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
  ...

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

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

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);

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 @@ static int ext4_readdir(struct file *filp,
 		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
 				 */

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,

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,

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;
 }
 

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);
@@ -3518,6 +3534,12 @@ void ext4_ext_truncate(struct inode *inode)
 	handle_t *handle;
 	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
 	 */
@@ -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;

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);
 }
 

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;
 

fs/ext4/ialloc.c

@@ -1027,7 +1027,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode,
 	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;

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 @@ static int ext4_alloc_blocks(handle_t *handle, struct inode *inode,
 
 /**
  *	ext4_alloc_branch - allocate and set up a chain of blocks.
+ *	@handle: handle for this transaction
  *	@inode: owner
  *	@indirect_blks: number of allocated indirect blocks
  *	@blks: number of allocated direct blocks
+ *	@goal: preferred place for allocation
  *	@offsets: offsets (in the blocks) to store the pointers to next.
  *	@branch: place to store the chain in.
  *
@@ -826,6 +842,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
 
 /**
  * ext4_splice_branch - splice the allocated branch onto inode.
+ * @handle: handle for this transaction
  * @inode: owner
  * @block: (logical) number of block we are adding
  * @chain: chain of indirect blocks (with a missing link - see
@@ -1081,7 +1098,7 @@ static int ext4_indirect_calc_metadata_amount(struct inode *inode,
  * Calculate the number of metadata blocks need to reserve
  * to allocate a block located at @lblock
  */
-static int ext4_calc_metadata_amount(struct inode *inode, sector_t lblock)
+static int ext4_calc_metadata_amount(struct inode *inode, ext4_lblk_t lblock)
 {
 	if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
 		return ext4_ext_calc_metadata_amount(inode, lblock);
@@ -1320,7 +1337,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 	 * avoid double accounting
 	 */
 	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
-		EXT4_I(inode)->i_delalloc_reserved_flag = 1;
+		ext4_set_inode_state(inode, EXT4_STATE_DELALLOC_RESERVED);
 	/*
 	 * We need to check for EXT4 here because migrate
 	 * could have changed the inode type in between
@@ -1350,7 +1367,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 			ext4_da_update_reserve_space(inode, retval, 1);
 	}
 	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
-		EXT4_I(inode)->i_delalloc_reserved_flag = 0;
+		ext4_clear_inode_state(inode, EXT4_STATE_DELALLOC_RESERVED);
 
 	up_write((&EXT4_I(inode)->i_data_sem));
 	if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) {
@@ -1878,7 +1895,7 @@ static int ext4_journalled_write_end(struct file *file,
 /*
  * Reserve a single block located at lblock
  */
-static int ext4_da_reserve_space(struct inode *inode, sector_t lblock)
+static int ext4_da_reserve_space(struct inode *inode, ext4_lblk_t lblock)
 {
 	int retries = 0;
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
@@ -2239,7 +2256,7 @@ static void mpage_da_map_and_submit(struct mpage_da_data *mpd)
 	 * affects functions in many different parts of the allocation
 	 * call path.  This flag exists primarily because we don't
 	 * want to change *many* call functions, so ext4_map_blocks()
-	 * will set the magic i_delalloc_reserved_flag once the
+	 * will set the EXT4_STATE_DELALLOC_RESERVED flag once the
 	 * inode's allocation semaphore is taken.
 	 *
 	 * If the blocks in questions were delalloc blocks, set
@@ -3720,8 +3737,7 @@ static int ext4_set_bh_endio(struct buffer_head *bh, struct inode *inode)
 retry:
 	io_end = ext4_init_io_end(inode, GFP_ATOMIC);
 	if (!io_end) {
-		if (printk_ratelimit())
-			printk(KERN_WARNING "%s: allocation fail\n", __func__);
+		pr_warning_ratelimited("%s: allocation fail\n", __func__);
 		schedule();
 		goto retry;
 	}
@@ -4045,7 +4061,7 @@ int ext4_block_truncate_page(handle_t *handle,
 	if (ext4_should_journal_data(inode)) {
 		err = ext4_handle_dirty_metadata(handle, inode, bh);
 	} else {
-		if (ext4_should_order_data(inode))
+		if (ext4_should_order_data(inode) && EXT4_I(inode)->jinode)
 			err = ext4_jbd2_file_inode(handle, inode);
 		mark_buffer_dirty(bh);
 	}
@@ -4169,6 +4185,7 @@ static int ext4_clear_blocks(handle_t *handle, struct inode *inode,
 {
 	__le32 *p;
 	int	flags = EXT4_FREE_BLOCKS_FORGET | EXT4_FREE_BLOCKS_VALIDATED;
+	int	err;
 
 	if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
 		flags |= EXT4_FREE_BLOCKS_METADATA;
@@ -4184,11 +4201,23 @@ static int ext4_clear_blocks(handle_t *handle, struct inode *inode,
 	if (try_to_extend_transaction(handle, inode)) {
 		if (bh) {
 			BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
-			ext4_handle_dirty_metadata(handle, inode, bh);
+			err = ext4_handle_dirty_metadata(handle, inode, bh);
+			if (unlikely(err)) {
+				ext4_std_error(inode->i_sb, err);
+				return 1;
+			}
+		}
+		err = ext4_mark_inode_dirty(handle, inode);
+		if (unlikely(err)) {
+			ext4_std_error(inode->i_sb, err);
+			return 1;
+		}
+		err = ext4_truncate_restart_trans(handle, inode,
+						  blocks_for_truncate(inode));
+		if (unlikely(err)) {
+			ext4_std_error(inode->i_sb, err);
+			return 1;
 		}
-		ext4_mark_inode_dirty(handle, inode);
-		ext4_truncate_restart_trans(handle, inode,
-					    blocks_for_truncate(inode));
 		if (bh) {
 			BUFFER_TRACE(bh, "retaking write access");
 			ext4_journal_get_write_access(handle, bh);
@@ -4349,6 +4378,7 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 					(__le32 *) bh->b_data,
 					(__le32 *) bh->b_data + addr_per_block,
 					depth);
+			brelse(bh);
 
 			/*
 			 * Everything below this this pointer has been
@@ -4859,7 +4889,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	}
 	inode->i_nlink = le16_to_cpu(raw_inode->i_links_count);
 
-	ei->i_state_flags = 0;
+	ext4_clear_state_flags(ei);	/* Only relevant on 32-bit archs */
 	ei->i_dir_start_lookup = 0;
 	ei->i_dtime = le32_to_cpu(raw_inode->i_dtime);
 	/* We now have enough fields to check if the inode was active or not.
@@ -5118,7 +5148,7 @@ static int ext4_do_update_inode(handle_t *handle,
 	if (ext4_inode_blocks_set(handle, raw_inode, ei))
 		goto out_brelse;
 	raw_inode->i_dtime = cpu_to_le32(ei->i_dtime);
-	raw_inode->i_flags = cpu_to_le32(ei->i_flags);
+	raw_inode->i_flags = cpu_to_le32(ei->i_flags & 0xFFFFFFFF);
 	if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
 	    cpu_to_le32(EXT4_OS_HURD))
 		raw_inode->i_file_acl_high =

fs/ext4/mballoc.c

@@ -2608,18 +2608,12 @@ int ext4_mb_release(struct super_block *sb)
 static inline int ext4_issue_discard(struct super_block *sb,
 		ext4_group_t block_group, ext4_grpblk_t block, int count)
 {
-	int ret;
 	ext4_fsblk_t discard_block;
 
 	discard_block = block + ext4_group_first_block_no(sb, block_group);
 	trace_ext4_discard_blocks(sb,
 			(unsigned long long) discard_block, count);
-	ret = sb_issue_discard(sb, discard_block, count, GFP_NOFS, 0);
-	if (ret == -EOPNOTSUPP) {
-		ext4_warning(sb, "discard not supported, disabling");
-		clear_opt(EXT4_SB(sb)->s_mount_opt, DISCARD);
-	}
-	return ret;
+	return sb_issue_discard(sb, discard_block, count, GFP_NOFS, 0);
 }
 
 /*
@@ -2631,7 +2625,7 @@ static void release_blocks_on_commit(journal_t *journal, transaction_t *txn)
 	struct super_block *sb = journal->j_private;
 	struct ext4_buddy e4b;
 	struct ext4_group_info *db;
-	int err, count = 0, count2 = 0;
+	int err, ret, count = 0, count2 = 0;
 	struct ext4_free_data *entry;
 	struct list_head *l, *ltmp;
 
@@ -2641,9 +2635,15 @@ static void release_blocks_on_commit(journal_t *journal, transaction_t *txn)
 		mb_debug(1, "gonna free %u blocks in group %u (0x%p):",
 			 entry->count, entry->group, entry);
 
-		if (test_opt(sb, DISCARD))
-			ext4_issue_discard(sb, entry->group,
+		if (test_opt(sb, DISCARD)) {
+			ret = ext4_issue_discard(sb, entry->group,
 					entry->start_blk, entry->count);
+			if (unlikely(ret == -EOPNOTSUPP)) {
+				ext4_warning(sb, "discard not supported, "
+						 "disabling");
+				clear_opt(sb, DISCARD);
+			}
+		}
 
 		err = ext4_mb_load_buddy(sb, entry->group, &e4b);
 		/* we expect to find existing buddy because it's pinned */
@@ -3881,19 +3881,6 @@ void ext4_discard_preallocations(struct inode *inode)
 	}
 }
 
-/*
- * finds all preallocated spaces and return blocks being freed to them
- * if preallocated space becomes full (no block is used from the space)
- * then the function frees space in buddy
- * XXX: at the moment, truncate (which is the only way to free blocks)
- * discards all preallocations
- */
-static void ext4_mb_return_to_preallocation(struct inode *inode,
-					struct ext4_buddy *e4b,
-					sector_t block, int count)
-{
-	BUG_ON(!list_empty(&EXT4_I(inode)->i_prealloc_list));
-}
 #ifdef CONFIG_EXT4_DEBUG
 static void ext4_mb_show_ac(struct ext4_allocation_context *ac)
 {
@@ -4283,7 +4270,7 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 	 * EDQUOT check, as blocks and quotas have been already
 	 * reserved when data being copied into pagecache.
 	 */
-	if (EXT4_I(ar->inode)->i_delalloc_reserved_flag)
+	if (ext4_test_inode_state(ar->inode, EXT4_STATE_DELALLOC_RESERVED))
 		ar->flags |= EXT4_MB_DELALLOC_RESERVED;
 	else {
 		/* Without delayed allocation we need to verify
@@ -4380,7 +4367,8 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 	if (inquota && ar->len < inquota)
 		dquot_free_block(ar->inode, inquota - ar->len);
 	if (!ar->len) {
-		if (!EXT4_I(ar->inode)->i_delalloc_reserved_flag)
+		if (!ext4_test_inode_state(ar->inode,
+					   EXT4_STATE_DELALLOC_RESERVED))
 			/* release all the reserved blocks if non delalloc */
 			percpu_counter_sub(&sbi->s_dirtyblocks_counter,
 						reserv_blks);
@@ -4626,7 +4614,11 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 		 * blocks being freed are metadata. these blocks shouldn't
 		 * be used until this transaction is committed
 		 */
-		new_entry  = kmem_cache_alloc(ext4_free_ext_cachep, GFP_NOFS);
+		new_entry = kmem_cache_alloc(ext4_free_ext_cachep, GFP_NOFS);
+		if (!new_entry) {
+			err = -ENOMEM;
+			goto error_return;
+		}
 		new_entry->start_blk = bit;
 		new_entry->group  = block_group;
 		new_entry->count = count;
@@ -4643,7 +4635,6 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 		ext4_lock_group(sb, block_group);
 		mb_clear_bits(bitmap_bh->b_data, bit, count);
 		mb_free_blocks(inode, &e4b, bit, count);
-		ext4_mb_return_to_preallocation(inode, &e4b, block, count);
 	}
 
 	ret = ext4_free_blks_count(sb, gdp) + count;
@@ -4718,8 +4709,6 @@ static int ext4_trim_extent(struct super_block *sb, int start, int count,
 	ext4_unlock_group(sb, group);
 
 	ret = ext4_issue_discard(sb, group, start, count);
-	if (ret)
-		ext4_std_error(sb, ret);
 
 	ext4_lock_group(sb, group);
 	mb_free_blocks(NULL, e4b, start, ex.fe_len);
@@ -4819,6 +4808,8 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
 	ext4_group_t group, ngroups = ext4_get_groups_count(sb);
 	ext4_grpblk_t cnt = 0, first_block, last_block;
 	uint64_t start, len, minlen, trimmed;
+	ext4_fsblk_t first_data_blk =
+			le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
 	int ret = 0;
 
 	start = range->start >> sb->s_blocksize_bits;
@@ -4828,6 +4819,10 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
 
 	if (unlikely(minlen > EXT4_BLOCKS_PER_GROUP(sb)))
 		return -EINVAL;
+	if (start < first_data_blk) {
+		len -= first_data_blk - start;
+		start = first_data_blk;
+	}
 
 	/* Determine first and last group to examine based on start and len */
 	ext4_get_group_no_and_offset(sb, (ext4_fsblk_t) start,
@@ -4851,7 +4846,7 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
 		if (len >= EXT4_BLOCKS_PER_GROUP(sb))
 			len -= (EXT4_BLOCKS_PER_GROUP(sb) - first_block);
 		else
-			last_block = len;
+			last_block = first_block + len;
 
 		if (e4b.bd_info->bb_free >= minlen) {
 			cnt = ext4_trim_all_free(sb, &e4b, first_block,

fs/ext4/migrate.c

@@ -496,7 +496,7 @@ int ext4_ext_migrate(struct inode *inode)
 	goal = (((inode->i_ino - 1) / EXT4_INODES_PER_GROUP(inode->i_sb)) *
 		EXT4_INODES_PER_GROUP(inode->i_sb)) + 1;
 	tmp_inode = ext4_new_inode(handle, inode->i_sb->s_root->d_inode,
-				   S_IFREG, 0, goal);
+				   S_IFREG, NULL, goal);
 	if (IS_ERR(tmp_inode)) {
 		retval = -ENOMEM;
 		ext4_journal_stop(handle);

fs/ext4/namei.c

@@ -581,9 +581,9 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 					   dir->i_sb->s_blocksize -
 					   EXT4_DIR_REC_LEN(0));
 	for (; de < top; de = ext4_next_entry(de, dir->i_sb->s_blocksize)) {
-		if (!ext4_check_dir_entry(dir, de, bh,
-					(block<<EXT4_BLOCK_SIZE_BITS(dir->i_sb))
-						+((char *)de - bh->b_data))) {
+		if (ext4_check_dir_entry(dir, NULL, de, bh,
+				(block<<EXT4_BLOCK_SIZE_BITS(dir->i_sb))
+					 + ((char *)de - bh->b_data))) {
 			/* On error, skip the f_pos to the next block. */
 			dir_file->f_pos = (dir_file->f_pos |
 					(dir->i_sb->s_blocksize - 1)) + 1;
@@ -820,7 +820,7 @@ static inline int search_dirblock(struct buffer_head *bh,
 		if ((char *) de + namelen <= dlimit &&
 		    ext4_match (namelen, name, de)) {
 			/* found a match - just to be sure, do a full check */
-			if (!ext4_check_dir_entry(dir, de, bh, offset))
+			if (ext4_check_dir_entry(dir, NULL, de, bh, offset))
 				return -1;
 			*res_dir = de;
 			return 1;
@@ -1036,7 +1036,7 @@ static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, stru
 			return ERR_PTR(-EIO);
 		}
 		inode = ext4_iget(dir->i_sb, ino);
-		if (unlikely(IS_ERR(inode))) {
+		if (IS_ERR(inode)) {
 			if (PTR_ERR(inode) == -ESTALE) {
 				EXT4_ERROR_INODE(dir,
 						 "deleted inode referenced: %u",
@@ -1269,7 +1269,7 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry,
 		de = (struct ext4_dir_entry_2 *)bh->b_data;
 		top = bh->b_data + blocksize - reclen;
 		while ((char *) de <= top) {
-			if (!ext4_check_dir_entry(dir, de, bh, offset))
+			if (ext4_check_dir_entry(dir, NULL, de, bh, offset))
 				return -EIO;
 			if (ext4_match(namelen, name, de))
 				return -EEXIST;
@@ -1602,7 +1602,11 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 			if (err)
 				goto journal_error;
 		}
-		ext4_handle_dirty_metadata(handle, inode, frames[0].bh);
+		err = ext4_handle_dirty_metadata(handle, inode, frames[0].bh);
+		if (err) {
+			ext4_std_error(inode->i_sb, err);
+			goto cleanup;
+		}
 	}
 	de = do_split(handle, dir, &bh, frame, &hinfo, &err);
 	if (!de)
@@ -1630,17 +1634,21 @@ static int ext4_delete_entry(handle_t *handle,
 {
 	struct ext4_dir_entry_2 *de, *pde;
 	unsigned int blocksize = dir->i_sb->s_blocksize;
-	int i;
+	int i, err;
 
 	i = 0;
 	pde = NULL;
 	de = (struct ext4_dir_entry_2 *) bh->b_data;
 	while (i < bh->b_size) {
-		if (!ext4_check_dir_entry(dir, de, bh, i))
+		if (ext4_check_dir_entry(dir, NULL, de, bh, i))
 			return -EIO;
 		if (de == de_del)  {
 			BUFFER_TRACE(bh, "get_write_access");
-			ext4_journal_get_write_access(handle, bh);
+			err = ext4_journal_get_write_access(handle, bh);
+			if (unlikely(err)) {
+				ext4_std_error(dir->i_sb, err);
+				return err;
+			}
 			if (pde)
 				pde->rec_len = ext4_rec_len_to_disk(
 					ext4_rec_len_from_disk(pde->rec_len,
@@ -1652,7 +1660,11 @@ static int ext4_delete_entry(handle_t *handle,
 				de->inode = 0;
 			dir->i_version++;
 			BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
-			ext4_handle_dirty_metadata(handle, dir, bh);
+			err = ext4_handle_dirty_metadata(handle, dir, bh);
+			if (unlikely(err)) {
+				ext4_std_error(dir->i_sb, err);
+				return err;
+			}
 			return 0;
 		}
 		i += ext4_rec_len_from_disk(de->rec_len, blocksize);
@@ -1789,7 +1801,7 @@ static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 {
 	handle_t *handle;
 	struct inode *inode;
-	struct buffer_head *dir_block;
+	struct buffer_head *dir_block = NULL;
 	struct ext4_dir_entry_2 *de;
 	unsigned int blocksize = dir->i_sb->s_blocksize;
 	int err, retries = 0;
@@ -1822,7 +1834,9 @@ static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	if (!dir_block)
 		goto out_clear_inode;
 	BUFFER_TRACE(dir_block, "get_write_access");
-	ext4_journal_get_write_access(handle, dir_block);
+	err = ext4_journal_get_write_access(handle, dir_block);
+	if (err)
+		goto out_clear_inode;
 	de = (struct ext4_dir_entry_2 *) dir_block->b_data;
 	de->inode = cpu_to_le32(inode->i_ino);
 	de->name_len = 1;
@@ -1839,10 +1853,12 @@ static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	ext4_set_de_type(dir->i_sb, de, S_IFDIR);
 	inode->i_nlink = 2;
 	BUFFER_TRACE(dir_block, "call ext4_handle_dirty_metadata");
-	ext4_handle_dirty_metadata(handle, dir, dir_block);
-	brelse(dir_block);
-	ext4_mark_inode_dirty(handle, inode);
-	err = ext4_add_entry(handle, dentry, inode);
+	err = ext4_handle_dirty_metadata(handle, dir, dir_block);
+	if (err)
+		goto out_clear_inode;
+	err = ext4_mark_inode_dirty(handle, inode);
+	if (!err)
+		err = ext4_add_entry(handle, dentry, inode);
 	if (err) {
 out_clear_inode:
 		clear_nlink(inode);
@@ -1853,10 +1869,13 @@ static int ext4_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	}
 	ext4_inc_count(handle, dir);
 	ext4_update_dx_flag(dir);
-	ext4_mark_inode_dirty(handle, dir);
+	err = ext4_mark_inode_dirty(handle, dir);
+	if (err)
+		goto out_clear_inode;
 	d_instantiate(dentry, inode);
 	unlock_new_inode(inode);
 out_stop:
+	brelse(dir_block);
 	ext4_journal_stop(handle);
 	if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries))
 		goto retry;
@@ -1919,7 +1938,7 @@ static int empty_dir(struct inode *inode)
 			}
 			de = (struct ext4_dir_entry_2 *) bh->b_data;
 		}
-		if (!ext4_check_dir_entry(inode, de, bh, offset)) {
+		if (ext4_check_dir_entry(inode, NULL, de, bh, offset)) {
 			de = (struct ext4_dir_entry_2 *)(bh->b_data +
 							 sb->s_blocksize);
 			offset = (offset | (sb->s_blocksize - 1)) + 1;
@@ -2407,7 +2426,11 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
 					ext4_current_time(new_dir);
 		ext4_mark_inode_dirty(handle, new_dir);
 		BUFFER_TRACE(new_bh, "call ext4_handle_dirty_metadata");
-		ext4_handle_dirty_metadata(handle, new_dir, new_bh);
+		retval = ext4_handle_dirty_metadata(handle, new_dir, new_bh);
+		if (unlikely(retval)) {
+			ext4_std_error(new_dir->i_sb, retval);
+			goto end_rename;
+		}
 		brelse(new_bh);
 		new_bh = NULL;
 	}
@@ -2459,7 +2482,11 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry,
 		PARENT_INO(dir_bh->b_data, new_dir->i_sb->s_blocksize) =
 						cpu_to_le32(new_dir->i_ino);
 		BUFFER_TRACE(dir_bh, "call ext4_handle_dirty_metadata");
-		ext4_handle_dirty_metadata(handle, old_dir, dir_bh);
+		retval = ext4_handle_dirty_metadata(handle, old_dir, dir_bh);
+		if (retval) {
+			ext4_std_error(old_dir->i_sb, retval);
+			goto end_rename;
+		}
 		ext4_dec_count(handle, old_dir);
 		if (new_inode) {
 			/* checked empty_dir above, can't have another parent,

fs/ext4/page-io.c

@@ -44,7 +44,7 @@ int __init ext4_init_pageio(void)
 	if (io_page_cachep == NULL)
 		return -ENOMEM;
 	io_end_cachep = KMEM_CACHE(ext4_io_end, SLAB_RECLAIM_ACCOUNT);
-	if (io_page_cachep == NULL) {
+	if (io_end_cachep == NULL) {
 		kmem_cache_destroy(io_page_cachep);
 		return -ENOMEM;
 	}
@@ -158,11 +158,8 @@ static void ext4_end_io_work(struct work_struct *work)
 
 ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags)
 {
-	ext4_io_end_t *io = NULL;
-
-	io = kmem_cache_alloc(io_end_cachep, flags);
+	ext4_io_end_t *io = kmem_cache_zalloc(io_end_cachep, flags);
 	if (io) {
-		memset(io, 0, sizeof(*io));
 		atomic_inc(&EXT4_I(inode)->i_ioend_count);
 		io->inode = inode;
 		INIT_WORK(&io->work, ext4_end_io_work);

fs/ext4/resize.c

@@ -220,7 +220,11 @@ static int setup_new_group_blocks(struct super_block *sb,
 		memcpy(gdb->b_data, sbi->s_group_desc[i]->b_data, gdb->b_size);
 		set_buffer_uptodate(gdb);
 		unlock_buffer(gdb);
-		ext4_handle_dirty_metadata(handle, NULL, gdb);
+		err = ext4_handle_dirty_metadata(handle, NULL, gdb);
+		if (unlikely(err)) {
+			brelse(gdb);
+			goto exit_bh;
+		}
 		ext4_set_bit(bit, bh->b_data);
 		brelse(gdb);
 	}
@@ -258,7 +262,11 @@ static int setup_new_group_blocks(struct super_block *sb,
 
 	ext4_mark_bitmap_end(input->blocks_count, sb->s_blocksize * 8,
 			     bh->b_data);
-	ext4_handle_dirty_metadata(handle, NULL, bh);
+	err = ext4_handle_dirty_metadata(handle, NULL, bh);
+	if (unlikely(err)) {
+		ext4_std_error(sb, err);
+		goto exit_bh;
+	}
 	brelse(bh);
 	/* Mark unused entries in inode bitmap used */
 	ext4_debug("clear inode bitmap %#04llx (+%llu)\n",
@@ -270,7 +278,9 @@ static int setup_new_group_blocks(struct super_block *sb,
 
 	ext4_mark_bitmap_end(EXT4_INODES_PER_GROUP(sb), sb->s_blocksize * 8,
 			     bh->b_data);
-	ext4_handle_dirty_metadata(handle, NULL, bh);
+	err = ext4_handle_dirty_metadata(handle, NULL, bh);
+	if (unlikely(err))
+		ext4_std_error(sb, err);
 exit_bh:
 	brelse(bh);
 
@@ -422,17 +432,21 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 		goto exit_dind;
 	}
 
-	if ((err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh)))
+	err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh);
+	if (unlikely(err))
 		goto exit_dind;
 
-	if ((err = ext4_journal_get_write_access(handle, *primary)))
+	err = ext4_journal_get_write_access(handle, *primary);
+	if (unlikely(err))
 		goto exit_sbh;
 
-	if ((err = ext4_journal_get_write_access(handle, dind)))
-		goto exit_primary;
+	err = ext4_journal_get_write_access(handle, dind);
+	if (unlikely(err))
+		ext4_std_error(sb, err);
 
 	/* ext4_reserve_inode_write() gets a reference on the iloc */
-	if ((err = ext4_reserve_inode_write(handle, inode, &iloc)))
+	err = ext4_reserve_inode_write(handle, inode, &iloc);
+	if (unlikely(err))
 		goto exit_dindj;
 
 	n_group_desc = kmalloc((gdb_num + 1) * sizeof(struct buffer_head *),
@@ -454,12 +468,20 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	 * reserved inode, and will become GDT blocks (primary and backup).
 	 */
 	data[gdb_num % EXT4_ADDR_PER_BLOCK(sb)] = 0;
-	ext4_handle_dirty_metadata(handle, NULL, dind);
-	brelse(dind);
+	err = ext4_handle_dirty_metadata(handle, NULL, dind);
+	if (unlikely(err)) {
+		ext4_std_error(sb, err);
+		goto exit_inode;
+	}
 	inode->i_blocks -= (gdbackups + 1) * sb->s_blocksize >> 9;
 	ext4_mark_iloc_dirty(handle, inode, &iloc);
 	memset((*primary)->b_data, 0, sb->s_blocksize);
-	ext4_handle_dirty_metadata(handle, NULL, *primary);
+	err = ext4_handle_dirty_metadata(handle, NULL, *primary);
+	if (unlikely(err)) {
+		ext4_std_error(sb, err);
+		goto exit_inode;
+	}
+	brelse(dind);
 
 	o_group_desc = EXT4_SB(sb)->s_group_desc;
 	memcpy(n_group_desc, o_group_desc,
@@ -470,19 +492,19 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	kfree(o_group_desc);
 
 	le16_add_cpu(&es->s_reserved_gdt_blocks, -1);
-	ext4_handle_dirty_metadata(handle, NULL, EXT4_SB(sb)->s_sbh);
+	err = ext4_handle_dirty_metadata(handle, NULL, EXT4_SB(sb)->s_sbh);
+	if (err)
+		ext4_std_error(sb, err);
 
-	return 0;
+	return err;
 
 exit_inode:
 	/* ext4_journal_release_buffer(handle, iloc.bh); */
 	brelse(iloc.bh);
 exit_dindj:
 	/* ext4_journal_release_buffer(handle, dind); */
-exit_primary:
-	/* ext4_journal_release_buffer(handle, *primary); */
 exit_sbh:
-	/* ext4_journal_release_buffer(handle, *primary); */
+	/* ext4_journal_release_buffer(handle, EXT4_SB(sb)->s_sbh); */
 exit_dind:
 	brelse(dind);
 exit_bh:
@@ -665,7 +687,9 @@ static void update_backups(struct super_block *sb,
 			memset(bh->b_data + size, 0, rest);
 		set_buffer_uptodate(bh);
 		unlock_buffer(bh);
-		ext4_handle_dirty_metadata(handle, NULL, bh);
+		err = ext4_handle_dirty_metadata(handle, NULL, bh);
+		if (unlikely(err))
+			ext4_std_error(sb, err);
 		brelse(bh);
 	}
 	if ((err2 = ext4_journal_stop(handle)) && !err)
@@ -883,7 +907,11 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 	/* Update the global fs size fields */
 	sbi->s_groups_count++;
 
-	ext4_handle_dirty_metadata(handle, NULL, primary);
+	err = ext4_handle_dirty_metadata(handle, NULL, primary);
+	if (unlikely(err)) {
+		ext4_std_error(sb, err);
+		goto exit_journal;
+	}
 
 	/* Update the reserved block counts only once the new group is
 	 * active. */

fs/ext4/xattr.c

@@ -427,23 +427,23 @@ ext4_xattr_ibody_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 static int
 ext4_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
-	int i_error, b_error;
+	int ret, ret2;
 
 	down_read(&EXT4_I(dentry->d_inode)->xattr_sem);
-	i_error = ext4_xattr_ibody_list(dentry, buffer, buffer_size);
-	if (i_error < 0) {
-		b_error = 0;
-	} else {
-		if (buffer) {
-			buffer += i_error;
-			buffer_size -= i_error;
-		}
-		b_error = ext4_xattr_block_list(dentry, buffer, buffer_size);
-		if (b_error < 0)
-			i_error = 0;
+	ret = ret2 = ext4_xattr_ibody_list(dentry, buffer, buffer_size);
+	if (ret < 0)
+		goto errout;
+	if (buffer) {
+		buffer += ret;
+		buffer_size -= ret;
 	}
+	ret = ext4_xattr_block_list(dentry, buffer, buffer_size);
+	if (ret < 0)
+		goto errout;
+	ret += ret2;
+errout:
 	up_read(&EXT4_I(dentry->d_inode)->xattr_sem);
-	return i_error + b_error;
+	return ret;
 }
 
 /*
@@ -947,7 +947,7 @@ ext4_xattr_ibody_set(handle_t *handle, struct inode *inode,
 /*
  * ext4_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

fs/jbd2/journal.c

@@ -43,6 +43,7 @@
 #include <linux/vmalloc.h>
 #include <linux/backing-dev.h>
 #include <linux/bitops.h>
+#include <linux/ratelimit.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/jbd2.h>
@@ -93,6 +94,7 @@ EXPORT_SYMBOL(jbd2_journal_file_inode);
 EXPORT_SYMBOL(jbd2_journal_init_jbd_inode);
 EXPORT_SYMBOL(jbd2_journal_release_jbd_inode);
 EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate);
+EXPORT_SYMBOL(jbd2_inode_cache);
 
 static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *);
 static void __journal_abort_soft (journal_t *journal, int errno);
@@ -827,7 +829,7 @@ static journal_t * journal_init_common (void)
 
 	journal = kzalloc(sizeof(*journal), GFP_KERNEL);
 	if (!journal)
-		goto fail;
+		return NULL;
 
 	init_waitqueue_head(&journal->j_wait_transaction_locked);
 	init_waitqueue_head(&journal->j_wait_logspace);
@@ -852,14 +854,12 @@ static journal_t * journal_init_common (void)
 	err = jbd2_journal_init_revoke(journal, JOURNAL_REVOKE_DEFAULT_HASH);
 	if (err) {
 		kfree(journal);
-		goto fail;
+		return NULL;
 	}
 
 	spin_lock_init(&journal->j_history_lock);
 
 	return journal;
-fail:
-	return NULL;
 }
 
 /* jbd2_journal_init_dev and jbd2_journal_init_inode:
@@ -1982,7 +1982,6 @@ static void jbd2_journal_destroy_jbd2_journal_head_cache(void)
 static struct journal_head *journal_alloc_journal_head(void)
 {
 	struct journal_head *ret;
-	static unsigned long last_warning;
 
 #ifdef CONFIG_JBD2_DEBUG
 	atomic_inc(&nr_journal_heads);
@@ -1990,11 +1989,7 @@ static struct journal_head *journal_alloc_journal_head(void)
 	ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS);
 	if (!ret) {
 		jbd_debug(1, "out of memory for journal_head\n");
-		if (time_after(jiffies, last_warning + 5*HZ)) {
-			printk(KERN_NOTICE "ENOMEM in %s, retrying.\n",
-			       __func__);
-			last_warning = jiffies;
-		}
+		pr_notice_ratelimited("ENOMEM in %s, retrying.\n", __func__);
 		while (!ret) {
 			yield();
 			ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS);
@@ -2292,17 +2287,19 @@ static void __exit jbd2_remove_jbd_stats_proc_entry(void)
 
 #endif
 
-struct kmem_cache *jbd2_handle_cache;
+struct kmem_cache *jbd2_handle_cache, *jbd2_inode_cache;
 
 static int __init journal_init_handle_cache(void)
 {
-	jbd2_handle_cache = kmem_cache_create("jbd2_journal_handle",
-				sizeof(handle_t),
-				0,		/* offset */
-				SLAB_TEMPORARY,	/* flags */
-				NULL);		/* ctor */
+	jbd2_handle_cache = KMEM_CACHE(jbd2_journal_handle, SLAB_TEMPORARY);
 	if (jbd2_handle_cache == NULL) {
-		printk(KERN_EMERG "JBD: failed to create handle cache\n");
+		printk(KERN_EMERG "JBD2: failed to create handle cache\n");
+		return -ENOMEM;
+	}
+	jbd2_inode_cache = KMEM_CACHE(jbd2_inode, 0);
+	if (jbd2_inode_cache == NULL) {
+		printk(KERN_EMERG "JBD2: failed to create inode cache\n");
+		kmem_cache_destroy(jbd2_handle_cache);
 		return -ENOMEM;
 	}
 	return 0;
@@ -2312,6 +2309,9 @@ static void jbd2_journal_destroy_handle_cache(void)
 {
 	if (jbd2_handle_cache)
 		kmem_cache_destroy(jbd2_handle_cache);
+	if (jbd2_inode_cache)
+		kmem_cache_destroy(jbd2_inode_cache);
+
 }
 
 /*

fs/jbd2/recovery.c

@@ -299,10 +299,10 @@ int jbd2_journal_skip_recovery(journal_t *journal)
 #ifdef CONFIG_JBD2_DEBUG
 		int dropped = info.end_transaction - 
 			be32_to_cpu(journal->j_superblock->s_sequence);
-#endif
 		jbd_debug(1,
 			  "JBD: ignoring %d transaction%s from the journal.\n",
 			  dropped, (dropped == 1) ? "" : "s");
+#endif
 		journal->j_transaction_sequence = ++info.end_transaction;
 	}
 

fs/jbd2/transaction.c

@@ -340,9 +340,7 @@ handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int gfp_mask)
 		jbd2_free_handle(handle);
 		current->journal_info = NULL;
 		handle = ERR_PTR(err);
-		goto out;
 	}
-out:
 	return handle;
 }
 EXPORT_SYMBOL(jbd2__journal_start);
@@ -589,7 +587,7 @@ do_get_write_access(handle_t *handle, struct journal_head *jh,
 	transaction = handle->h_transaction;
 	journal = transaction->t_journal;
 
-	jbd_debug(5, "buffer_head %p, force_copy %d\n", jh, force_copy);
+	jbd_debug(5, "journal_head %p, force_copy %d\n", jh, force_copy);
 
 	JBUFFER_TRACE(jh, "entry");
 repeat:
@@ -774,7 +772,7 @@ do_get_write_access(handle_t *handle, struct journal_head *jh,
 		J_EXPECT_JH(jh, buffer_uptodate(jh2bh(jh)),
 			    "Possible IO failure.\n");
 		page = jh2bh(jh)->b_page;
-		offset = ((unsigned long) jh2bh(jh)->b_data) & ~PAGE_MASK;
+		offset = offset_in_page(jh2bh(jh)->b_data);
 		source = kmap_atomic(page, KM_USER0);
 		/* Fire data frozen trigger just before we copy the data */
 		jbd2_buffer_frozen_trigger(jh, source + offset,

include/linux/jbd2.h

@@ -94,7 +94,7 @@ extern void jbd2_free(void *ptr, size_t size);
  *
  * This is an opaque datatype.
  **/
-typedef struct handle_s		handle_t;	/* Atomic operation type */
+typedef struct jbd2_journal_handle handle_t;	/* Atomic operation type */
 
 
 /**
@@ -416,7 +416,7 @@ struct jbd2_revoke_table_s;
  * in so it can be fixed later.
  */
 
-struct handle_s
+struct jbd2_journal_handle
 {
 	/* Which compound transaction is this update a part of? */
 	transaction_t		*h_transaction;
@@ -1158,6 +1158,22 @@ static inline void jbd2_free_handle(handle_t *handle)
 	kmem_cache_free(jbd2_handle_cache, handle);
 }
 
+/*
+ * jbd2_inode management (optional, for those file systems that want to use
+ * dynamically allocated jbd2_inode structures)
+ */
+extern struct kmem_cache *jbd2_inode_cache;
+
+static inline struct jbd2_inode *jbd2_alloc_inode(gfp_t gfp_flags)
+{
+	return kmem_cache_alloc(jbd2_inode_cache, gfp_flags);
+}
+
+static inline void jbd2_free_inode(struct jbd2_inode *jinode)
+{
+	kmem_cache_free(jbd2_inode_cache, jinode);
+}
+
 /* Primary revoke support */
 #define JOURNAL_REVOKE_DEFAULT_HASH 256
 extern int	   jbd2_journal_init_revoke(journal_t *, int);