
-	if (cache_type) {
-		if (cache_type == EXT4_EXT_CACHE_GAP) {
+	if (ext4_ext_in_cache(inode, map->m_lblk, &newex)) {
+		if (!newex.ee_start_lo && !newex.ee_start_hi) {
 			if ((flags & EXT4_GET_BLOCKS_CREATE) == 0) {
 				/*
 				 * block isn't allocated yet and
@@ -3296,7 +3316,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 				goto out2;
 			}
 			/* we should allocate requested block */
-		} else if (cache_type == EXT4_EXT_CACHE_EXTENT) {
+		} else {
 			/* block is already allocated */
 			newblock = map->m_lblk
 				   - le32_to_cpu(newex.ee_block)
@@ -3305,8 +3325,6 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			allocated = ext4_ext_get_actual_len(&newex) -
 				(map->m_lblk - le32_to_cpu(newex.ee_block));
 			goto out;
-		} else {
-			BUG();
 		}
 	}
 
@@ -3357,8 +3375,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			/* Do not put uninitialized extent in the cache */
 			if (!ext4_ext_is_uninitialized(ex)) {
 				ext4_ext_put_in_cache(inode, ee_block,
-							ee_len, ee_start,
-							EXT4_EXT_CACHE_EXTENT);
+							ee_len, ee_start);
 				goto out;
 			}
 			ret = ext4_ext_handle_uninitialized_extents(handle,
@@ -3456,7 +3473,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			map->m_flags |= EXT4_MAP_UNINIT;
 	}
 
-	err = check_eofblocks_fl(handle, inode, map, path, ar.len);
+	err = check_eofblocks_fl(handle, inode, map->m_lblk, path, ar.len);
 	if (err)
 		goto out2;
 
@@ -3490,8 +3507,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 	 * when it is _not_ an uninitialized extent.
 	 */
 	if ((flags & EXT4_GET_BLOCKS_UNINIT_EXT) == 0) {
-		ext4_ext_put_in_cache(inode, map->m_lblk, allocated, newblock,
-						EXT4_EXT_CACHE_EXTENT);
+		ext4_ext_put_in_cache(inode, map->m_lblk, allocated, newblock);
 		ext4_update_inode_fsync_trans(handle, inode, 1);
 	} else
 		ext4_update_inode_fsync_trans(handle, inode, 0);
@@ -3519,6 +3535,12 @@ void ext4_ext_truncate(struct inode *inode)
 	int err = 0;
 
 	/*
+	 * finish any pending end_io work so we won't run the risk of
+	 * converting any truncated blocks to initialized later
+	 */
+	ext4_flush_completed_IO(inode);
+
+	/*
 	 * probably first extent we're gonna free will be last in block
 	 */
 	err = ext4_writepage_trans_blocks(inode);
@@ -3767,7 +3789,7 @@ static int ext4_ext_fiemap_cb(struct inode *inode, struct ext4_ext_path *path,
 
 	logical =  (__u64)newex->ec_block << blksize_bits;
 
-	if (newex->ec_type == EXT4_EXT_CACHE_GAP) {
+	if (newex->ec_start == 0) {
 		pgoff_t offset;
 		struct page *page;
 		struct buffer_head *bh = NULL;
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 5a5c55ddceef..bb003dc9ffff 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -104,6 +104,7 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
 {
 	struct super_block *sb = inode->i_sb;
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
+	struct ext4_inode_info *ei = EXT4_I(inode);
 	struct vfsmount *mnt = filp->f_path.mnt;
 	struct path path;
 	char buf[64], *cp;
@@ -127,6 +128,27 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
 			ext4_mark_super_dirty(sb);
 		}
 	}
+	/*
+	 * Set up the jbd2_inode if we are opening the inode for
+	 * writing and the journal is present
+	 */
+	if (sbi->s_journal && !ei->jinode && (filp->f_mode & FMODE_WRITE)) {
+		struct jbd2_inode *jinode = jbd2_alloc_inode(GFP_KERNEL);
+
+		spin_lock(&inode->i_lock);
+		if (!ei->jinode) {
+			if (!jinode) {
+				spin_unlock(&inode->i_lock);
+				return -ENOMEM;
+			}
+			ei->jinode = jinode;
+			jbd2_journal_init_jbd_inode(ei->jinode, inode);
+			jinode = NULL;
+		}
+		spin_unlock(&inode->i_lock);
+		if (unlikely(jinode != NULL))
+			jbd2_free_inode(jinode);
+	}
 	return dquot_file_open(inode, filp);
 }
 
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c
index c1a7bc923cf6..7829b287822a 100644
--- a/fs/ext4/fsync.c
+++ b/fs/ext4/fsync.c
@@ -75,7 +75,7 @@ static void dump_completed_IO(struct inode * inode)
  * to written.
  * The function return the number of pending IOs on success.
  */
-static int flush_completed_IO(struct inode *inode)
+extern int ext4_flush_completed_IO(struct inode *inode)
 {
 	ext4_io_end_t *io;
 	struct ext4_inode_info *ei = EXT4_I(inode);
@@ -169,7 +169,7 @@ int ext4_sync_file(struct file *file, int datasync)
 	if (inode->i_sb->s_flags & MS_RDONLY)
 		return 0;
 
-	ret = flush_completed_IO(inode);
+	ret = ext4_flush_completed_IO(inode);
 	if (ret < 0)
 		return ret;
 
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 1ce240a23ebb..eb9097aec6f0 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -1027,7 +1027,7 @@ got:
 	inode->i_generation = sbi->s_next_generation++;
 	spin_unlock(&sbi->s_next_gen_lock);
 
-	ei->i_state_flags = 0;
+	ext4_clear_state_flags(ei); /* Only relevant on 32-bit archs */
 	ext4_set_inode_state(inode, EXT4_STATE_NEW);
 
 	ei->i_extra_isize = EXT4_SB(sb)->s_want_extra_isize;
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index e659597b690b..e80fc513eacc 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -40,6 +40,7 @@
 #include <linux/workqueue.h>
 #include <linux/kernel.h>
 #include <linux/slab.h>
+#include <linux/ratelimit.h>
 
 #include "ext4_jbd2.h"
 #include "xattr.h"
@@ -54,10 +55,17 @@ static inline int ext4_begin_ordered_truncate(struct inode *inode,
 					      loff_t new_size)
 {
 	trace_ext4_begin_ordered_truncate(inode, new_size);
-	return jbd2_journal_begin_ordered_truncate(
-					EXT4_SB(inode->i_sb)->s_journal,
-					&EXT4_I(inode)->jinode,
-					new_size);
+	/*
+	 * If jinode is zero, then we never opened the file for
+	 * writing, so there's no need to call
+	 * jbd2_journal_begin_ordered_truncate() since there's no
+	 * outstanding writes we need to flush.
+	 */
+	if (!EXT4_I(inode)->jinode)
+		return 0;
+	return jbd2_journal_begin_ordered_truncate(EXT4_JOURNAL(inode),
+						   EXT4_I(inode)->jinode,
+						   new_size);
 }
 
 static void ext4_invalidatepage(struct page *page, unsigned long offset);
@@ -552,7 +560,7 @@ static ext4_fsblk_t ext4_find_goal(struct inode *inode, ext4_lblk_t block,
 }
 
 /**
- *	ext4_blks_to_allocate: Look up the block map and count the number
+ *	ext4_blks_to_allocate - Look up the block map and count the number
  *	of direct blocks need to be allocated for the given branch.
  *
  *	@branch: chain of indirect blocks
@@ -591,13 +599,19 @@ static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned int blks,
 
 /**
  *	ext4_alloc_blocks: multiple allocate blocks needed for a branch
+ *	@handle: handle for this transaction
+ *	@inode: inode which needs allocated blocks
+ *	@iblock: the logical block to start allocated at
+ *	@goal: preferred physical block of allocation
  *	@indirect_blks: the number of blocks need to allocate for indirect
  *			blocks
- *
+ *	@blks: number of desired blocks
  *	@new_blocks: on return it will store the new block numbers for
  *	the indirect blocks(if needed) and the first direct block,
- *	@blks:	on return it will store the total number of allocated
- *		direct blocks
+ *	@err: on return it will store the error code
+ *
+ *	This function will return the number of blocks allocated as
+ *	requested by the passed-in parameters.
  */
 static int ext4_alloc_blocks(handle_t *handle, struct inode *inode,
 			     ext4_lblk_t iblock, ext4_fsblk_t goal,
@@ -711,9 +725,11 @@ failed_out:
 
 /**