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

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (40 commits)
  ext4: Adding error check after calling ext4_mb_regular_allocator()
  ext4: Fix dirtying of journalled buffers in data=journal mode
  ext4: re-inline ext4_rec_len_(to|from)_disk functions
  jbd2: Remove t_handle_lock from start_this_handle()
  jbd2: Change j_state_lock to be a rwlock_t
  jbd2: Use atomic variables to avoid taking t_handle_lock in jbd2_journal_stop
  ext4: Add mount options in superblock
  ext4: force block allocation on quota_off
  ext4: fix freeze deadlock under IO
  ext4: drop inode from orphan list if ext4_delete_inode() fails
  ext4: check to make make sure bd_dev is set before dereferencing it
  jbd2: Make barrier messages less scary
  ext4: don't print scary messages for allocation failures post-abort
  ext4: fix EFBIG edge case when writing to large non-extent file
  ext4: fix ext4_get_blocks references
  ext4: Always journal quota file modifications
  ext4: Fix potential memory leak in ext4_fill_super
  ext4: Don't error out the fs if the user tries to make a file too big
  ext4: allocate stripe-multiple IOs on stripe boundaries
  ext4: move aio completion after unwritten extent conversion
  ...

Fix up conflicts in fs/ext4/inode.c as per Ted.

Fix up xfs conflicts as per earlier xfs merge.

fs/ext4/acl.c

@@ -204,6 +204,7 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type,
 				return error;
 			else {
 				inode->i_mode = mode;
+				inode->i_ctime = ext4_current_time(inode);
 				ext4_mark_inode_dirty(handle, inode);
 				if (error == 0)
 					acl = NULL;

fs/ext4/balloc.c

@@ -377,14 +377,11 @@ void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
 	ext4_grpblk_t bit;
 	unsigned int i;
 	struct ext4_group_desc *desc;
-	struct ext4_super_block *es;
-	struct ext4_sb_info *sbi;
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	int err = 0, ret, blk_free_count;
 	ext4_grpblk_t blocks_freed;
 	struct ext4_group_info *grp;
 
-	sbi = EXT4_SB(sb);
-	es = sbi->s_es;
 	ext4_debug("Adding block(s) %llu-%llu\n", block, block + count - 1);
 
 	ext4_get_group_no_and_offset(sb, block, &block_group, &bit);
@@ -477,7 +474,6 @@ void ext4_add_groupblocks(handle_t *handle, struct super_block *sb,
 	ret = ext4_handle_dirty_metadata(handle, NULL, gd_bh);
 	if (!err)
 		err = ret;
-	sb->s_dirt = 1;
 
 error_return:
 	brelse(bitmap_bh);

fs/ext4/block_validity.c

@@ -229,16 +229,20 @@ int ext4_data_block_valid(struct ext4_sb_info *sbi, ext4_fsblk_t start_blk,
 
 	if ((start_blk <= le32_to_cpu(sbi->s_es->s_first_data_block)) ||
 	    (start_blk + count < start_blk) ||
-	    (start_blk + count > ext4_blocks_count(sbi->s_es)))
+	    (start_blk + count > ext4_blocks_count(sbi->s_es))) {
+		sbi->s_es->s_last_error_block = cpu_to_le64(start_blk);
 		return 0;
+	}
 	while (n) {
 		entry = rb_entry(n, struct ext4_system_zone, node);
 		if (start_blk + count - 1 < entry->start_blk)
 			n = n->rb_left;
 		else if (start_blk >= (entry->start_blk + entry->count))
 			n = n->rb_right;
-		else
+		else {
+			sbi->s_es->s_last_error_block = cpu_to_le64(start_blk);
 			return 0;
+		}
 	}
 	return 1;
 }

fs/ext4/dir.c

@@ -61,10 +61,11 @@ static unsigned char get_dtype(struct super_block *sb, int filetype)
 }
 
 
-int ext4_check_dir_entry(const char *function, struct inode *dir,
-			 struct ext4_dir_entry_2 *de,
-			 struct buffer_head *bh,
-			 unsigned int offset)
+int __ext4_check_dir_entry(const char *function, unsigned int line,
+			   struct inode *dir,
+			   struct ext4_dir_entry_2 *de,
+			   struct buffer_head *bh,
+			   unsigned int offset)
 {
 	const char *error_msg = NULL;
 	const int rlen = ext4_rec_len_from_disk(de->rec_len,
@@ -83,11 +84,10 @@ int ext4_check_dir_entry(const char *function, struct inode *dir,
 		error_msg = "inode out of bounds";
 
 	if (error_msg != NULL)
-		ext4_error_inode(function, dir,
-			"bad entry in directory: %s - block=%llu"
+		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 long long) bh->b_blocknr,
-			(unsigned) (offset%bh->b_size), offset,
+			error_msg, (unsigned) (offset%bh->b_size), offset,
 			le32_to_cpu(de->inode),
 			rlen, de->name_len);
 	return error_msg == NULL ? 1 : 0;
@@ -121,7 +121,8 @@ static int ext4_readdir(struct file *filp,
 		 * We don't set the inode dirty flag since it's not
 		 * critical that it get flushed back to the disk.
 		 */
-		ext4_clear_inode_flag(filp->f_path.dentry->d_inode, EXT4_INODE_INDEX);
+		ext4_clear_inode_flag(filp->f_path.dentry->d_inode,
+				      EXT4_INODE_INDEX);
 	}
 	stored = 0;
 	offset = filp->f_pos & (sb->s_blocksize - 1);
@@ -193,7 +194,7 @@ 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("ext4_readdir", inode, de,
+			if (!ext4_check_dir_entry(inode, de,
 						  bh, offset)) {
 				/*
 				 * On error, skip the f_pos to the next block
@@ -343,7 +344,7 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 	struct dir_private_info *info;
 	int len;
 
-	info = (struct dir_private_info *) dir_file->private_data;
+	info = dir_file->private_data;
 	p = &info->root.rb_node;
 
 	/* Create and allocate the fname structure */

fs/ext4/ext4.h

@@ -57,10 +57,13 @@
 #endif
 
 #define EXT4_ERROR_INODE(inode, fmt, a...) \
-	ext4_error_inode(__func__, (inode), (fmt), ## a)
+	ext4_error_inode((inode), __func__, __LINE__, 0, (fmt), ## a)
+
+#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__, (file), (fmt), ## a)
+	ext4_error_file(__func__, __LINE__, (file), (fmt), ## a)
 
 /* data type for block offset of block group */
 typedef int ext4_grpblk_t;
@@ -167,13 +170,15 @@ struct mpage_da_data {
 };
 #define	EXT4_IO_UNWRITTEN	0x1
 typedef struct ext4_io_end {
-	struct list_head	list;		/* per-file finished AIO list */
+	struct list_head	list;		/* per-file finished IO list */
 	struct inode		*inode;		/* file being written to */
 	unsigned int		flag;		/* unwritten or not */
 	struct page		*page;		/* page struct for buffer write */
 	loff_t			offset;		/* offset in the file */
 	ssize_t			size;		/* size of the extent */
 	struct work_struct	work;		/* data work queue */
+	struct kiocb		*iocb;		/* iocb struct for AIO */
+	int			result;		/* error value for AIO */
 } ext4_io_end_t;
 
 /*
@@ -460,7 +465,7 @@ struct ext4_new_group_data {
 };
 
 /*
- * Flags used by ext4_get_blocks()
+ * Flags used by ext4_map_blocks()
  */
 	/* Allocate any needed blocks and/or convert an unitialized
 	   extent to be an initialized ext4 */
@@ -873,7 +878,6 @@ struct ext4_inode_info {
 #define EXT4_MOUNT_POSIX_ACL		0x08000	/* POSIX Access Control Lists */
 #define EXT4_MOUNT_NO_AUTO_DA_ALLOC	0x10000	/* No auto delalloc mapping */
 #define EXT4_MOUNT_BARRIER		0x20000 /* Use block barriers */
-#define EXT4_MOUNT_NOBH			0x40000 /* No bufferheads */
 #define EXT4_MOUNT_QUOTA		0x80000 /* Some quota option set */
 #define EXT4_MOUNT_USRQUOTA		0x100000 /* "old" user quota */
 #define EXT4_MOUNT_GRPQUOTA		0x200000 /* "old" group quota */
@@ -982,7 +986,7 @@ struct ext4_super_block {
 	__le32	s_last_orphan;		/* start of list of inodes to delete */
 	__le32	s_hash_seed[4];		/* HTREE hash seed */
 	__u8	s_def_hash_version;	/* Default hash version to use */
-	__u8	s_reserved_char_pad;
+	__u8	s_jnl_backup_type;
 	__le16  s_desc_size;		/* size of group descriptor */
 /*100*/	__le32	s_default_mount_opts;
 	__le32	s_first_meta_bg;	/* First metablock block group */
@@ -1000,12 +1004,34 @@ struct ext4_super_block {
 	__le64  s_mmp_block;            /* Block for multi-mount protection */
 	__le32  s_raid_stripe_width;    /* blocks on all data disks (N*stride)*/
 	__u8	s_log_groups_per_flex;  /* FLEX_BG group size */
-	__u8	s_reserved_char_pad2;
+	__u8	s_reserved_char_pad;
 	__le16  s_reserved_pad;
 	__le64	s_kbytes_written;	/* nr of lifetime kilobytes written */
-	__u32   s_reserved[160];        /* Padding to the end of the block */
+	__le32	s_snapshot_inum;	/* Inode number of active snapshot */
+	__le32	s_snapshot_id;		/* sequential ID of active snapshot */
+	__le64	s_snapshot_r_blocks_count; /* reserved blocks for active
+					      snapshot's future use */
+	__le32	s_snapshot_list;	/* inode number of the head of the
+					   on-disk snapshot list */
+#define EXT4_S_ERR_START offsetof(struct ext4_super_block, s_error_count)
+	__le32	s_error_count;		/* number of fs errors */
+	__le32	s_first_error_time;	/* first time an error happened */
+	__le32	s_first_error_ino;	/* inode involved in first error */
+	__le64	s_first_error_block;	/* block involved of first error */
+	__u8	s_first_error_func[32];	/* function where the error happened */
+	__le32	s_first_error_line;	/* line number where error happened */
+	__le32	s_last_error_time;	/* most recent time of an error */
+	__le32	s_last_error_ino;	/* inode involved in last error */
+	__le32	s_last_error_line;	/* line number where error happened */
+	__le64	s_last_error_block;	/* block involved of last error */
+	__u8	s_last_error_func[32];	/* function where the error happened */
+#define EXT4_S_ERR_END offsetof(struct ext4_super_block, s_mount_opts)
+	__u8	s_mount_opts[64];
+	__le32	s_reserved[112];        /* Padding to the end of the block */
 };
 
+#define EXT4_S_ERR_LEN (EXT4_S_ERR_END - EXT4_S_ERR_START)
+
 #ifdef __KERNEL__
 
 /*
@@ -1143,6 +1169,9 @@ struct ext4_sb_info {
 
 	/* workqueue for dio unwritten */
 	struct workqueue_struct *dio_unwritten_wq;
+
+	/* timer for periodic error stats printing */
+	struct timer_list s_err_report;
 };
 
 static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb)
@@ -1313,6 +1342,10 @@ EXT4_INODE_BIT_FNS(state, state_flags)
 #define EXT4_DEFM_JMODE_DATA	0x0020
 #define EXT4_DEFM_JMODE_ORDERED	0x0040
 #define EXT4_DEFM_JMODE_WBACK	0x0060
+#define EXT4_DEFM_NOBARRIER	0x0100
+#define EXT4_DEFM_BLOCK_VALIDITY 0x0200
+#define EXT4_DEFM_DISCARD	0x0400
+#define EXT4_DEFM_NODELALLOC	0x0800
 
 /*
  * Default journal batch times
@@ -1378,6 +1411,43 @@ struct ext4_dir_entry_2 {
 					 ~EXT4_DIR_ROUND)
 #define EXT4_MAX_REC_LEN		((1<<16)-1)
 
+/*
+ * If we ever get support for fs block sizes > page_size, we'll need
+ * to remove the #if statements in the next two functions...
+ */
+static inline unsigned int
+ext4_rec_len_from_disk(__le16 dlen, unsigned blocksize)
+{
+	unsigned len = le16_to_cpu(dlen);
+
+#if (PAGE_CACHE_SIZE >= 65536)
+	if (len == EXT4_MAX_REC_LEN || len == 0)
+		return blocksize;
+	return (len & 65532) | ((len & 3) << 16);
+#else
+	return len;
+#endif
+}
+
+static inline __le16 ext4_rec_len_to_disk(unsigned len, unsigned blocksize)
+{
+	if ((len > blocksize) || (blocksize > (1 << 18)) || (len & 3))
+		BUG();
+#if (PAGE_CACHE_SIZE >= 65536)
+	if (len < 65536)
+		return cpu_to_le16(len);
+	if (len == blocksize) {
+		if (blocksize == 65536)
+			return cpu_to_le16(EXT4_MAX_REC_LEN);
+		else
+			return cpu_to_le16(0);
+	}
+	return cpu_to_le16((len & 65532) | ((len >> 16) & 3));
+#else
+	return cpu_to_le16(len);
+#endif
+}
+
 /*
  * Hash Tree Directory indexing
  * (c) Daniel Phillips, 2001
@@ -1510,9 +1580,11 @@ extern unsigned ext4_init_block_bitmap(struct super_block *sb,
 		ext4_init_block_bitmap(sb, NULL, group, desc)
 
 /* dir.c */
-extern int ext4_check_dir_entry(const char *, struct inode *,
-				struct ext4_dir_entry_2 *,
-				struct buffer_head *, unsigned int);
+extern int __ext4_check_dir_entry(const char *, unsigned int, struct inode *,
+				  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))
 extern int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 				    __u32 minor_hash,
 				    struct ext4_dir_entry_2 *dirent);
@@ -1601,8 +1673,6 @@ extern long ext4_compat_ioctl(struct file *, unsigned int, unsigned long);
 extern int ext4_ext_migrate(struct inode *);
 
 /* namei.c */
-extern unsigned int ext4_rec_len_from_disk(__le16 dlen, unsigned blocksize);
-extern __le16 ext4_rec_len_to_disk(unsigned len, unsigned blocksize);
 extern int ext4_orphan_add(handle_t *, struct inode *);
 extern int ext4_orphan_del(handle_t *, struct inode *);
 extern int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash,
@@ -1616,25 +1686,38 @@ extern int ext4_group_extend(struct super_block *sb,
 				ext4_fsblk_t n_blocks_count);
 
 /* super.c */
-extern void __ext4_error(struct super_block *, const char *, const char *, ...)
-	__attribute__ ((format (printf, 3, 4)));
-#define ext4_error(sb, message...)	__ext4_error(sb, __func__, ## message)
-extern void ext4_error_inode(const char *, struct inode *, const char *, ...)
-	__attribute__ ((format (printf, 3, 4)));
-extern void ext4_error_file(const char *, struct file *, const char *, ...)
-	__attribute__ ((format (printf, 3, 4)));
-extern void __ext4_std_error(struct super_block *, const char *, int);
-extern void ext4_abort(struct super_block *, const char *, const char *, ...)
-	__attribute__ ((format (printf, 3, 4)));
-extern void __ext4_warning(struct super_block *, const char *,
+extern void __ext4_error(struct super_block *, const char *, unsigned int,
+			 const char *, ...)
+	__attribute__ ((format (printf, 4, 5)));
+#define ext4_error(sb, message...)	__ext4_error(sb, __func__,	\
+						     __LINE__, ## message)
+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)));
+extern void __ext4_std_error(struct super_block *, const char *,
+			     unsigned int, int);
+extern void __ext4_abort(struct super_block *, const char *, unsigned int,
+		       const char *, ...)
+	__attribute__ ((format (printf, 4, 5)));
+#define ext4_abort(sb, message...)	__ext4_abort(sb, __func__, \
+						       __LINE__, ## message)
+extern void __ext4_warning(struct super_block *, const char *, unsigned int,
 			  const char *, ...)
-	__attribute__ ((format (printf, 3, 4)));
-#define ext4_warning(sb, message...)	__ext4_warning(sb, __func__, ## message)
+	__attribute__ ((format (printf, 4, 5)));
+#define ext4_warning(sb, message...)	__ext4_warning(sb, __func__, \
+						       __LINE__, ## message)
 extern void ext4_msg(struct super_block *, const char *, const char *, ...)
 	__attribute__ ((format (printf, 3, 4)));
-extern void ext4_grp_locked_error(struct super_block *, ext4_group_t,
-				const char *, const char *, ...)
-	__attribute__ ((format (printf, 4, 5)));
+extern void __ext4_grp_locked_error(const char *, unsigned int, \
+				    struct super_block *, ext4_group_t, \
+				    unsigned long, ext4_fsblk_t, \
+				    const char *, ...)
+	__attribute__ ((format (printf, 7, 8)));
+#define ext4_grp_locked_error(sb, grp, message...) \
+	__ext4_grp_locked_error(__func__, __LINE__, (sb), (grp), ## message)
 extern void ext4_update_dynamic_rev(struct super_block *sb);
 extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb,
 					__u32 compat);
@@ -1768,7 +1851,7 @@ static inline unsigned int ext4_flex_bg_size(struct ext4_sb_info *sbi)
 #define ext4_std_error(sb, errno)				\
 do {								\
 	if ((errno))						\
-		__ext4_std_error((sb), __func__, (errno));	\
+		__ext4_std_error((sb), __func__, __LINE__, (errno));	\
 } while (0)
 
 #ifdef CONFIG_SMP
@@ -1860,6 +1943,12 @@ static inline void ext4_unlock_group(struct super_block *sb,
 	spin_unlock(ext4_group_lock_ptr(sb, group));
 }
 
+static inline void ext4_mark_super_dirty(struct super_block *sb)
+{
+	if (EXT4_SB(sb)->s_journal == NULL)
+		sb->s_dirt =1;
+}
+
 /*
  * Inodes and files operations
  */
@@ -1905,9 +1994,6 @@ extern int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset,
 			  ssize_t len);
 extern int ext4_map_blocks(handle_t *handle, struct inode *inode,
 			   struct ext4_map_blocks *map, int flags);
-extern int ext4_get_blocks(handle_t *handle, struct inode *inode,
-			   sector_t block, unsigned int max_blocks,
-			   struct buffer_head *bh, int flags);
 extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			__u64 start, __u64 len);
 /* move_extent.c */

fs/ext4/ext4_jbd2.c

@@ -6,29 +6,29 @@
 
 #include <trace/events/ext4.h>
 
-int __ext4_journal_get_undo_access(const char *where, handle_t *handle,
-				struct buffer_head *bh)
+int __ext4_journal_get_undo_access(const char *where, unsigned int line,
+				   handle_t *handle, struct buffer_head *bh)
 {
 	int err = 0;
 
 	if (ext4_handle_valid(handle)) {
 		err = jbd2_journal_get_undo_access(handle, bh);
 		if (err)
-			ext4_journal_abort_handle(where, __func__, bh,
+			ext4_journal_abort_handle(where, line, __func__, bh,
 						  handle, err);
 	}
 	return err;
 }
 
-int __ext4_journal_get_write_access(const char *where, handle_t *handle,
-				struct buffer_head *bh)
+int __ext4_journal_get_write_access(const char *where, unsigned int line,
+				    handle_t *handle, struct buffer_head *bh)
 {
 	int err = 0;
 
 	if (ext4_handle_valid(handle)) {
 		err = jbd2_journal_get_write_access(handle, bh);
 		if (err)
-			ext4_journal_abort_handle(where, __func__, bh,
+			ext4_journal_abort_handle(where, line, __func__, bh,
 						  handle, err);
 	}
 	return err;
@@ -46,9 +46,9 @@ int __ext4_journal_get_write_access(const char *where, handle_t *handle,
  * If the handle isn't valid we're not journaling, but we still need to
  * call into ext4_journal_revoke() to put the buffer head.
  */
-int __ext4_forget(const char *where, handle_t *handle, int is_metadata,
-		  struct inode *inode, struct buffer_head *bh,
-		  ext4_fsblk_t blocknr)
+int __ext4_forget(const char *where, unsigned int line, handle_t *handle,
+		  int is_metadata, struct inode *inode,
+		  struct buffer_head *bh, ext4_fsblk_t blocknr)
 {
 	int err;
 
@@ -79,8 +79,8 @@ int __ext4_forget(const char *where, handle_t *handle, int is_metadata,
 			BUFFER_TRACE(bh, "call jbd2_journal_forget");
 			err = jbd2_journal_forget(handle, bh);
 			if (err)
-				ext4_journal_abort_handle(where, __func__, bh,
-							  handle, err);
+				ext4_journal_abort_handle(where, line, __func__,
+							  bh, handle, err);
 			return err;
 		}
 		return 0;
@@ -92,15 +92,16 @@ int __ext4_forget(const char *where, handle_t *handle, int is_metadata,
 	BUFFER_TRACE(bh, "call jbd2_journal_revoke");
 	err = jbd2_journal_revoke(handle, blocknr, bh);
 	if (err) {
-		ext4_journal_abort_handle(where, __func__, bh, handle, err);
-		ext4_abort(inode->i_sb, __func__,
+		ext4_journal_abort_handle(where, line, __func__,
+					  bh, handle, err);
+		__ext4_abort(inode->i_sb, where, line,
 			   "error %d when attempting revoke", err);
 	}
 	BUFFER_TRACE(bh, "exit");
 	return err;
 }
 
-int __ext4_journal_get_create_access(const char *where,
+int __ext4_journal_get_create_access(const char *where, unsigned int line,
 				handle_t *handle, struct buffer_head *bh)
 {
 	int err = 0;
@@ -108,22 +109,23 @@ int __ext4_journal_get_create_access(const char *where,
 	if (ext4_handle_valid(handle)) {
 		err = jbd2_journal_get_create_access(handle, bh);
 		if (err)
-			ext4_journal_abort_handle(where, __func__, bh,
-						  handle, err);
+			ext4_journal_abort_handle(where, line, __func__,
+						  bh, handle, err);
 	}
 	return err;
 }
 
-int __ext4_handle_dirty_metadata(const char *where, handle_t *handle,
-				 struct inode *inode, struct buffer_head *bh)
+int __ext4_handle_dirty_metadata(const char *where, unsigned int line,
+				 handle_t *handle, struct inode *inode,
+				 struct buffer_head *bh)
 {
 	int err = 0;
 
 	if (ext4_handle_valid(handle)) {
 		err = jbd2_journal_dirty_metadata(handle, bh);
 		if (err)
-			ext4_journal_abort_handle(where, __func__, bh,
-						  handle, err);
+			ext4_journal_abort_handle(where, line, __func__,
+						  bh, handle, err);
 	} else {
 		if (inode)
 			mark_buffer_dirty_inode(bh, inode);
@@ -132,14 +134,33 @@ int __ext4_handle_dirty_metadata(const char *where, handle_t *handle,
 		if (inode && inode_needs_sync(inode)) {
 			sync_dirty_buffer(bh);
 			if (buffer_req(bh) && !buffer_uptodate(bh)) {
-				ext4_error(inode->i_sb,
-					   "IO error syncing inode, "
-					   "inode=%lu, block=%llu",
-					   inode->i_ino,
-					   (unsigned long long) bh->b_blocknr);
+				struct ext4_super_block *es;
+
+				es = EXT4_SB(inode->i_sb)->s_es;
+				es->s_last_error_block =
+					cpu_to_le64(bh->b_blocknr);
+				ext4_error_inode(inode, where, line,
+						 bh->b_blocknr,
+					"IO error syncing itable block");
 				err = -EIO;
 			}
 		}
 	}
 	return err;
 }
+
+int __ext4_handle_dirty_super(const char *where, unsigned int line,
+			      handle_t *handle, struct super_block *sb)
+{
+	struct buffer_head *bh = EXT4_SB(sb)->s_sbh;
+	int err = 0;
+
+	if (ext4_handle_valid(handle)) {
+		err = jbd2_journal_dirty_metadata(handle, bh);
+		if (err)
+			ext4_journal_abort_handle(where, line, __func__,
+						  bh, handle, err);
+	} else
+		sb->s_dirt = 1;
+	return err;
+}

fs/ext4/ext4_jbd2.h

@@ -122,39 +122,47 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode);
 /*
  * Wrapper functions with which ext4 calls into JBD.
  */
-void ext4_journal_abort_handle(const char *caller, const char *err_fn,
+void ext4_journal_abort_handle(const char *caller, unsigned int line,
+			       const char *err_fn,
 		struct buffer_head *bh, handle_t *handle, int err);
 
-int __ext4_journal_get_undo_access(const char *where, handle_t *handle,
-				struct buffer_head *bh);
+int __ext4_journal_get_undo_access(const char *where, unsigned int line,
+				   handle_t *handle, struct buffer_head *bh);
 
-int __ext4_journal_get_write_access(const char *where, handle_t *handle,
-				struct buffer_head *bh);
+int __ext4_journal_get_write_access(const char *where, unsigned int line,
+				    handle_t *handle, struct buffer_head *bh);
 
-int __ext4_forget(const char *where, handle_t *handle, int is_metadata,
-		  struct inode *inode, struct buffer_head *bh,
-		  ext4_fsblk_t blocknr);
+int __ext4_forget(const char *where, unsigned int line, handle_t *handle,
+		  int is_metadata, struct inode *inode,
+		  struct buffer_head *bh, ext4_fsblk_t blocknr);
 
-int __ext4_journal_get_create_access(const char *where,
+int __ext4_journal_get_create_access(const char *where, unsigned int line,
 				handle_t *handle, struct buffer_head *bh);
 
-int __ext4_handle_dirty_metadata(const char *where, handle_t *handle,
-				 struct inode *inode, struct buffer_head *bh);
+int __ext4_handle_dirty_metadata(const char *where, unsigned int line,
+				 handle_t *handle, struct inode *inode,
+				 struct buffer_head *bh);
+
+int __ext4_handle_dirty_super(const char *where, unsigned int line,
+			      handle_t *handle, struct super_block *sb);
 
 #define ext4_journal_get_undo_access(handle, bh) \
-	__ext4_journal_get_undo_access(__func__, (handle), (bh))
+	__ext4_journal_get_undo_access(__func__, __LINE__, (handle), (bh))
 #define ext4_journal_get_write_access(handle, bh) \
-	__ext4_journal_get_write_access(__func__, (handle), (bh))
+	__ext4_journal_get_write_access(__func__, __LINE__, (handle), (bh))
 #define ext4_forget(handle, is_metadata, inode, bh, block_nr) \
-	__ext4_forget(__func__, (handle), (is_metadata), (inode), (bh),\
-		      (block_nr))
+	__ext4_forget(__func__, __LINE__, (handle), (is_metadata), (inode), \
+		      (bh), (block_nr))
 #define ext4_journal_get_create_access(handle, bh) \
-	__ext4_journal_get_create_access(__func__, (handle), (bh))
+	__ext4_journal_get_create_access(__func__, __LINE__, (handle), (bh))
 #define ext4_handle_dirty_metadata(handle, inode, bh) \
-	__ext4_handle_dirty_metadata(__func__, (handle), (inode), (bh))
+	__ext4_handle_dirty_metadata(__func__, __LINE__, (handle), (inode), \
+				     (bh))
+#define ext4_handle_dirty_super(handle, sb) \
+	__ext4_handle_dirty_super(__func__, __LINE__, (handle), (sb))
 
 handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks);
-int __ext4_journal_stop(const char *where, handle_t *handle);
+int __ext4_journal_stop(const char *where, unsigned int line, handle_t *handle);
 
 #define EXT4_NOJOURNAL_MAX_REF_COUNT ((unsigned long) 4096)
 
@@ -207,7 +215,7 @@ static inline handle_t *ext4_journal_start(struct inode *inode, int nblocks)
 }
 
 #define ext4_journal_stop(handle) \
-	__ext4_journal_stop(__func__, (handle))
+	__ext4_journal_stop(__func__, __LINE__, (handle))
 
 static inline handle_t *ext4_journal_current_handle(void)
 {
@@ -308,17 +316,15 @@ static inline int ext4_should_writeback_data(struct inode *inode)
  * This function controls whether or not we should try to go down the
  * dioread_nolock code paths, which makes it safe to avoid taking
  * i_mutex for direct I/O reads.  This only works for extent-based
- * files, and it doesn't work for nobh or if data journaling is
- * enabled, since the dioread_nolock code uses b_private to pass
- * information back to the I/O completion handler, and this conflicts
- * with the jbd's use of b_private.
+ * files, and it doesn't work if data journaling is enabled, since the
+ * dioread_nolock code uses b_private to pass information back to the
+ * I/O completion handler, and this conflicts with the jbd's use of
+ * b_private.
  */
 static inline int ext4_should_dioread_nolock(struct inode *inode)
 {
 	if (!test_opt(inode->i_sb, DIOREAD_NOLOCK))
 		return 0;
-	if (test_opt(inode->i_sb, NOBH))
-		return 0;
 	if (!S_ISREG(inode->i_mode))
 		return 0;
 	if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))

fs/ext4/extents.c

@@ -401,9 +401,9 @@ static int ext4_valid_extent_entries(struct inode *inode,
 	return 1;
 }
 
-static int __ext4_ext_check(const char *function, struct inode *inode,
-					struct ext4_extent_header *eh,
-					int depth)
+static int __ext4_ext_check(const char *function, unsigned int line,
+			    struct inode *inode, struct ext4_extent_header *eh,
+			    int depth)
 {
 	const char *error_msg;
 	int max = 0;
@@ -436,7 +436,7 @@ static int __ext4_ext_check(const char *function, struct inode *inode,
 	return 0;
 
 corrupted:
-	ext4_error_inode(function, inode,
+	ext4_error_inode(inode, function, line, 0,
 			"bad header/extent: %s - magic %x, "
 			"entries %u, max %u(%u), depth %u(%u)",
 			error_msg, le16_to_cpu(eh->eh_magic),
@@ -447,7 +447,7 @@ static int __ext4_ext_check(const char *function, struct inode *inode,
 }
 
 #define ext4_ext_check(inode, eh, depth)	\
-	__ext4_ext_check(__func__, inode, eh, depth)
+	__ext4_ext_check(__func__, __LINE__, inode, eh, depth)
 
 int ext4_ext_check_inode(struct inode *inode)
 {
@@ -1083,7 +1083,6 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode,
 {
 	struct ext4_ext_path *curp = path;
 	struct ext4_extent_header *neh;
-	struct ext4_extent_idx *fidx;
 	struct buffer_head *bh;
 	ext4_fsblk_t newblock;
 	int err = 0;
@@ -1144,10 +1143,10 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode,
 	ext4_idx_store_pblock(curp->p_idx, newblock);
 
 	neh = ext_inode_hdr(inode);
-	fidx = EXT_FIRST_INDEX(neh);
 	ext_debug("new root: num %d(%d), lblock %d, ptr %llu\n",
 		  le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max),
-		  le32_to_cpu(fidx->ei_block), idx_pblock(fidx));
+		  le32_to_cpu(EXT_FIRST_INDEX(neh)->ei_block),
+		  idx_pblock(EXT_FIRST_INDEX(neh)));
 
 	neh->eh_depth = cpu_to_le16(path->p_depth + 1);
 	err = ext4_ext_dirty(handle, inode, curp);
@@ -2954,7 +2953,6 @@ static int ext4_split_unwritten_extents(handle_t *handle,
 	struct ext4_extent *ex1 = NULL;
 	struct ext4_extent *ex2 = NULL;
 	struct ext4_extent *ex3 = NULL;
-	struct ext4_extent_header *eh;
 	ext4_lblk_t ee_block, eof_block;
 	unsigned int allocated, ee_len, depth;
 	ext4_fsblk_t newblock;
@@ -2971,7 +2969,6 @@ static int ext4_split_unwritten_extents(handle_t *handle,
 		eof_block = map->m_lblk + map->m_len;
 
 	depth = ext_depth(inode);
-	eh = path[depth].p_hdr;
 	ex = path[depth].p_ext;
 	ee_block = le32_to_cpu(ex->ee_block);
 	ee_len = ext4_ext_get_actual_len(ex);
@@ -3058,7 +3055,6 @@ static int ext4_split_unwritten_extents(handle_t *handle,
 			err = PTR_ERR(path);
 			goto out;
 		}
-		eh = path[depth].p_hdr;
 		ex = path[depth].p_ext;
 		if (ex2 != &newex)
 			ex2 = ex;

fs/ext4/file.c

@@ -70,7 +70,8 @@ ext4_file_write(struct kiocb *iocb, const struct iovec *iov,
 		struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 		size_t length = iov_length(iov, nr_segs);
 
-		if (pos > sbi->s_bitmap_maxbytes)
+		if ((pos > sbi->s_bitmap_maxbytes ||
+		    (pos == sbi->s_bitmap_maxbytes && length > 0)))
 			return -EFBIG;
 
 		if (pos + length > sbi->s_bitmap_maxbytes) {
@@ -123,7 +124,7 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
 		if (!IS_ERR(cp)) {
 			memcpy(sbi->s_es->s_last_mounted, cp,
 			       sizeof(sbi->s_es->s_last_mounted));
-			sb->s_dirt = 1;
+			ext4_mark_super_dirty(sb);
 		}
 	}
 	return dquot_file_open(inode, filp);

fs/ext4/ialloc.c

@@ -279,7 +279,7 @@ void ext4_free_inode(handle_t *handle, struct inode *inode)
 		err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh);
 		if (!fatal)
 			fatal = err;
-		sb->s_dirt = 1;
+		ext4_mark_super_dirty(sb);
 	} else
 		ext4_error(sb, "bit already cleared for inode %lu", ino);
 
@@ -965,7 +965,7 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, int mode,
 	percpu_counter_dec(&sbi->s_freeinodes_counter);
 	if (S_ISDIR(mode))
 		percpu_counter_inc(&sbi->s_dirs_counter);
-	sb->s_dirt = 1;
+	ext4_mark_super_dirty(sb);
 
 	if (sbi->s_log_groups_per_flex) {
 		flex_group = ext4_flex_group(sbi, group);

fs/ext4/inode.c

@@ -221,6 +221,7 @@ void ext4_delete_inode(struct inode *inode)
 				     "couldn't extend journal (err %d)", err);
 		stop_handle:
 			ext4_journal_stop(handle);
+			ext4_orphan_del(NULL, inode);
 			goto no_delete;
 		}
 	}
@@ -337,9 +338,11 @@ static int ext4_block_to_path(struct inode *inode,
 	return n;
 }
 
-static int __ext4_check_blockref(const char *function, struct inode *inode,
+static int __ext4_check_blockref(const char *function, unsigned int line,
+				 struct inode *inode,
 				 __le32 *p, unsigned int max)
 {
+	struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es;
 	__le32 *bref = p;
 	unsigned int blk;
 
@@ -348,8 +351,9 @@ static int __ext4_check_blockref(const char *function, struct inode *inode,
 		if (blk &&
 		    unlikely(!ext4_data_block_valid(EXT4_SB(inode->i_sb),
 						    blk, 1))) {
-			ext4_error_inode(function, inode,
-					 "invalid block reference %u", blk);
+			es->s_last_error_block = cpu_to_le64(blk);
+			ext4_error_inode(inode, function, line, blk,
+					 "invalid block");
 			return -EIO;
 		}
 	}
@@ -358,11 +362,13 @@ static int __ext4_check_blockref(const char *function, struct inode *inode,
 
 
 #define ext4_check_indirect_blockref(inode, bh)                         \
-	__ext4_check_blockref(__func__, inode, (__le32 *)(bh)->b_data,  \
+	__ext4_check_blockref(__func__, __LINE__, inode,		\
+			      (__le32 *)(bh)->b_data,			\
 			      EXT4_ADDR_PER_BLOCK((inode)->i_sb))
 
 #define ext4_check_inode_blockref(inode)                                \
-	__ext4_check_blockref(__func__, inode, EXT4_I(inode)->i_data,   \
+	__ext4_check_blockref(__func__, __LINE__, inode,		\
+			      EXT4_I(inode)->i_data,			\
 			      EXT4_NDIR_BLOCKS)
 
 /**
@@ -1128,20 +1134,24 @@ void ext4_da_update_reserve_space(struct inode *inode,
 		ext4_discard_preallocations(inode);
 }
 
-static int check_block_validity(struct inode *inode, const char *func,
+static int __check_block_validity(struct inode *inode, const char *func,
+				unsigned int line,
 				struct ext4_map_blocks *map)
 {
 	if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), map->m_pblk,
 				   map->m_len)) {
-		ext4_error_inode(func, inode,
-			   "lblock %lu mapped to illegal pblock %llu "
-			   "(length %d)", (unsigned long) map->m_lblk,
-				 map->m_pblk, map->m_len);
+		ext4_error_inode(inode, func, line, map->m_pblk,
+				 "lblock %lu mapped to illegal pblock "
+				 "(length %d)", (unsigned long) map->m_lblk,
+				 map->m_len);
 		return -EIO;
 	}
 	return 0;
 }
 
+#define check_block_validity(inode, map)	\
+	__check_block_validity((inode), __func__, __LINE__, (map))
+
 /*
  * Return the number of contiguous dirty pages in a given inode
  * starting at page frame idx.
@@ -1244,7 +1254,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 	up_read((&EXT4_I(inode)->i_data_sem));
 
 	if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) {
-		int ret = check_block_validity(inode, __func__, map);
+		int ret = check_block_validity(inode, map);
 		if (ret != 0)
 			return ret;
 	}
@@ -1324,9 +1334,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 
 	up_write((&EXT4_I(inode)->i_data_sem));
 	if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) {
-		int ret = check_block_validity(inode,
-					       "ext4_map_blocks_after_alloc",
-					       map);
+		int ret = check_block_validity(inode, map);
 		if (ret != 0)
 			return ret;
 	}
@@ -1519,9 +1527,25 @@ static int walk_page_buffers(handle_t *handle,
 static int do_journal_get_write_access(handle_t *handle,
 				       struct buffer_head *bh)
 {
+	int dirty = buffer_dirty(bh);
+	int ret;
+
 	if (!buffer_mapped(bh) || buffer_freed(bh))
 		return 0;
-	return ext4_journal_get_write_access(handle, bh);
+	/*
+	 * __block_prepare_write() could have dirtied some buffers. Clean
+	 * the dirty bit as jbd2_journal_get_write_access() could complain
+	 * otherwise about fs integrity issues. Setting of the dirty bit
+	 * by __block_prepare_write() isn't a real problem here as we clear
+	 * the bit before releasing a page lock and thus writeback cannot
+	 * ever write the buffer.
+	 */
+	if (dirty)
+		clear_buffer_dirty(bh);
+	ret = ext4_journal_get_write_access(handle, bh);
+	if (!ret && dirty)
+		ret = ext4_handle_dirty_metadata(handle, NULL, bh);
+	return ret;
 }
 
 /*
@@ -2194,7 +2218,7 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
 	BUG_ON(!handle);
 
 	/*
-	 * Call ext4_get_blocks() to allocate any delayed allocation
+	 * Call ext4_map_blocks() to allocate any delayed allocation
 	 * blocks, or to convert an uninitialized extent to be
 	 * initialized (in the case where we have written into
 	 * one or more preallocated blocks).
@@ -2203,7 +2227,7 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
 	 * indicate that we are on the delayed allocation path.  This
 	 * 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_get_blocks()
+	 * want to change *many* call functions, so ext4_map_blocks()
 	 * will set the magic i_delalloc_reserved_flag once the
 	 * inode's allocation semaphore is taken.
 	 *
@@ -2221,6 +2245,8 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
 
 	blks = ext4_map_blocks(handle, mpd->inode, &map, get_blocks_flags);
 	if (blks < 0) {
+		struct super_block *sb = mpd->inode->i_sb;
+
 		err = blks;
 		/*
 		 * If get block returns with error we simply
@@ -2231,7 +2257,7 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
 			return 0;
 
 		if (err == -ENOSPC &&
-		    ext4_count_free_blocks(mpd->inode->i_sb)) {
+		    ext4_count_free_blocks(sb)) {
 			mpd->retval = err;
 			return 0;
 		}
@@ -2243,16 +2269,17 @@ static int mpage_da_map_blocks(struct mpage_da_data *mpd)
 		 * writepage and writepages will again try to write
 		 * the same.
 		 */
-		ext4_msg(mpd->inode->i_sb, KERN_CRIT,
-			 "delayed block allocation failed for inode %lu at "
-			 "logical offset %llu with max blocks %zd with "
-			 "error %d", mpd->inode->i_ino,
-			 (unsigned long long) next,
-			 mpd->b_size >> mpd->inode->i_blkbits, err);
-		printk(KERN_CRIT "This should not happen!!  "
-		       "Data will be lost\n");
-		if (err == -ENOSPC) {
-			ext4_print_free_blocks(mpd->inode);
+		if (!(EXT4_SB(sb)->s_mount_flags & EXT4_MF_FS_ABORTED)) {
+			ext4_msg(sb, KERN_CRIT,
+				 "delayed block allocation failed for inode %lu "
+				 "at logical offset %llu with max blocks %zd "
+				 "with error %d", mpd->inode->i_ino,
+				 (unsigned long long) next,
+				 mpd->b_size >> mpd->inode->i_blkbits, err);
+			ext4_msg(sb, KERN_CRIT,
+				"This should not happen!! Data will be lost\n");
+			if (err == -ENOSPC)
+				ext4_print_free_blocks(mpd->inode);
 		}
 		/* invalidate all the pages */
 		ext4_da_block_invalidatepages(mpd, next,
@@ -2320,7 +2347,7 @@ static void mpage_add_bh_to_extent(struct mpage_da_data *mpd,
 	 * XXX Don't go larger than mballoc is willing to allocate
 	 * This is a stopgap solution.  We eventually need to fold
 	 * mpage_da_submit_io() into this function and then call
-	 * ext4_get_blocks() multiple times in a loop
+	 * ext4_map_blocks() multiple times in a loop
 	 */
 	if (nrblocks >= 8*1024*1024/mpd->inode->i_sb->s_blocksize)
 		goto flush_it;
@@ -2553,18 +2580,16 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock,
 /*
  * This function is used as a standard get_block_t calback function
  * when there is no desire to allocate any blocks.  It is used as a
- * callback function for block_prepare_write(), nobh_writepage(), and
- * block_write_full_page().  These functions should only try to map a
- * single block at a time.
+ * callback function for block_prepare_write() and block_write_full_page().
+ * These functions should only try to map a single block at a time.
  *
  * Since this function doesn't do block allocations even if the caller
  * requests it by passing in create=1, it is critically important that
  * any caller checks to make sure that any buffer heads are returned
  * by this function are either all already mapped or marked for
- * delayed allocation before calling nobh_writepage() or
- * block_write_full_page().  Otherwise, b_blocknr could be left
- * unitialized, and the page write functions will be taken by
- * surprise.
+ * delayed allocation before calling  block_write_full_page().  Otherwise,
+ * b_blocknr could be left unitialized, and the page write functions will
+ * be taken by surprise.
  */
 static int noalloc_get_block_write(struct inode *inode, sector_t iblock,
 				   struct buffer_head *bh_result, int create)
@@ -2749,9 +2774,7 @@ static int ext4_writepage(struct page *page,
 		return __ext4_journalled_writepage(page, len);
 	}
 
-	if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode))
-		ret = nobh_writepage(page, noalloc_get_block_write, wbc);
-	else if (page_bufs && buffer_uninit(page_bufs)) {
+	if (page_bufs && buffer_uninit(page_bufs)) {
 		ext4_set_bh_endio(page_bufs, inode);
 		ret = block_write_full_page_endio(page, noalloc_get_block_write,
 					    wbc, ext4_end_io_buffer_write);
@@ -3146,13 +3169,10 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
 	int ret, retries = 0;
 	struct page *page;
 	pgoff_t index;
-	unsigned from, to;
 	struct inode *inode = mapping->host;
 	handle_t *handle;
 
 	index = pos >> PAGE_CACHE_SHIFT;
-	from = pos & (PAGE_CACHE_SIZE - 1);
-	to = from + len;
 
 	if (ext4_nonda_switch(inode->i_sb)) {
 		*fsdata = (void *)FALL_BACK_TO_NONDELALLOC;
@@ -3668,6 +3688,8 @@ static int ext4_end_io_nolock(ext4_io_end_t *io)
 		return ret;
 	}
 
+	if (io->iocb)
+		aio_complete(io->iocb, io->result, 0);
 	/* clear the DIO AIO unwritten flag */
 	io->flag = 0;
 	return ret;
@@ -3767,6 +3789,8 @@ static ext4_io_end_t *ext4_init_io_end (struct inode *inode, gfp_t flags)
 		io->offset = 0;
 		io->size = 0;
 		io->page = NULL;
+		io->iocb = NULL;
+		io->result = 0;
 		INIT_WORK(&io->work, ext4_end_io_work);
 		INIT_LIST_HEAD(&io->list);
 	}
@@ -3796,12 +3820,18 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
 	if (io_end->flag != EXT4_IO_UNWRITTEN){
 		ext4_free_io_end(io_end);
 		iocb->private = NULL;
-		goto out;
+out:
+		if (is_async)
+			aio_complete(iocb, ret, 0);
+		return;
 	}
 
 	io_end->offset = offset;
 	io_end->size = size;
-	io_end->flag = EXT4_IO_UNWRITTEN;
+	if (is_async) {
+		io_end->iocb = iocb;
+		io_end->result = ret;
+	}
 	wq = EXT4_SB(io_end->inode->i_sb)->dio_unwritten_wq;
 
 	/* queue the work to convert unwritten extents to written */
@@ -3813,9 +3843,6 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
 	list_add_tail(&io_end->list, &ei->i_completed_io_list);
 	spin_unlock_irqrestore(&ei->i_completed_io_lock, flags);
 	iocb->private = NULL;
-out:
-	if (is_async)
-		aio_complete(iocb, ret, 0);
 }
 
 static void ext4_end_io_buffer_write(struct buffer_head *bh, int uptodate)
@@ -3941,7 +3968,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb,
 				return -ENOMEM;
 			/*
 			 * we save the io structure for current async
-			 * direct IO, so that later ext4_get_blocks()
+			 * direct IO, so that later ext4_map_blocks()
 			 * could flag the io structure whether there
 			 * is a unwritten extents needs to be converted
 			 * when IO is completed.
@@ -4132,17 +4159,6 @@ int ext4_block_truncate_page(handle_t *handle,
 	length = blocksize - (offset & (blocksize - 1));
 	iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits);
 
-	/*
-	 * For "nobh" option,  we can only work if we don't need to
-	 * read-in the page - otherwise we create buffers to do the IO.
-	 */
-	if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) &&
-	     ext4_should_writeback_data(inode) && PageUptodate(page)) {
-		zero_user(page, offset, length);
-		set_page_dirty(page);
-		goto unlock;
-	}
-
 	if (!page_has_buffers(page))
 		create_empty_buffers(page, blocksize, 0);
 
@@ -4492,9 +4508,8 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 			 * (should be rare).
 			 */
 			if (!bh) {
-				EXT4_ERROR_INODE(inode,
-						 "Read failure block=%llu",
-						 (unsigned long long) nr);
+				EXT4_ERROR_INODE_BLOCK(inode, nr,
+						       "Read failure");
 				continue;
 			}
 
@@ -4505,27 +4520,6 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 					(__le32 *) bh->b_data + addr_per_block,
 					depth);
 
-			/*
-			 * We've probably journalled the indirect block several
-			 * times during the truncate.  But it's no longer
-			 * needed and we now drop it from the transaction via
-			 * jbd2_journal_revoke().
-			 *
-			 * That's easy if it's exclusively part of this
-			 * transaction.  But if it's part of the committing
-			 * transaction then jbd2_journal_forget() will simply
-			 * brelse() it.  That means that if the underlying
-			 * block is reallocated in ext4_get_block(),
-			 * unmap_underlying_metadata() will find this block
-			 * and will try to get rid of it.  damn, damn.
-			 *
-			 * If this block has already been committed to the
-			 * journal, a revoke record will be written.  And
-			 * revoke records must be emitted *before* clearing
-			 * this block's bit in the bitmaps.
-			 */
-			ext4_forget(handle, 1, inode, bh, bh->b_blocknr);
-
 			/*
 			 * Everything below this this pointer has been
 			 * released.  Now let this top-of-subtree go.
@@ -4550,8 +4544,20 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
 					    blocks_for_truncate(inode));
 			}
 
+			/*
+			 * The forget flag here is critical because if
+			 * we are journaling (and not doing data
+			 * journaling), we have to make sure a revoke
+			 * record is written to prevent the journal
+			 * replay from overwriting the (former)
+			 * indirect block if it gets reallocated as a
+			 * data block.  This must happen in the same
+			 * transaction where the data blocks are
+			 * actually freed.
+			 */
 			ext4_free_blocks(handle, inode, 0, nr, 1,
-					 EXT4_FREE_BLOCKS_METADATA);
+					 EXT4_FREE_BLOCKS_METADATA|
+					 EXT4_FREE_BLOCKS_FORGET);
 
 			if (parent_bh) {
 				/*
@@ -4809,8 +4815,8 @@ static int __ext4_get_inode_loc(struct inode *inode,
 
 	bh = sb_getblk(sb, block);
 	if (!bh) {
-		EXT4_ERROR_INODE(inode, "unable to read inode block - "
-				 "block %llu", block);
+		EXT4_ERROR_INODE_BLOCK(inode, block,
+				       "unable to read itable block");
 		return -EIO;
 	}
 	if (!buffer_uptodate(bh)) {
@@ -4908,8 +4914,8 @@ static int __ext4_get_inode_loc(struct inode *inode,
 		submit_bh(READ_META, bh);
 		wait_on_buffer(bh);
 		if (!buffer_uptodate(bh)) {
-			EXT4_ERROR_INODE(inode, "unable to read inode "
-					 "block %llu", block);
+			EXT4_ERROR_INODE_BLOCK(inode, block,
+					       "unable to read itable block");
 			brelse(bh);
 			return -EIO;
 		}
@@ -4980,7 +4986,7 @@ static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode,
 		/* we are using combined 48 bit field */
 		i_blocks = ((u64)le16_to_cpu(raw_inode->i_blocks_high)) << 32 |
 					le32_to_cpu(raw_inode->i_blocks_lo);
-		if (ei->i_flags & EXT4_HUGE_FILE_FL) {
+		if (ext4_test_inode_flag(inode, EXT4_INODE_HUGE_FILE)) {
 			/* i_blocks represent file system block size */
 			return i_blocks  << (inode->i_blkbits - 9);
 		} else {
@@ -5076,7 +5082,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		transaction_t *transaction;
 		tid_t tid;
 
-		spin_lock(&journal->j_state_lock);
+		read_lock(&journal->j_state_lock);
 		if (journal->j_running_transaction)
 			transaction = journal->j_running_transaction;
 		else
@@ -5085,7 +5091,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 			tid = transaction->t_tid;
 		else
 			tid = journal->j_commit_sequence;
-		spin_unlock(&journal->j_state_lock);
+		read_unlock(&journal->j_state_lock);
 		ei->i_sync_tid = tid;
 		ei->i_datasync_tid = tid;
 	}
@@ -5130,7 +5136,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 				 ei->i_file_acl);
 		ret = -EIO;
 		goto bad_inode;
-	} else if (ei->i_flags & EXT4_EXTENTS_FL) {
+	} else if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
 		if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
 		    (S_ISLNK(inode->i_mode) &&
 		     !ext4_inode_is_fast_symlink(inode)))
@@ -5410,9 +5416,8 @@ int ext4_write_inode(struct inode *inode, struct writeback_control *wbc)
 		if (wbc->sync_mode == WB_SYNC_ALL)
 			sync_dirty_buffer(iloc.bh);
 		if (buffer_req(iloc.bh) && !buffer_uptodate(iloc.bh)) {
-			EXT4_ERROR_INODE(inode,
-				"IO error syncing inode (block=%llu)",
-				(unsigned long long) iloc.bh->b_blocknr);
+			EXT4_ERROR_INODE_BLOCK(inode, iloc.bh->b_blocknr,
+					 "IO error syncing inode");
 			err = -EIO;
 		}
 		brelse(iloc.bh);
@@ -5487,10 +5492,8 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
 		if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) {
 			struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 
-			if (attr->ia_size > sbi->s_bitmap_maxbytes) {
-				error = -EFBIG;
-				goto err_out;
-			}
+			if (attr->ia_size > sbi->s_bitmap_maxbytes)
+				return -EFBIG;
 		}
 	}
 
@@ -5692,7 +5695,7 @@ int ext4_writepage_trans_blocks(struct inode *inode)
  * Calculate the journal credits for a chunk of data modification.
  *
  * This is called from DIO, fallocate or whoever calling
- * ext4_get_blocks() to map/allocate a chunk of contiguous disk blocks.
+ * ext4_map_blocks() to map/allocate a chunk of contiguous disk blocks.
  *
  * journal buffers for data blocks are not included here, as DIO
  * and fallocate do no need to journal data buffers.
@@ -5758,7 +5761,6 @@ static int ext4_expand_extra_isize(struct inode *inode,
 {
 	struct ext4_inode *raw_inode;
 	struct ext4_xattr_ibody_header *header;
-	struct ext4_xattr_entry *entry;
 
 	if (EXT4_I(inode)->i_extra_isize >= new_extra_isize)
 		return 0;
@@ -5766,7 +5768,6 @@ static int ext4_expand_extra_isize(struct inode *inode,
 	raw_inode = ext4_raw_inode(&iloc);
 
 	header = IHDR(inode, raw_inode);
-	entry = IFIRST(header);
 
 	/* No extended attributes present */
 	if (!ext4_test_inode_state(inode, EXT4_STATE_XATTR) ||

fs/ext4/mballoc.c

@@ -446,10 +446,11 @@ static void mb_free_blocks_double(struct inode *inode, struct ext4_buddy *e4b,
 			blocknr = ext4_group_first_block_no(sb, e4b->bd_group);
 			blocknr += first + i;
 			ext4_grp_locked_error(sb, e4b->bd_group,
-				   __func__, "double-free of inode"
-				   " %lu's block %llu(bit %u in group %u)",
-				   inode ? inode->i_ino : 0, blocknr,
-				   first + i, e4b->bd_group);
+					      inode ? inode->i_ino : 0,
+					      blocknr,
+					      "freeing block already freed "
+					      "(bit %u)",
+					      first + i);
 		}
 		mb_clear_bit(first + i, e4b->bd_info->bb_bitmap);
 	}
@@ -712,9 +713,9 @@ void ext4_mb_generate_buddy(struct super_block *sb,
 	grp->bb_fragments = fragments;
 
 	if (free != grp->bb_free) {
-		ext4_grp_locked_error(sb, group,  __func__,
-			"EXT4-fs: group %u: %u blocks in bitmap, %u in gd",
-			group, free, grp->bb_free);
+		ext4_grp_locked_error(sb, group, 0, 0,
+				      "%u blocks in bitmap, %u in gd",
+				      free, grp->bb_free);
 		/*
 		 * If we intent to continue, we consider group descritor
 		 * corrupt and update bb_free using bitmap value
@@ -1296,10 +1297,10 @@ static void mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b,
 			blocknr = ext4_group_first_block_no(sb, e4b->bd_group);
 			blocknr += block;
 			ext4_grp_locked_error(sb, e4b->bd_group,
-				   __func__, "double-free of inode"
-				   " %lu's block %llu(bit %u in group %u)",
-				   inode ? inode->i_ino : 0, blocknr, block,
-				   e4b->bd_group);
+					      inode ? inode->i_ino : 0,
+					      blocknr,
+					      "freeing already freed block "
+					      "(bit %u)", block);
 		}
 		mb_clear_bit(block, EXT4_MB_BITMAP(e4b));
 		e4b->bd_info->bb_counters[order]++;
@@ -1788,8 +1789,8 @@ void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac,
 			 * free blocks even though group info says we
 			 * we have free blocks
 			 */
-			ext4_grp_locked_error(sb, e4b->bd_group,
-					__func__, "%d free blocks as per "
+			ext4_grp_locked_error(sb, e4b->bd_group, 0, 0,
+					"%d free blocks as per "
 					"group info. But bitmap says 0",
 					free);
 			break;
@@ -1798,8 +1799,8 @@ void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac,
 		mb_find_extent(e4b, 0, i, ac->ac_g_ex.fe_len, &ex);
 		BUG_ON(ex.fe_len <= 0);
 		if (free < ex.fe_len) {
-			ext4_grp_locked_error(sb, e4b->bd_group,
-					__func__, "%d free blocks as per "
+			ext4_grp_locked_error(sb, e4b->bd_group, 0, 0,
+					"%d free blocks as per "
 					"group info. But got %d blocks",
 					free, ex.fe_len);
 			/*
@@ -1821,8 +1822,7 @@ void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac,
 
 /*
  * This is a special case for storages like raid5
- * we try to find stripe-aligned chunks for stripe-size requests
- * XXX should do so at least for multiples of stripe size as well
+ * we try to find stripe-aligned chunks for stripe-size-multiple requests
  */
 static noinline_for_stack
 void ext4_mb_scan_aligned(struct ext4_allocation_context *ac,
@@ -1999,7 +1999,6 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
 	ext4_group_t ngroups, group, i;
 	int cr;
 	int err = 0;
-	int bsbits;
 	struct ext4_sb_info *sbi;
 	struct super_block *sb;
 	struct ext4_buddy e4b;
@@ -2041,8 +2040,6 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
 			ac->ac_2order = i - 1;
 	}
 
-	bsbits = ac->ac_sb->s_blocksize_bits;
-
 	/* if stream allocation is enabled, use global goal */
 	if (ac->ac_flags & EXT4_MB_STREAM_ALLOC) {
 		/* TBD: may be hot point */
@@ -2094,8 +2091,8 @@ ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
 			ac->ac_groups_scanned++;
 			if (cr == 0)
 				ext4_mb_simple_scan_group(ac, &e4b);
-			else if (cr == 1 &&
-					ac->ac_g_ex.fe_len == sbi->s_stripe)
+			else if (cr == 1 && sbi->s_stripe &&
+					!(ac->ac_g_ex.fe_len % sbi->s_stripe))
 				ext4_mb_scan_aligned(ac, &e4b);
 			else
 				ext4_mb_complex_scan_group(ac, &e4b);
@@ -2221,7 +2218,7 @@ static int ext4_mb_seq_groups_open(struct inode *inode, struct file *file)
 
 	rc = seq_open(file, &ext4_mb_seq_groups_ops);
 	if (rc == 0) {
-		struct seq_file *m = (struct seq_file *)file->private_data;
+		struct seq_file *m = file->private_data;
 		m->private = sb;
 	}
 	return rc;
@@ -2560,6 +2557,22 @@ int ext4_mb_release(struct super_block *sb)
 	return 0;
 }
 
+static inline void 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);
+	if (ret == EOPNOTSUPP) {
+		ext4_warning(sb, "discard not supported, disabling");
+		clear_opt(EXT4_SB(sb)->s_mount_opt, DISCARD);
+	}
+}
+
 /*
  * This function is called by the jbd2 layer once the commit has finished,
  * so we know we can free the blocks that were released with that commit.
@@ -2579,22 +2592,9 @@ 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)) {
-			int ret;
-			ext4_fsblk_t discard_block;
-
-			discard_block = entry->start_blk +
-				ext4_group_first_block_no(sb, entry->group);
-			trace_ext4_discard_blocks(sb,
-					(unsigned long long)discard_block,
-					entry->count);
-			ret = sb_issue_discard(sb, discard_block, entry->count);
-			if (ret == EOPNOTSUPP) {
-				ext4_warning(sb,
-					"discard not supported, disabling");
-				clear_opt(EXT4_SB(sb)->s_mount_opt, DISCARD);
-			}
-		}
+		if (test_opt(sb, DISCARD))
+			ext4_issue_discard(sb, entry->group,
+					entry->start_blk, entry->count);
 
 		err = ext4_mb_load_buddy(sb, entry->group, &e4b);
 		/* we expect to find existing buddy because it's pinned */
@@ -2712,7 +2712,6 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 				handle_t *handle, unsigned int reserv_blks)
 {
 	struct buffer_head *bitmap_bh = NULL;
-	struct ext4_super_block *es;
 	struct ext4_group_desc *gdp;
 	struct buffer_head *gdp_bh;
 	struct ext4_sb_info *sbi;
@@ -2725,8 +2724,6 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 
 	sb = ac->ac_sb;
 	sbi = EXT4_SB(sb);
-	es = sbi->s_es;
-
 
 	err = -EIO;
 	bitmap_bh = ext4_read_block_bitmap(sb, ac->ac_b_ex.fe_group);
@@ -2812,7 +2809,7 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 	err = ext4_handle_dirty_metadata(handle, NULL, gdp_bh);
 
 out_err:
-	sb->s_dirt = 1;
+	ext4_mark_super_dirty(sb);
 	brelse(bitmap_bh);
 	return err;
 }
@@ -2850,7 +2847,7 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac,
 	int bsbits, max;
 	ext4_lblk_t end;
 	loff_t size, orig_size, start_off;
-	ext4_lblk_t start, orig_start;
+	ext4_lblk_t start;
 	struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
 	struct ext4_prealloc_space *pa;
 
@@ -2881,6 +2878,7 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac,
 	size = size << bsbits;
 	if (size < i_size_read(ac->ac_inode))
 		size = i_size_read(ac->ac_inode);
+	orig_size = size;
 
 	/* max size of free chunks */
 	max = 2 << bsbits;
@@ -2922,8 +2920,8 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac,
 		start_off = (loff_t)ac->ac_o_ex.fe_logical << bsbits;
 		size	  = ac->ac_o_ex.fe_len << bsbits;
 	}
-	orig_size = size = size >> bsbits;
-	orig_start = start = start_off >> bsbits;
+	size = size >> bsbits;
+	start = start_off >> bsbits;
 
 	/* don't cover already allocated blocks in selected range */
 	if (ar->pleft && start <= ar->lleft) {
@@ -3547,7 +3545,6 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh,
 	ext4_group_t group;
 	ext4_grpblk_t bit;
 	unsigned long long grp_blk_start;
-	sector_t start;
 	int err = 0;
 	int free = 0;
 
@@ -3567,10 +3564,9 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh,
 		if (bit >= end)
 			break;
 		next = mb_find_next_bit(bitmap_bh->b_data, end, bit);
-		start = ext4_group_first_block_no(sb, group) + bit;
 		mb_debug(1, "    free preallocated %u/%u in group %u\n",
-				(unsigned) start, (unsigned) next - bit,
-				(unsigned) group);
+			 (unsigned) ext4_group_first_block_no(sb, group) + bit,
+			 (unsigned) next - bit, (unsigned) group);
 		free += next - bit;
 
 		if (ac) {
@@ -3581,7 +3577,7 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh,
 			trace_ext4_mballoc_discard(ac);
 		}
 
-		trace_ext4_mb_release_inode_pa(ac, pa, grp_blk_start + bit,
+		trace_ext4_mb_release_inode_pa(sb, ac, pa, grp_blk_start + bit,
 					       next - bit);
 		mb_free_blocks(pa->pa_inode, e4b, bit, next - bit);
 		bit = next + 1;
@@ -3591,8 +3587,7 @@ ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh,
 			pa, (unsigned long) pa->pa_lstart,
 			(unsigned long) pa->pa_pstart,
 			(unsigned long) pa->pa_len);
-		ext4_grp_locked_error(sb, group,
-					__func__, "free %u, pa_free %u",
+		ext4_grp_locked_error(sb, group, 0, 0, "free %u, pa_free %u",
 					free, pa->pa_free);
 		/*
 		 * pa is already deleted so we use the value obtained
@@ -3613,7 +3608,7 @@ ext4_mb_release_group_pa(struct ext4_buddy *e4b,
 	ext4_group_t group;
 	ext4_grpblk_t bit;
 
-	trace_ext4_mb_release_group_pa(ac, pa);
+	trace_ext4_mb_release_group_pa(sb, ac, pa);
 	BUG_ON(pa->pa_deleted == 0);
 	ext4_get_group_no_and_offset(sb, pa->pa_pstart, &group, &bit);
 	BUG_ON(group != e4b->bd_group && pa->pa_len != 0);
@@ -3889,6 +3884,9 @@ static void ext4_mb_show_ac(struct ext4_allocation_context *ac)
 	struct super_block *sb = ac->ac_sb;
 	ext4_group_t ngroups, i;
 
+	if (EXT4_SB(sb)->s_mount_flags & EXT4_MF_FS_ABORTED)
+		return;
+
 	printk(KERN_ERR "EXT4-fs: Can't allocate:"
 			" Allocation context details:\n");
 	printk(KERN_ERR "EXT4-fs: status %d flags %d\n",
@@ -4255,7 +4253,7 @@ static int ext4_mb_discard_preallocations(struct super_block *sb, int needed)
  * to usual allocation
  */
 ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
-				 struct ext4_allocation_request *ar, int *errp)
+				struct ext4_allocation_request *ar, int *errp)
 {
 	int freed;
 	struct ext4_allocation_context *ac = NULL;
@@ -4299,7 +4297,7 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 		inquota = ar->len;
 		if (ar->len == 0) {
 			*errp = -EDQUOT;
-			goto out3;
+			goto out;
 		}
 	}
 
@@ -4307,13 +4305,13 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 	if (!ac) {
 		ar->len = 0;
 		*errp = -ENOMEM;
-		goto out1;
+		goto out;
 	}
 
 	*errp = ext4_mb_initialize_context(ac, ar);
 	if (*errp) {
 		ar->len = 0;
-		goto out2;
+		goto out;
 	}
 
 	ac->ac_op = EXT4_MB_HISTORY_PREALLOC;
@@ -4322,7 +4320,9 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 		ext4_mb_normalize_request(ac, ar);
 repeat:
 		/* allocate space in core */
-		ext4_mb_regular_allocator(ac);
+		*errp = ext4_mb_regular_allocator(ac);
+		if (*errp)
+			goto errout;
 
 		/* as we've just preallocated more space than
 		 * user requested orinally, we store allocated
@@ -4333,7 +4333,7 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 	}
 	if (likely(ac->ac_status == AC_STATUS_FOUND)) {
 		*errp = ext4_mb_mark_diskspace_used(ac, handle, reserv_blks);
-		if (*errp ==  -EAGAIN) {
+		if (*errp == -EAGAIN) {
 			/*
 			 * drop the reference that we took
 			 * in ext4_mb_use_best_found
@@ -4344,12 +4344,10 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 			ac->ac_b_ex.fe_len = 0;
 			ac->ac_status = AC_STATUS_CONTINUE;
 			goto repeat;
-		} else if (*errp) {
+		} else if (*errp)
+		errout:
 			ext4_discard_allocated_blocks(ac);
-			ac->ac_b_ex.fe_len = 0;
-			ar->len = 0;
-			ext4_mb_show_ac(ac);
-		} else {
+		else {
 			block = ext4_grp_offs_to_block(sb, &ac->ac_b_ex);
 			ar->len = ac->ac_b_ex.fe_len;
 		}
@@ -4358,19 +4356,19 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 		if (freed)
 			goto repeat;
 		*errp = -ENOSPC;
+	}
+
+	if (*errp) {
 		ac->ac_b_ex.fe_len = 0;
 		ar->len = 0;
 		ext4_mb_show_ac(ac);
 	}
-
 	ext4_mb_release_context(ac);
-
-out2:
-	kmem_cache_free(ext4_ac_cachep, ac);
-out1:
+out:
+	if (ac)
+		kmem_cache_free(ext4_ac_cachep, ac);
 	if (inquota && ar->len < inquota)
 		dquot_free_block(ar->inode, inquota - ar->len);
-out3:
 	if (!ar->len) {
 		if (!EXT4_I(ar->inode)->i_delalloc_reserved_flag)
 			/* release all the reserved blocks if non delalloc */
@@ -4402,6 +4400,7 @@ static noinline_for_stack int
 ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b,
 		      struct ext4_free_data *new_entry)
 {
+	ext4_group_t group = e4b->bd_group;
 	ext4_grpblk_t block;
 	struct ext4_free_data *entry;
 	struct ext4_group_info *db = e4b->bd_info;
@@ -4434,9 +4433,9 @@ 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_grp_locked_error(sb, e4b->bd_group, __func__,
-					"Double free of blocks %d (%d %d)",
-					block, entry->start_blk, entry->count);
+			ext4_grp_locked_error(sb, group, 0,
+				ext4_group_first_block_no(sb, group) + block,
+				"Block already on to-be-freed list");
 			return 0;
 		}
 	}
@@ -4494,7 +4493,6 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 	struct super_block *sb = inode->i_sb;
 	struct ext4_allocation_context *ac = NULL;
 	struct ext4_group_desc *gdp;
-	struct ext4_super_block *es;
 	unsigned long freed = 0;
 	unsigned int overflow;
 	ext4_grpblk_t bit;
@@ -4513,7 +4511,6 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 	}
 
 	sbi = EXT4_SB(sb);
-	es = EXT4_SB(sb)->s_es;
 	if (!(flags & EXT4_FREE_BLOCKS_VALIDATED) &&
 	    !ext4_data_block_valid(sbi, block, count)) {
 		ext4_error(sb, "Freeing blocks not in datazone - "
@@ -4647,6 +4644,8 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 		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);
+		if (test_opt(sb, DISCARD))
+			ext4_issue_discard(sb, block_group, bit, count);
 	}
 
 	ret = ext4_free_blks_count(sb, gdp) + count;
@@ -4680,7 +4679,7 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 		put_bh(bitmap_bh);
 		goto do_more;
 	}
-	sb->s_dirt = 1;
+	ext4_mark_super_dirty(sb);
 error_return:
 	if (freed)
 		dquot_free_block(inode, freed);

fs/ext4/migrate.c

@@ -376,7 +376,7 @@ static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
 	 * We have the extent map build with the tmp inode.
 	 * Now copy the i_data across
 	 */
-	ei->i_flags |= EXT4_EXTENTS_FL;
+	ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS);
 	memcpy(ei->i_data, tmp_ei->i_data, sizeof(ei->i_data));
 
 	/*

fs/ext4/move_extent.c

@@ -148,17 +148,17 @@ mext_next_extent(struct inode *inode, struct ext4_ext_path *path,
  */
 static int
 mext_check_null_inode(struct inode *inode1, struct inode *inode2,
-		const char *function)
+		      const char *function, unsigned int line)
 {
 	int ret = 0;
 
 	if (inode1 == NULL) {
-		__ext4_error(inode2->i_sb, function,
+		__ext4_error(inode2->i_sb, function, line,
 			"Both inodes should not be NULL: "
 			"inode1 NULL inode2 %lu", inode2->i_ino);
 		ret = -EIO;
 	} else if (inode2 == NULL) {
-		__ext4_error(inode1->i_sb, function,
+		__ext4_error(inode1->i_sb, function, line,
 			"Both inodes should not be NULL: "
 			"inode1 %lu inode2 NULL", inode1->i_ino);
 		ret = -EIO;
@@ -1084,7 +1084,7 @@ mext_inode_double_lock(struct inode *inode1, struct inode *inode2)
 
 	BUG_ON(inode1 == NULL && inode2 == NULL);
 
-	ret = mext_check_null_inode(inode1, inode2, __func__);
+	ret = mext_check_null_inode(inode1, inode2, __func__, __LINE__);
 	if (ret < 0)
 		goto out;
 
@@ -1121,7 +1121,7 @@ mext_inode_double_unlock(struct inode *inode1, struct inode *inode2)
 
 	BUG_ON(inode1 == NULL && inode2 == NULL);
 
-	ret = mext_check_null_inode(inode1, inode2, __func__);
+	ret = mext_check_null_inode(inode1, inode2, __func__, __LINE__);
 	if (ret < 0)
 		goto out;
 

fs/ext4/namei.c

@@ -179,30 +179,6 @@ static struct buffer_head * ext4_dx_find_entry(struct inode *dir,
 static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 			     struct inode *inode);
 
-unsigned int ext4_rec_len_from_disk(__le16 dlen, unsigned blocksize)
-{
-	unsigned len = le16_to_cpu(dlen);
-
-	if (len == EXT4_MAX_REC_LEN || len == 0)
-		return blocksize;
-	return (len & 65532) | ((len & 3) << 16);
-}
-
-__le16 ext4_rec_len_to_disk(unsigned len, unsigned blocksize)
-{
-	if ((len > blocksize) || (blocksize > (1 << 18)) || (len & 3))
-		BUG();
-	if (len < 65536)
-		return cpu_to_le16(len);
-	if (len == blocksize) {
-		if (blocksize == 65536)
-			return cpu_to_le16(EXT4_MAX_REC_LEN);
-		else
-			return cpu_to_le16(0);
-	}
-	return cpu_to_le16((len & 65532) | ((len >> 16) & 3));
-}
-
 /*
  * p is at least 6 bytes before the end of page
  */
@@ -605,7 +581,7 @@ 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("htree_dirblock_to_tree", dir, de, bh,
+		if (!ext4_check_dir_entry(dir, 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. */
@@ -844,8 +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("ext4_find_entry",
-						  dir, de, bh, offset))
+			if (!ext4_check_dir_entry(dir, de, bh, offset))
 				return -1;
 			*res_dir = de;
 			return 1;
@@ -1019,7 +994,7 @@ static struct buffer_head * ext4_dx_find_entry(struct inode *dir, const struct q
 			int off = (block << EXT4_BLOCK_SIZE_BITS(sb))
 				  + ((char *) de - bh->b_data);
 
-			if (!ext4_check_dir_entry(__func__, dir, de, bh, off)) {
+			if (!ext4_check_dir_entry(dir, de, bh, off)) {
 				brelse(bh);
 				*err = ERR_BAD_DX_DIR;
 				goto errout;
@@ -1088,7 +1063,6 @@ static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, stru
 struct dentry *ext4_get_parent(struct dentry *child)
 {
 	__u32 ino;
-	struct inode *inode;
 	static const struct qstr dotdot = {
 		.name = "..",
 		.len = 2,
@@ -1097,7 +1071,6 @@ struct dentry *ext4_get_parent(struct dentry *child)
 	struct buffer_head *bh;
 
 	bh = ext4_find_entry(child->d_inode, &dotdot, &de);
-	inode = NULL;
 	if (!bh)
 		return ERR_PTR(-ENOENT);
 	ino = le32_to_cpu(de->inode);
@@ -1305,8 +1278,7 @@ 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("ext4_add_entry", dir, de,
-						  bh, offset))
+			if (!ext4_check_dir_entry(dir, de, bh, offset))
 				return -EIO;
 			if (ext4_match(namelen, name, de))
 				return -EEXIST;
@@ -1673,7 +1645,7 @@ static int ext4_delete_entry(handle_t *handle,
 	pde = NULL;
 	de = (struct ext4_dir_entry_2 *) bh->b_data;
 	while (i < bh->b_size) {
-		if (!ext4_check_dir_entry("ext4_delete_entry", dir, de, bh, i))
+		if (!ext4_check_dir_entry(dir, de, bh, i))
 			return -EIO;
 		if (de == de_del)  {
 			BUFFER_TRACE(bh, "get_write_access");
@@ -1956,7 +1928,7 @@ static int empty_dir(struct inode *inode)
 			}
 			de = (struct ext4_dir_entry_2 *) bh->b_data;
 		}
-		if (!ext4_check_dir_entry("empty_dir", inode, de, bh, offset)) {
+		if (!ext4_check_dir_entry(inode, de, bh, offset)) {
 			de = (struct ext4_dir_entry_2 *)(bh->b_data +
 							 sb->s_blocksize);
 			offset = (offset | (sb->s_blocksize - 1)) + 1;

fs/ext4/resize.c

@@ -921,8 +921,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 			   &sbi->s_flex_groups[flex_group].free_inodes);
 	}
 
-	ext4_handle_dirty_metadata(handle, NULL, sbi->s_sbh);
-	sb->s_dirt = 1;
+	ext4_handle_dirty_super(handle, sb);
 
 exit_journal:
 	mutex_unlock(&sbi->s_resize_lock);
@@ -953,7 +952,6 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 		      ext4_fsblk_t n_blocks_count)
 {
 	ext4_fsblk_t o_blocks_count;
-	ext4_group_t o_groups_count;
 	ext4_grpblk_t last;
 	ext4_grpblk_t add;
 	struct buffer_head *bh;
@@ -965,7 +963,6 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 	 * yet: we're going to revalidate es->s_blocks_count after
 	 * taking the s_resize_lock below. */
 	o_blocks_count = ext4_blocks_count(es);
-	o_groups_count = EXT4_SB(sb)->s_groups_count;
 
 	if (test_opt(sb, DEBUG))
 		printk(KERN_DEBUG "EXT4-fs: extending last group from %llu uto %llu blocks\n",
@@ -1045,13 +1042,12 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 		goto exit_put;
 	}
 	ext4_blocks_count_set(es, o_blocks_count + add);
-	ext4_handle_dirty_metadata(handle, NULL, EXT4_SB(sb)->s_sbh);
-	sb->s_dirt = 1;
 	mutex_unlock(&EXT4_SB(sb)->s_resize_lock);
 	ext4_debug("freeing blocks %llu through %llu\n", o_blocks_count,
 		   o_blocks_count + add);
 	/* We add the blocks to the bitmap and set the group need init bit */
 	ext4_add_groupblocks(handle, sb, o_blocks_count, add);
+	ext4_handle_dirty_super(handle, sb);
 	ext4_debug("freed blocks %llu through %llu\n", o_blocks_count,
 		   o_blocks_count + add);
 	if ((err = ext4_journal_stop(handle)))

fs/ext4/xattr.c

@@ -458,8 +458,7 @@ static void ext4_xattr_update_super_block(handle_t *handle,
 
 	if (ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh) == 0) {
 		EXT4_SET_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_EXT_ATTR);
-		sb->s_dirt = 1;
-		ext4_handle_dirty_metadata(handle, NULL, EXT4_SB(sb)->s_sbh);
+		ext4_handle_dirty_super(handle, sb);
 	}
 }
 

fs/jbd2/checkpoint.c

@@ -118,13 +118,13 @@ static int __try_to_free_cp_buf(struct journal_head *jh)
 void __jbd2_log_wait_for_space(journal_t *journal)
 {
 	int nblocks, space_left;
-	assert_spin_locked(&journal->j_state_lock);
+	/* assert_spin_locked(&journal->j_state_lock); */
 
 	nblocks = jbd_space_needed(journal);
 	while (__jbd2_log_space_left(journal) < nblocks) {
 		if (journal->j_flags & JBD2_ABORT)
 			return;
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		mutex_lock(&journal->j_checkpoint_mutex);
 
 		/*
@@ -138,7 +138,7 @@ void __jbd2_log_wait_for_space(journal_t *journal)
 		 * filesystem, so abort the journal and leave a stack
 		 * trace for forensic evidence.
 		 */
-		spin_lock(&journal->j_state_lock);
+		write_lock(&journal->j_state_lock);
 		spin_lock(&journal->j_list_lock);
 		nblocks = jbd_space_needed(journal);
 		space_left = __jbd2_log_space_left(journal);
@@ -149,7 +149,7 @@ void __jbd2_log_wait_for_space(journal_t *journal)
 			if (journal->j_committing_transaction)
 				tid = journal->j_committing_transaction->t_tid;
 			spin_unlock(&journal->j_list_lock);
-			spin_unlock(&journal->j_state_lock);
+			write_unlock(&journal->j_state_lock);
 			if (chkpt) {
 				jbd2_log_do_checkpoint(journal);
 			} else if (jbd2_cleanup_journal_tail(journal) == 0) {
@@ -167,7 +167,7 @@ void __jbd2_log_wait_for_space(journal_t *journal)
 				WARN_ON(1);
 				jbd2_journal_abort(journal, 0);
 			}
-			spin_lock(&journal->j_state_lock);
+			write_lock(&journal->j_state_lock);
 		} else {
 			spin_unlock(&journal->j_list_lock);
 		}
@@ -474,7 +474,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	 * next transaction ID we will write, and where it will
 	 * start. */
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	spin_lock(&journal->j_list_lock);
 	transaction = journal->j_checkpoint_transactions;
 	if (transaction) {
@@ -496,7 +496,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	/* If the oldest pinned transaction is at the tail of the log
            already then there's not much we can do right now. */
 	if (journal->j_tail_sequence == first_tid) {
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		return 1;
 	}
 
@@ -516,7 +516,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	journal->j_free += freed;
 	journal->j_tail_sequence = first_tid;
 	journal->j_tail = blocknr;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 
 	/*
 	 * If there is an external journal, we need to make sure that
@@ -775,7 +775,7 @@ void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transact
 	J_ASSERT(transaction->t_log_list == NULL);
 	J_ASSERT(transaction->t_checkpoint_list == NULL);
 	J_ASSERT(transaction->t_checkpoint_io_list == NULL);
-	J_ASSERT(transaction->t_updates == 0);
+	J_ASSERT(atomic_read(&transaction->t_updates) == 0);
 	J_ASSERT(journal->j_committing_transaction != transaction);
 	J_ASSERT(journal->j_running_transaction != transaction);
 

fs/jbd2/commit.c

@@ -150,11 +150,11 @@ static int journal_submit_commit_record(journal_t *journal,
 	 */
 	if (ret == -EOPNOTSUPP && barrier_done) {
 		printk(KERN_WARNING
-		       "JBD: barrier-based sync failed on %s - "
-		       "disabling barriers\n", journal->j_devname);
-		spin_lock(&journal->j_state_lock);
+		       "JBD2: Disabling barriers on %s, "
+		       "not supported by device\n", journal->j_devname);
+		write_lock(&journal->j_state_lock);
 		journal->j_flags &= ~JBD2_BARRIER;
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 
 		/* And try again, without the barrier */
 		lock_buffer(bh);
@@ -180,11 +180,11 @@ static int journal_wait_on_commit_record(journal_t *journal,
 	wait_on_buffer(bh);
 	if (buffer_eopnotsupp(bh) && (journal->j_flags & JBD2_BARRIER)) {
 		printk(KERN_WARNING
-		       "JBD2: wait_on_commit_record: sync failed on %s - "
-		       "disabling barriers\n", journal->j_devname);
-		spin_lock(&journal->j_state_lock);
+		       "JBD2: %s: disabling barries on %s - not supported "
+		       "by device\n", __func__, journal->j_devname);
+		write_lock(&journal->j_state_lock);
 		journal->j_flags &= ~JBD2_BARRIER;
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 
 		lock_buffer(bh);
 		clear_buffer_dirty(bh);
@@ -400,7 +400,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	jbd_debug(1, "JBD: starting commit of transaction %d\n",
 			commit_transaction->t_tid);
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	commit_transaction->t_state = T_LOCKED;
 
 	/*
@@ -417,23 +417,23 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 					      stats.run.rs_locked);
 
 	spin_lock(&commit_transaction->t_handle_lock);
-	while (commit_transaction->t_updates) {
+	while (atomic_read(&commit_transaction->t_updates)) {
 		DEFINE_WAIT(wait);
 
 		prepare_to_wait(&journal->j_wait_updates, &wait,
 					TASK_UNINTERRUPTIBLE);
-		if (commit_transaction->t_updates) {
+		if (atomic_read(&commit_transaction->t_updates)) {
 			spin_unlock(&commit_transaction->t_handle_lock);
-			spin_unlock(&journal->j_state_lock);
+			write_unlock(&journal->j_state_lock);
 			schedule();
-			spin_lock(&journal->j_state_lock);
+			write_lock(&journal->j_state_lock);
 			spin_lock(&commit_transaction->t_handle_lock);
 		}
 		finish_wait(&journal->j_wait_updates, &wait);
 	}
 	spin_unlock(&commit_transaction->t_handle_lock);
 
-	J_ASSERT (commit_transaction->t_outstanding_credits <=
+	J_ASSERT (atomic_read(&commit_transaction->t_outstanding_credits) <=
 			journal->j_max_transaction_buffers);
 
 	/*
@@ -497,7 +497,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	start_time = ktime_get();
 	commit_transaction->t_log_start = journal->j_head;
 	wake_up(&journal->j_wait_transaction_locked);
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 
 	jbd_debug (3, "JBD: commit phase 2\n");
 
@@ -519,19 +519,20 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 * transaction!  Now comes the tricky part: we need to write out
 	 * metadata.  Loop over the transaction's entire buffer list:
 	 */
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	commit_transaction->t_state = T_COMMIT;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 
 	trace_jbd2_commit_logging(journal, commit_transaction);
 	stats.run.rs_logging = jiffies;
 	stats.run.rs_flushing = jbd2_time_diff(stats.run.rs_flushing,
 					       stats.run.rs_logging);
-	stats.run.rs_blocks = commit_transaction->t_outstanding_credits;
+	stats.run.rs_blocks =
+		atomic_read(&commit_transaction->t_outstanding_credits);
 	stats.run.rs_blocks_logged = 0;
 
 	J_ASSERT(commit_transaction->t_nr_buffers <=
-		 commit_transaction->t_outstanding_credits);
+		 atomic_read(&commit_transaction->t_outstanding_credits));
 
 	err = 0;
 	descriptor = NULL;
@@ -616,7 +617,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		 * the free space in the log, but this counter is changed
 		 * by jbd2_journal_next_log_block() also.
 		 */
-		commit_transaction->t_outstanding_credits--;
+		atomic_dec(&commit_transaction->t_outstanding_credits);
 
 		/* Bump b_count to prevent truncate from stumbling over
                    the shadowed buffer!  @@@ This can go if we ever get
@@ -977,7 +978,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 * __jbd2_journal_drop_transaction(). Otherwise we could race with
 	 * other checkpointing code processing the transaction...
 	 */
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	spin_lock(&journal->j_list_lock);
 	/*
 	 * Now recheck if some buffers did not get attached to the transaction
@@ -985,7 +986,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 */
 	if (commit_transaction->t_forget) {
 		spin_unlock(&journal->j_list_lock);
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		goto restart_loop;
 	}
 
@@ -1003,7 +1004,8 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	 * File the transaction statistics
 	 */
 	stats.ts_tid = commit_transaction->t_tid;
-	stats.run.rs_handle_count = commit_transaction->t_handle_count;
+	stats.run.rs_handle_count =
+		atomic_read(&commit_transaction->t_handle_count);
 	trace_jbd2_run_stats(journal->j_fs_dev->bd_dev,
 			     commit_transaction->t_tid, &stats.run);
 
@@ -1037,7 +1039,7 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 				journal->j_average_commit_time*3) / 4;
 	else
 		journal->j_average_commit_time = commit_time;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 
 	if (commit_transaction->t_checkpoint_list == NULL &&
 	    commit_transaction->t_checkpoint_io_list == NULL) {

fs/jbd2/journal.c

@@ -41,6 +41,7 @@
 #include <linux/hash.h>
 #include <linux/log2.h>
 #include <linux/vmalloc.h>
+#include <linux/backing-dev.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/jbd2.h>
@@ -48,8 +49,6 @@
 #include <asm/uaccess.h>
 #include <asm/page.h>
 
-EXPORT_SYMBOL(jbd2_journal_start);
-EXPORT_SYMBOL(jbd2_journal_restart);
 EXPORT_SYMBOL(jbd2_journal_extend);
 EXPORT_SYMBOL(jbd2_journal_stop);
 EXPORT_SYMBOL(jbd2_journal_lock_updates);
@@ -143,7 +142,7 @@ static int kjournald2(void *arg)
 	/*
 	 * And now, wait forever for commit wakeup events.
 	 */
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 
 loop:
 	if (journal->j_flags & JBD2_UNMOUNT)
@@ -154,10 +153,10 @@ static int kjournald2(void *arg)
 
 	if (journal->j_commit_sequence != journal->j_commit_request) {
 		jbd_debug(1, "OK, requests differ\n");
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		del_timer_sync(&journal->j_commit_timer);
 		jbd2_journal_commit_transaction(journal);
-		spin_lock(&journal->j_state_lock);
+		write_lock(&journal->j_state_lock);
 		goto loop;
 	}
 
@@ -169,9 +168,9 @@ static int kjournald2(void *arg)
 		 * be already stopped.
 		 */
 		jbd_debug(1, "Now suspending kjournald2\n");
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		refrigerator();
-		spin_lock(&journal->j_state_lock);
+		write_lock(&journal->j_state_lock);
 	} else {
 		/*
 		 * We assume on resume that commits are already there,
@@ -191,9 +190,9 @@ static int kjournald2(void *arg)
 		if (journal->j_flags & JBD2_UNMOUNT)
 			should_sleep = 0;
 		if (should_sleep) {
-			spin_unlock(&journal->j_state_lock);
+			write_unlock(&journal->j_state_lock);
 			schedule();
-			spin_lock(&journal->j_state_lock);
+			write_lock(&journal->j_state_lock);
 		}
 		finish_wait(&journal->j_wait_commit, &wait);
 	}
@@ -211,7 +210,7 @@ static int kjournald2(void *arg)
 	goto loop;
 
 end_loop:
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 	del_timer_sync(&journal->j_commit_timer);
 	journal->j_task = NULL;
 	wake_up(&journal->j_wait_done_commit);
@@ -234,16 +233,16 @@ static int jbd2_journal_start_thread(journal_t *journal)
 
 static void journal_kill_thread(journal_t *journal)
 {
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	journal->j_flags |= JBD2_UNMOUNT;
 
 	while (journal->j_task) {
 		wake_up(&journal->j_wait_commit);
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		wait_event(journal->j_wait_done_commit, journal->j_task == NULL);
-		spin_lock(&journal->j_state_lock);
+		write_lock(&journal->j_state_lock);
 	}
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 }
 
 /*
@@ -310,7 +309,17 @@ int jbd2_journal_write_metadata_buffer(transaction_t *transaction,
 	 */
 	J_ASSERT_BH(bh_in, buffer_jbddirty(bh_in));
 
-	new_bh = alloc_buffer_head(GFP_NOFS|__GFP_NOFAIL);
+retry_alloc:
+	new_bh = alloc_buffer_head(GFP_NOFS);
+	if (!new_bh) {
+		/*
+		 * Failure is not an option, but __GFP_NOFAIL is going
+		 * away; so we retry ourselves here.
+		 */
+		congestion_wait(BLK_RW_ASYNC, HZ/50);
+		goto retry_alloc;
+	}
+
 	/* keep subsequent assertions sane */
 	new_bh->b_state = 0;
 	init_buffer(new_bh, NULL, NULL);
@@ -442,7 +451,7 @@ int __jbd2_log_space_left(journal_t *journal)
 {
 	int left = journal->j_free;
 
-	assert_spin_locked(&journal->j_state_lock);
+	/* assert_spin_locked(&journal->j_state_lock); */
 
 	/*
 	 * Be pessimistic here about the number of those free blocks which
@@ -487,9 +496,9 @@ int jbd2_log_start_commit(journal_t *journal, tid_t tid)
 {
 	int ret;
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	ret = __jbd2_log_start_commit(journal, tid);
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 	return ret;
 }
 
@@ -508,7 +517,7 @@ int jbd2_journal_force_commit_nested(journal_t *journal)
 	transaction_t *transaction = NULL;
 	tid_t tid;
 
-	spin_lock(&journal->j_state_lock);
+	read_lock(&journal->j_state_lock);
 	if (journal->j_running_transaction && !current->journal_info) {
 		transaction = journal->j_running_transaction;
 		__jbd2_log_start_commit(journal, transaction->t_tid);
@@ -516,12 +525,12 @@ int jbd2_journal_force_commit_nested(journal_t *journal)
 		transaction = journal->j_committing_transaction;
 
 	if (!transaction) {
-		spin_unlock(&journal->j_state_lock);
+		read_unlock(&journal->j_state_lock);
 		return 0;	/* Nothing to retry */
 	}
 
 	tid = transaction->t_tid;
-	spin_unlock(&journal->j_state_lock);
+	read_unlock(&journal->j_state_lock);
 	jbd2_log_wait_commit(journal, tid);
 	return 1;
 }
@@ -535,7 +544,7 @@ int jbd2_journal_start_commit(journal_t *journal, tid_t *ptid)
 {
 	int ret = 0;
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	if (journal->j_running_transaction) {
 		tid_t tid = journal->j_running_transaction->t_tid;
 
@@ -554,7 +563,7 @@ int jbd2_journal_start_commit(journal_t *journal, tid_t *ptid)
 			*ptid = journal->j_committing_transaction->t_tid;
 		ret = 1;
 	}
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 	return ret;
 }
 
@@ -566,26 +575,24 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid)
 {
 	int err = 0;
 
+	read_lock(&journal->j_state_lock);
 #ifdef CONFIG_JBD2_DEBUG
-	spin_lock(&journal->j_state_lock);
 	if (!tid_geq(journal->j_commit_request, tid)) {
 		printk(KERN_EMERG
 		       "%s: error: j_commit_request=%d, tid=%d\n",
 		       __func__, journal->j_commit_request, tid);
 	}
-	spin_unlock(&journal->j_state_lock);
 #endif
-	spin_lock(&journal->j_state_lock);
 	while (tid_gt(tid, journal->j_commit_sequence)) {
 		jbd_debug(1, "JBD: want %d, j_commit_sequence=%d\n",
 				  tid, journal->j_commit_sequence);
 		wake_up(&journal->j_wait_commit);
-		spin_unlock(&journal->j_state_lock);
+		read_unlock(&journal->j_state_lock);
 		wait_event(journal->j_wait_done_commit,
 				!tid_gt(tid, journal->j_commit_sequence));
-		spin_lock(&journal->j_state_lock);
+		read_lock(&journal->j_state_lock);
 	}
-	spin_unlock(&journal->j_state_lock);
+	read_unlock(&journal->j_state_lock);
 
 	if (unlikely(is_journal_aborted(journal))) {
 		printk(KERN_EMERG "journal commit I/O error\n");
@@ -602,7 +609,7 @@ int jbd2_journal_next_log_block(journal_t *journal, unsigned long long *retp)
 {
 	unsigned long blocknr;
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	J_ASSERT(journal->j_free > 1);
 
 	blocknr = journal->j_head;
@@ -610,7 +617,7 @@ int jbd2_journal_next_log_block(journal_t *journal, unsigned long long *retp)
 	journal->j_free--;
 	if (journal->j_head == journal->j_last)
 		journal->j_head = journal->j_first;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 	return jbd2_journal_bmap(journal, blocknr, retp);
 }
 
@@ -830,7 +837,7 @@ static journal_t * journal_init_common (void)
 	mutex_init(&journal->j_checkpoint_mutex);
 	spin_lock_init(&journal->j_revoke_lock);
 	spin_lock_init(&journal->j_list_lock);
-	spin_lock_init(&journal->j_state_lock);
+	rwlock_init(&journal->j_state_lock);
 
 	journal->j_commit_interval = (HZ * JBD2_DEFAULT_MAX_COMMIT_AGE);
 	journal->j_min_batch_time = 0;
@@ -1096,14 +1103,14 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait)
 		set_buffer_uptodate(bh);
 	}
 
-	spin_lock(&journal->j_state_lock);
+	read_lock(&journal->j_state_lock);
 	jbd_debug(1,"JBD: updating superblock (start %ld, seq %d, errno %d)\n",
 		  journal->j_tail, journal->j_tail_sequence, journal->j_errno);
 
 	sb->s_sequence = cpu_to_be32(journal->j_tail_sequence);
 	sb->s_start    = cpu_to_be32(journal->j_tail);
 	sb->s_errno    = cpu_to_be32(journal->j_errno);
-	spin_unlock(&journal->j_state_lock);
+	read_unlock(&journal->j_state_lock);
 
 	BUFFER_TRACE(bh, "marking dirty");
 	mark_buffer_dirty(bh);
@@ -1124,12 +1131,12 @@ void jbd2_journal_update_superblock(journal_t *journal, int wait)
 	 * any future commit will have to be careful to update the
 	 * superblock again to re-record the true start of the log. */
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	if (sb->s_start)
 		journal->j_flags &= ~JBD2_FLUSHED;
 	else
 		journal->j_flags |= JBD2_FLUSHED;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 }
 
 /*
@@ -1391,13 +1398,9 @@ int jbd2_journal_check_used_features (journal_t *journal, unsigned long compat,
 int jbd2_journal_check_available_features (journal_t *journal, unsigned long compat,
 				      unsigned long ro, unsigned long incompat)
 {
-	journal_superblock_t *sb;
-
 	if (!compat && !ro && !incompat)
 		return 1;
 
-	sb = journal->j_superblock;
-
 	/* We can support any known requested features iff the
 	 * superblock is in version 2.  Otherwise we fail to support any
 	 * extended sb features. */
@@ -1545,7 +1548,7 @@ int jbd2_journal_flush(journal_t *journal)
 	transaction_t *transaction = NULL;
 	unsigned long old_tail;
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 
 	/* Force everything buffered to the log... */
 	if (journal->j_running_transaction) {
@@ -1558,10 +1561,10 @@ int jbd2_journal_flush(journal_t *journal)
 	if (transaction) {
 		tid_t tid = transaction->t_tid;
 
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		jbd2_log_wait_commit(journal, tid);
 	} else {
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 	}
 
 	/* ...and flush everything in the log out to disk. */
@@ -1585,12 +1588,12 @@ int jbd2_journal_flush(journal_t *journal)
 	 * the magic code for a fully-recovered superblock.  Any future
 	 * commits of data to the journal will restore the current
 	 * s_start value. */
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	old_tail = journal->j_tail;
 	journal->j_tail = 0;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 	jbd2_journal_update_superblock(journal, 1);
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	journal->j_tail = old_tail;
 
 	J_ASSERT(!journal->j_running_transaction);
@@ -1598,7 +1601,7 @@ int jbd2_journal_flush(journal_t *journal)
 	J_ASSERT(!journal->j_checkpoint_transactions);
 	J_ASSERT(journal->j_head == journal->j_tail);
 	J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence);
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 	return 0;
 }
 
@@ -1617,7 +1620,6 @@ int jbd2_journal_flush(journal_t *journal)
 
 int jbd2_journal_wipe(journal_t *journal, int write)
 {
-	journal_superblock_t *sb;
 	int err = 0;
 
 	J_ASSERT (!(journal->j_flags & JBD2_LOADED));
@@ -1626,8 +1628,6 @@ int jbd2_journal_wipe(journal_t *journal, int write)
 	if (err)
 		return err;
 
-	sb = journal->j_superblock;
-
 	if (!journal->j_tail)
 		goto no_recovery;
 
@@ -1665,12 +1665,12 @@ void __jbd2_journal_abort_hard(journal_t *journal)
 	printk(KERN_ERR "Aborting journal on device %s.\n",
 	       journal->j_devname);
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	journal->j_flags |= JBD2_ABORT;
 	transaction = journal->j_running_transaction;
 	if (transaction)
 		__jbd2_log_start_commit(journal, transaction->t_tid);
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 }
 
 /* Soft abort: record the abort error status in the journal superblock,
@@ -1755,12 +1755,12 @@ int jbd2_journal_errno(journal_t *journal)
 {
 	int err;
 
-	spin_lock(&journal->j_state_lock);
+	read_lock(&journal->j_state_lock);
 	if (journal->j_flags & JBD2_ABORT)
 		err = -EROFS;
 	else
 		err = journal->j_errno;
-	spin_unlock(&journal->j_state_lock);
+	read_unlock(&journal->j_state_lock);
 	return err;
 }
 
@@ -1775,12 +1775,12 @@ int jbd2_journal_clear_err(journal_t *journal)
 {
 	int err = 0;
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	if (journal->j_flags & JBD2_ABORT)
 		err = -EROFS;
 	else
 		journal->j_errno = 0;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 	return err;
 }
 
@@ -1793,10 +1793,10 @@ int jbd2_journal_clear_err(journal_t *journal)
  */
 void jbd2_journal_ack_err(journal_t *journal)
 {
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	if (journal->j_errno)
 		journal->j_flags |= JBD2_ACK_ERR;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 }
 
 int jbd2_journal_blocks_per_page(struct inode *inode)
@@ -2201,8 +2201,6 @@ void jbd2_journal_init_jbd_inode(struct jbd2_inode *jinode, struct inode *inode)
 void jbd2_journal_release_jbd_inode(journal_t *journal,
 				    struct jbd2_inode *jinode)
 {
-	int writeout = 0;
-
 	if (!journal)
 		return;
 restart:
@@ -2219,9 +2217,6 @@ void jbd2_journal_release_jbd_inode(journal_t *journal,
 		goto restart;
 	}
 
-	/* Do we need to wait for data writeback? */
-	if (journal->j_committing_transaction == jinode->i_transaction)
-		writeout = 1;
 	if (jinode->i_transaction) {
 		list_del(&jinode->i_list);
 		jinode->i_transaction = NULL;

fs/jbd2/recovery.c

@@ -285,12 +285,10 @@ int jbd2_journal_recover(journal_t *journal)
 int jbd2_journal_skip_recovery(journal_t *journal)
 {
 	int			err;
-	journal_superblock_t *	sb;
 
 	struct recovery_info	info;
 
 	memset (&info, 0, sizeof(info));
-	sb = journal->j_superblock;
 
 	err = do_one_pass(journal, &info, PASS_SCAN);
 
@@ -299,7 +297,8 @@ int jbd2_journal_skip_recovery(journal_t *journal)
 		++journal->j_transaction_sequence;
 	} else {
 #ifdef CONFIG_JBD2_DEBUG
-		int dropped = info.end_transaction - be32_to_cpu(sb->s_sequence);
+		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",
@@ -365,11 +364,6 @@ static int do_one_pass(journal_t *journal,
 	int			tag_bytes = journal_tag_bytes(journal);
 	__u32			crc32_sum = ~0; /* Transactional Checksums */
 
-	/* Precompute the maximum metadata descriptors in a descriptor block */
-	int			MAX_BLOCKS_PER_DESC;
-	MAX_BLOCKS_PER_DESC = ((journal->j_blocksize-sizeof(journal_header_t))
-			       / tag_bytes);
-
 	/*
 	 * First thing is to establish what we expect to find in the log
 	 * (in terms of transaction IDs), and where (in terms of log

fs/jbd2/transaction.c

@@ -26,6 +26,8 @@
 #include <linux/mm.h>
 #include <linux/highmem.h>
 #include <linux/hrtimer.h>
+#include <linux/backing-dev.h>
+#include <linux/module.h>
 
 static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh);
 
@@ -53,6 +55,9 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction)
 	transaction->t_tid = journal->j_transaction_sequence++;
 	transaction->t_expires = jiffies + journal->j_commit_interval;
 	spin_lock_init(&transaction->t_handle_lock);
+	atomic_set(&transaction->t_updates, 0);
+	atomic_set(&transaction->t_outstanding_credits, 0);
+	atomic_set(&transaction->t_handle_count, 0);
 	INIT_LIST_HEAD(&transaction->t_inode_list);
 	INIT_LIST_HEAD(&transaction->t_private_list);
 
@@ -83,65 +88,75 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction)
  * transaction's buffer credits.
  */
 
-static int start_this_handle(journal_t *journal, handle_t *handle)
+static int start_this_handle(journal_t *journal, handle_t *handle,
+			     int gfp_mask)
 {
 	transaction_t *transaction;
 	int needed;
 	int nblocks = handle->h_buffer_credits;
 	transaction_t *new_transaction = NULL;
-	int ret = 0;
 	unsigned long ts = jiffies;
 
 	if (nblocks > journal->j_max_transaction_buffers) {
 		printk(KERN_ERR "JBD: %s wants too many credits (%d > %d)\n",
 		       current->comm, nblocks,
 		       journal->j_max_transaction_buffers);
-		ret = -ENOSPC;
-		goto out;
+		return -ENOSPC;
 	}
 
 alloc_transaction:
 	if (!journal->j_running_transaction) {
-		new_transaction = kzalloc(sizeof(*new_transaction),
-						GFP_NOFS|__GFP_NOFAIL);
+		new_transaction = kzalloc(sizeof(*new_transaction), gfp_mask);
 		if (!new_transaction) {
-			ret = -ENOMEM;
-			goto out;
+			/*
+			 * If __GFP_FS is not present, then we may be
+			 * being called from inside the fs writeback
+			 * layer, so we MUST NOT fail.  Since
+			 * __GFP_NOFAIL is going away, we will arrange
+			 * to retry the allocation ourselves.
+			 */
+			if ((gfp_mask & __GFP_FS) == 0) {
+				congestion_wait(BLK_RW_ASYNC, HZ/50);
+				goto alloc_transaction;
+			}
+			return -ENOMEM;
 		}
 	}
 
 	jbd_debug(3, "New handle %p going live.\n", handle);
 
-repeat:
-
 	/*
 	 * We need to hold j_state_lock until t_updates has been incremented,
 	 * for proper journal barrier handling
 	 */
-	spin_lock(&journal->j_state_lock);
-repeat_locked:
+repeat:
+	read_lock(&journal->j_state_lock);
 	if (is_journal_aborted(journal) ||
 	    (journal->j_errno != 0 && !(journal->j_flags & JBD2_ACK_ERR))) {
-		spin_unlock(&journal->j_state_lock);
-		ret = -EROFS;
-		goto out;
+		read_unlock(&journal->j_state_lock);
+		kfree(new_transaction);
+		return -EROFS;
 	}
 
 	/* Wait on the journal's transaction barrier if necessary */
 	if (journal->j_barrier_count) {
-		spin_unlock(&journal->j_state_lock);
+		read_unlock(&journal->j_state_lock);
 		wait_event(journal->j_wait_transaction_locked,
 				journal->j_barrier_count == 0);
 		goto repeat;
 	}
 
 	if (!journal->j_running_transaction) {
-		if (!new_transaction) {
-			spin_unlock(&journal->j_state_lock);
+		read_unlock(&journal->j_state_lock);
+		if (!new_transaction)
 			goto alloc_transaction;
+		write_lock(&journal->j_state_lock);
+		if (!journal->j_running_transaction) {
+			jbd2_get_transaction(journal, new_transaction);
+			new_transaction = NULL;
 		}
-		jbd2_get_transaction(journal, new_transaction);
-		new_transaction = NULL;
+		write_unlock(&journal->j_state_lock);
+		goto repeat;
 	}
 
 	transaction = journal->j_running_transaction;
@@ -155,7 +170,7 @@ static int start_this_handle(journal_t *journal, handle_t *handle)
 
 		prepare_to_wait(&journal->j_wait_transaction_locked,
 					&wait, TASK_UNINTERRUPTIBLE);
-		spin_unlock(&journal->j_state_lock);
+		read_unlock(&journal->j_state_lock);
 		schedule();
 		finish_wait(&journal->j_wait_transaction_locked, &wait);
 		goto repeat;
@@ -166,8 +181,8 @@ static int start_this_handle(journal_t *journal, handle_t *handle)
 	 * buffers requested by this operation, we need to stall pending a log
 	 * checkpoint to free some more log space.
 	 */
-	spin_lock(&transaction->t_handle_lock);
-	needed = transaction->t_outstanding_credits + nblocks;
+	needed = atomic_add_return(nblocks,
+				   &transaction->t_outstanding_credits);
 
 	if (needed > journal->j_max_transaction_buffers) {
 		/*
@@ -178,11 +193,11 @@ static int start_this_handle(journal_t *journal, handle_t *handle)
 		DEFINE_WAIT(wait);
 
 		jbd_debug(2, "Handle %p starting new commit...\n", handle);
-		spin_unlock(&transaction->t_handle_lock);
+		atomic_sub(nblocks, &transaction->t_outstanding_credits);
 		prepare_to_wait(&journal->j_wait_transaction_locked, &wait,
 				TASK_UNINTERRUPTIBLE);
 		__jbd2_log_start_commit(journal, transaction->t_tid);
-		spin_unlock(&journal->j_state_lock);
+		read_unlock(&journal->j_state_lock);
 		schedule();
 		finish_wait(&journal->j_wait_transaction_locked, &wait);
 		goto repeat;
@@ -215,35 +230,48 @@ static int start_this_handle(journal_t *journal, handle_t *handle)
 	 */
 	if (__jbd2_log_space_left(journal) < jbd_space_needed(journal)) {
 		jbd_debug(2, "Handle %p waiting for checkpoint...\n", handle);
-		spin_unlock(&transaction->t_handle_lock);
-		__jbd2_log_wait_for_space(journal);
-		goto repeat_locked;
+		atomic_sub(nblocks, &transaction->t_outstanding_credits);
+		read_unlock(&journal->j_state_lock);
+		write_lock(&journal->j_state_lock);
+		if (__jbd2_log_space_left(journal) < jbd_space_needed(journal))
+			__jbd2_log_wait_for_space(journal);
+		write_unlock(&journal->j_state_lock);
+		goto repeat;
 	}
 
 	/* OK, account for the buffers that this operation expects to
-	 * use and add the handle to the running transaction. */
-
-	if (time_after(transaction->t_start, ts)) {
+	 * use and add the handle to the running transaction. 
+	 *
+	 * In order for t_max_wait to be reliable, it must be
+	 * protected by a lock.  But doing so will mean that
+	 * start_this_handle() can not be run in parallel on SMP
+	 * systems, which limits our scalability.  So we only enable
+	 * it when debugging is enabled.  We may want to use a
+	 * separate flag, eventually, so we can enable this
+	 * independently of debugging.
+	 */
+#ifdef CONFIG_JBD2_DEBUG
+	if (jbd2_journal_enable_debug &&
+	    time_after(transaction->t_start, ts)) {
 		ts = jbd2_time_diff(ts, transaction->t_start);
+		spin_lock(&transaction->t_handle_lock);
 		if (ts > transaction->t_max_wait)
 			transaction->t_max_wait = ts;
+		spin_unlock(&transaction->t_handle_lock);
 	}
-
+#endif
 	handle->h_transaction = transaction;
-	transaction->t_outstanding_credits += nblocks;
-	transaction->t_updates++;
-	transaction->t_handle_count++;
+	atomic_inc(&transaction->t_updates);
+	atomic_inc(&transaction->t_handle_count);
 	jbd_debug(4, "Handle %p given %d credits (total %d, free %d)\n",
-		  handle, nblocks, transaction->t_outstanding_credits,
+		  handle, nblocks,
+		  atomic_read(&transaction->t_outstanding_credits),
 		  __jbd2_log_space_left(journal));
-	spin_unlock(&transaction->t_handle_lock);
-	spin_unlock(&journal->j_state_lock);
+	read_unlock(&journal->j_state_lock);
 
 	lock_map_acquire(&handle->h_lockdep_map);
-out:
-	if (unlikely(new_transaction))		/* It's usually NULL */
-		kfree(new_transaction);
-	return ret;
+	kfree(new_transaction);
+	return 0;
 }
 
 static struct lock_class_key jbd2_handle_key;
@@ -278,7 +306,7 @@ static handle_t *new_handle(int nblocks)
  *
  * Return a pointer to a newly allocated handle, or NULL on failure
  */
-handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
+handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int gfp_mask)
 {
 	handle_t *handle = journal_current_handle();
 	int err;
@@ -298,7 +326,7 @@ handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
 
 	current->journal_info = handle;
 
-	err = start_this_handle(journal, handle);
+	err = start_this_handle(journal, handle, gfp_mask);
 	if (err < 0) {
 		jbd2_free_handle(handle);
 		current->journal_info = NULL;
@@ -308,6 +336,15 @@ handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
 out:
 	return handle;
 }
+EXPORT_SYMBOL(jbd2__journal_start);
+
+
+handle_t *jbd2_journal_start(journal_t *journal, int nblocks)
+{
+	return jbd2__journal_start(journal, nblocks, GFP_NOFS);
+}
+EXPORT_SYMBOL(jbd2_journal_start);
+
 
 /**
  * int jbd2_journal_extend() - extend buffer credits.
@@ -342,7 +379,7 @@ int jbd2_journal_extend(handle_t *handle, int nblocks)
 
 	result = 1;
 
-	spin_lock(&journal->j_state_lock);
+	read_lock(&journal->j_state_lock);
 
 	/* Don't extend a locked-down transaction! */
 	if (handle->h_transaction->t_state != T_RUNNING) {
@@ -352,7 +389,7 @@ int jbd2_journal_extend(handle_t *handle, int nblocks)
 	}
 
 	spin_lock(&transaction->t_handle_lock);
-	wanted = transaction->t_outstanding_credits + nblocks;
+	wanted = atomic_read(&transaction->t_outstanding_credits) + nblocks;
 
 	if (wanted > journal->j_max_transaction_buffers) {
 		jbd_debug(3, "denied handle %p %d blocks: "
@@ -367,14 +404,14 @@ int jbd2_journal_extend(handle_t *handle, int nblocks)
 	}
 
 	handle->h_buffer_credits += nblocks;
-	transaction->t_outstanding_credits += nblocks;
+	atomic_add(nblocks, &transaction->t_outstanding_credits);
 	result = 0;
 
 	jbd_debug(3, "extended handle %p by %d\n", handle, nblocks);
 unlock:
 	spin_unlock(&transaction->t_handle_lock);
 error_out:
-	spin_unlock(&journal->j_state_lock);
+	read_unlock(&journal->j_state_lock);
 out:
 	return result;
 }
@@ -394,8 +431,7 @@ int jbd2_journal_extend(handle_t *handle, int nblocks)
  * transaction capabable of guaranteeing the requested number of
  * credits.
  */
-
-int jbd2_journal_restart(handle_t *handle, int nblocks)
+int jbd2__journal_restart(handle_t *handle, int nblocks, int gfp_mask)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal = transaction->t_journal;
@@ -410,29 +446,35 @@ int jbd2_journal_restart(handle_t *handle, int nblocks)
 	 * First unlink the handle from its current transaction, and start the
 	 * commit on that.
 	 */
-	J_ASSERT(transaction->t_updates > 0);
+	J_ASSERT(atomic_read(&transaction->t_updates) > 0);
 	J_ASSERT(journal_current_handle() == handle);
 
-	spin_lock(&journal->j_state_lock);
+	read_lock(&journal->j_state_lock);
 	spin_lock(&transaction->t_handle_lock);
-	transaction->t_outstanding_credits -= handle->h_buffer_credits;
-	transaction->t_updates--;
-
-	if (!transaction->t_updates)
+	atomic_sub(handle->h_buffer_credits,
+		   &transaction->t_outstanding_credits);
+	if (atomic_dec_and_test(&transaction->t_updates))
 		wake_up(&journal->j_wait_updates);
 	spin_unlock(&transaction->t_handle_lock);
 
 	jbd_debug(2, "restarting handle %p\n", handle);
 	__jbd2_log_start_commit(journal, transaction->t_tid);
-	spin_unlock(&journal->j_state_lock);
+	read_unlock(&journal->j_state_lock);
 
 	lock_map_release(&handle->h_lockdep_map);
 	handle->h_buffer_credits = nblocks;
-	ret = start_this_handle(journal, handle);
+	ret = start_this_handle(journal, handle, gfp_mask);
 	return ret;
 }
+EXPORT_SYMBOL(jbd2__journal_restart);
 
 
+int jbd2_journal_restart(handle_t *handle, int nblocks)
+{
+	return jbd2__journal_restart(handle, nblocks, GFP_NOFS);
+}
+EXPORT_SYMBOL(jbd2_journal_restart);
+
 /**
  * void jbd2_journal_lock_updates () - establish a transaction barrier.
  * @journal:  Journal to establish a barrier on.
@@ -447,7 +489,7 @@ void jbd2_journal_lock_updates(journal_t *journal)
 {
 	DEFINE_WAIT(wait);
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	++journal->j_barrier_count;
 
 	/* Wait until there are no running updates */
@@ -458,19 +500,19 @@ void jbd2_journal_lock_updates(journal_t *journal)
 			break;
 
 		spin_lock(&transaction->t_handle_lock);
-		if (!transaction->t_updates) {
+		if (!atomic_read(&transaction->t_updates)) {
 			spin_unlock(&transaction->t_handle_lock);
 			break;
 		}
 		prepare_to_wait(&journal->j_wait_updates, &wait,
 				TASK_UNINTERRUPTIBLE);
 		spin_unlock(&transaction->t_handle_lock);
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		schedule();
 		finish_wait(&journal->j_wait_updates, &wait);
-		spin_lock(&journal->j_state_lock);
+		write_lock(&journal->j_state_lock);
 	}
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 
 	/*
 	 * We have now established a barrier against other normal updates, but
@@ -494,9 +536,9 @@ void jbd2_journal_unlock_updates (journal_t *journal)
 	J_ASSERT(journal->j_barrier_count != 0);
 
 	mutex_unlock(&journal->j_barrier);
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	--journal->j_barrier_count;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 	wake_up(&journal->j_wait_transaction_locked);
 }
 
@@ -1238,7 +1280,8 @@ int jbd2_journal_stop(handle_t *handle)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal = transaction->t_journal;
-	int err;
+	int err, wait_for_commit = 0;
+	tid_t tid;
 	pid_t pid;
 
 	J_ASSERT(journal_current_handle() == handle);
@@ -1246,7 +1289,7 @@ int jbd2_journal_stop(handle_t *handle)
 	if (is_handle_aborted(handle))
 		err = -EIO;
 	else {
-		J_ASSERT(transaction->t_updates > 0);
+		J_ASSERT(atomic_read(&transaction->t_updates) > 0);
 		err = 0;
 	}
 
@@ -1291,9 +1334,9 @@ int jbd2_journal_stop(handle_t *handle)
 
 		journal->j_last_sync_writer = pid;
 
-		spin_lock(&journal->j_state_lock);
+		read_lock(&journal->j_state_lock);
 		commit_time = journal->j_average_commit_time;
-		spin_unlock(&journal->j_state_lock);
+		read_unlock(&journal->j_state_lock);
 
 		trans_time = ktime_to_ns(ktime_sub(ktime_get(),
 						   transaction->t_start_time));
@@ -1314,14 +1357,8 @@ int jbd2_journal_stop(handle_t *handle)
 	if (handle->h_sync)
 		transaction->t_synchronous_commit = 1;
 	current->journal_info = NULL;
-	spin_lock(&transaction->t_handle_lock);
-	transaction->t_outstanding_credits -= handle->h_buffer_credits;
-	transaction->t_updates--;
-	if (!transaction->t_updates) {
-		wake_up(&journal->j_wait_updates);
-		if (journal->j_barrier_count)
-			wake_up(&journal->j_wait_transaction_locked);
-	}
+	atomic_sub(handle->h_buffer_credits,
+		   &transaction->t_outstanding_credits);
 
 	/*
 	 * If the handle is marked SYNC, we need to set another commit
@@ -1330,15 +1367,13 @@ int jbd2_journal_stop(handle_t *handle)
 	 * transaction is too old now.
 	 */
 	if (handle->h_sync ||
-			transaction->t_outstanding_credits >
-				journal->j_max_transaction_buffers ||
-			time_after_eq(jiffies, transaction->t_expires)) {
+	    (atomic_read(&transaction->t_outstanding_credits) >
+	     journal->j_max_transaction_buffers) ||
+	    time_after_eq(jiffies, transaction->t_expires)) {
 		/* Do this even for aborted journals: an abort still
 		 * completes the commit thread, it just doesn't write
 		 * anything to disk. */
-		tid_t tid = transaction->t_tid;
 
-		spin_unlock(&transaction->t_handle_lock);
 		jbd_debug(2, "transaction too old, requesting commit for "
 					"handle %p\n", handle);
 		/* This is non-blocking */
@@ -1349,11 +1384,25 @@ int jbd2_journal_stop(handle_t *handle)
 		 * to wait for the commit to complete.
 		 */
 		if (handle->h_sync && !(current->flags & PF_MEMALLOC))
-			err = jbd2_log_wait_commit(journal, tid);
-	} else {
-		spin_unlock(&transaction->t_handle_lock);
+			wait_for_commit = 1;
 	}
 
+	/*
+	 * Once we drop t_updates, if it goes to zero the transaction
+	 * could start commiting on us and eventually disappear.  So
+	 * once we do this, we must not dereference transaction
+	 * pointer again.
+	 */
+	tid = transaction->t_tid;
+	if (atomic_dec_and_test(&transaction->t_updates)) {
+		wake_up(&journal->j_wait_updates);
+		if (journal->j_barrier_count)
+			wake_up(&journal->j_wait_transaction_locked);
+	}
+
+	if (wait_for_commit)
+		err = jbd2_log_wait_commit(journal, tid);
+
 	lock_map_release(&handle->h_lockdep_map);
 
 	jbd2_free_handle(handle);
@@ -1719,7 +1768,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 		goto zap_buffer_unlocked;
 
 	/* OK, we have data buffer in journaled mode */
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	jbd_lock_bh_state(bh);
 	spin_lock(&journal->j_list_lock);
 
@@ -1772,7 +1821,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 			jbd2_journal_put_journal_head(jh);
 			spin_unlock(&journal->j_list_lock);
 			jbd_unlock_bh_state(bh);
-			spin_unlock(&journal->j_state_lock);
+			write_unlock(&journal->j_state_lock);
 			return ret;
 		} else {
 			/* There is no currently-running transaction. So the
@@ -1786,7 +1835,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 				jbd2_journal_put_journal_head(jh);
 				spin_unlock(&journal->j_list_lock);
 				jbd_unlock_bh_state(bh);
-				spin_unlock(&journal->j_state_lock);
+				write_unlock(&journal->j_state_lock);
 				return ret;
 			} else {
 				/* The orphan record's transaction has
@@ -1810,7 +1859,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 		jbd2_journal_put_journal_head(jh);
 		spin_unlock(&journal->j_list_lock);
 		jbd_unlock_bh_state(bh);
-		spin_unlock(&journal->j_state_lock);
+		write_unlock(&journal->j_state_lock);
 		return 0;
 	} else {
 		/* Good, the buffer belongs to the running transaction.
@@ -1829,7 +1878,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh)
 zap_buffer_no_jh:
 	spin_unlock(&journal->j_list_lock);
 	jbd_unlock_bh_state(bh);
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 zap_buffer_unlocked:
 	clear_buffer_dirty(bh);
 	J_ASSERT_BH(bh, !buffer_jbddirty(bh));
@@ -2136,9 +2185,9 @@ int jbd2_journal_begin_ordered_truncate(journal_t *journal,
 	/* Locks are here just to force reading of recent values, it is
 	 * enough that the transaction was not committing before we started
 	 * a transaction adding the inode to orphan list */
-	spin_lock(&journal->j_state_lock);
+	read_lock(&journal->j_state_lock);
 	commit_trans = journal->j_committing_transaction;
-	spin_unlock(&journal->j_state_lock);
+	read_unlock(&journal->j_state_lock);
 	spin_lock(&journal->j_list_lock);
 	inode_trans = jinode->i_transaction;
 	spin_unlock(&journal->j_list_lock);

fs/ocfs2/journal.c

@@ -760,13 +760,13 @@ void ocfs2_set_journal_params(struct ocfs2_super *osb)
 	if (osb->osb_commit_interval)
 		commit_interval = osb->osb_commit_interval;
 
-	spin_lock(&journal->j_state_lock);
+	write_lock(&journal->j_state_lock);
 	journal->j_commit_interval = commit_interval;
 	if (osb->s_mount_opt & OCFS2_MOUNT_BARRIER)
 		journal->j_flags |= JBD2_BARRIER;
 	else
 		journal->j_flags &= ~JBD2_BARRIER;
-	spin_unlock(&journal->j_state_lock);
+	write_unlock(&journal->j_state_lock);
 }
 
 int ocfs2_journal_init(struct ocfs2_journal *journal, int *dirty)

include/linux/jbd2.h

@@ -601,13 +601,13 @@ struct transaction_s
 	 * Number of outstanding updates running on this transaction
 	 * [t_handle_lock]
 	 */
-	int			t_updates;
+	atomic_t		t_updates;
 
 	/*
 	 * Number of buffers reserved for use by all handles in this transaction
 	 * handle but not yet modified. [t_handle_lock]
 	 */
-	int			t_outstanding_credits;
+	atomic_t		t_outstanding_credits;
 
 	/*
 	 * Forward and backward links for the circular list of all transactions
@@ -629,7 +629,7 @@ struct transaction_s
 	/*
 	 * How many handles used this transaction? [t_handle_lock]
 	 */
-	int t_handle_count;
+	atomic_t		t_handle_count;
 
 	/*
 	 * This transaction is being forced and some process is
@@ -764,7 +764,7 @@ struct journal_s
 	/*
 	 * Protect the various scalars in the journal
 	 */
-	spinlock_t		j_state_lock;
+	rwlock_t		j_state_lock;
 
 	/*
 	 * Number of processes waiting to create a barrier lock [j_state_lock]
@@ -1082,7 +1082,9 @@ static inline handle_t *journal_current_handle(void)
  */
 
 extern handle_t *jbd2_journal_start(journal_t *, int nblocks);
-extern int	 jbd2_journal_restart (handle_t *, int nblocks);
+extern handle_t *jbd2__journal_start(journal_t *, int nblocks, int gfp_mask);
+extern int	 jbd2_journal_restart(handle_t *, int nblocks);
+extern int	 jbd2__journal_restart(handle_t *, int nblocks, int gfp_mask);
 extern int	 jbd2_journal_extend (handle_t *, int nblocks);
 extern int	 jbd2_journal_get_write_access(handle_t *, struct buffer_head *);
 extern int	 jbd2_journal_get_create_access (handle_t *, struct buffer_head *);
@@ -1257,8 +1259,8 @@ static inline int jbd_space_needed(journal_t *journal)
 {
 	int nblocks = journal->j_max_transaction_buffers;
 	if (journal->j_committing_transaction)
-		nblocks += journal->j_committing_transaction->
-					t_outstanding_credits;
+		nblocks += atomic_read(&journal->j_committing_transaction->
+				       t_outstanding_credits);
 	return nblocks;
 }
 

include/trace/events/ext4.h

@@ -395,11 +395,12 @@ DEFINE_EVENT(ext4__mb_new_pa, ext4_mb_new_group_pa,
 );
 
 TRACE_EVENT(ext4_mb_release_inode_pa,
-	TP_PROTO(struct ext4_allocation_context *ac,
+	TP_PROTO(struct super_block *sb,
+		 struct ext4_allocation_context *ac,
 		 struct ext4_prealloc_space *pa,
 		 unsigned long long block, unsigned int count),
 
-	TP_ARGS(ac, pa, block, count),
+	TP_ARGS(sb, ac, pa, block, count),
 
 	TP_STRUCT__entry(
 		__field(	dev_t,	dev			)
@@ -410,8 +411,9 @@ TRACE_EVENT(ext4_mb_release_inode_pa,
 	),
 
 	TP_fast_assign(
-		__entry->dev		= ac->ac_sb->s_dev;
-		__entry->ino		= ac->ac_inode->i_ino;
+		__entry->dev		= sb->s_dev;
+		__entry->ino		= (ac && ac->ac_inode) ? 
+						ac->ac_inode->i_ino : 0;
 		__entry->block		= block;
 		__entry->count		= count;
 	),
@@ -422,10 +424,11 @@ TRACE_EVENT(ext4_mb_release_inode_pa,
 );
 
 TRACE_EVENT(ext4_mb_release_group_pa,
-	TP_PROTO(struct ext4_allocation_context *ac,
+	TP_PROTO(struct super_block *sb,
+		 struct ext4_allocation_context *ac,
 		 struct ext4_prealloc_space *pa),
 
-	TP_ARGS(ac, pa),
+	TP_ARGS(sb, ac, pa),
 
 	TP_STRUCT__entry(
 		__field(	dev_t,	dev			)
@@ -436,8 +439,9 @@ TRACE_EVENT(ext4_mb_release_group_pa,
 	),
 
 	TP_fast_assign(
-		__entry->dev		= ac->ac_sb->s_dev;
-		__entry->ino		= ac->ac_inode->i_ino;
+		__entry->dev		= sb->s_dev;
+		__entry->ino		= (ac && ac->ac_inode) ?
+						ac->ac_inode->i_ino : 0;
 		__entry->pa_pstart	= pa->pa_pstart;
 		__entry->pa_len		= pa->pa_len;
 	),