Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 updates from Ted Ts'o:
 "Improve performance for ext4 by allowing multiple process to perform
  direct I/O writes to preallocated blocks by using a shared inode lock
  instead of taking an exclusive lock.

  In addition, multiple bug fixes and cleanups"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: fix incorrect options show of original mount_opt and extend mount_opt2
  ext4: Fix possible corruption when moving a directory
  ext4: init error handle resource before init group descriptors
  ext4: fix task hung in ext4_xattr_delete_inode
  jbd2: fix data missing when reusing bh which is ready to be checkpointed
  ext4: update s_journal_inum if it changes after journal replay
  ext4: fail ext4_iget if special inode unallocated
  ext4: fix function prototype mismatch for ext4_feat_ktype
  ext4: remove unnecessary variable initialization
  ext4: fix inode tree inconsistency caused by ENOMEM
  ext4: refuse to create ea block when umounted
  ext4: optimize ea_inode block expansion
  ext4: remove dead code in updating backup sb
  ext4: dio take shared inode lock when overwriting preallocated blocks
  ext4: don't show commit interval if it is zero
  ext4: use ext4_fc_tl_mem in fast-commit replay path
  ext4: improve xattr consistency checking and error reporting

fs/ext4/ext4.h

@@ -1529,6 +1529,7 @@ struct ext4_sb_info {
 	unsigned int s_mount_opt2;
 	unsigned long s_mount_flags;
 	unsigned int s_def_mount_opt;
+	unsigned int s_def_mount_opt2;
 	ext4_fsblk_t s_sb_block;
 	atomic64_t s_resv_clusters;
 	kuid_t s_resuid;

fs/ext4/extents.c

@@ -3251,7 +3251,7 @@ static int ext4_split_extent_at(handle_t *handle,
 		ext4_ext_mark_unwritten(ex2);
 
 	err = ext4_ext_insert_extent(handle, inode, ppath, &newex, flags);
-	if (err != -ENOSPC && err != -EDQUOT)
+	if (err != -ENOSPC && err != -EDQUOT && err != -ENOMEM)
 		goto out;
 
 	if (EXT4_EXT_MAY_ZEROOUT & split_flag) {

fs/ext4/fast_commit.c

@@ -1332,8 +1332,14 @@ struct dentry_info_args {
 	char *dname;
 };
 
+/* Same as struct ext4_fc_tl, but uses native endianness fields */
+struct ext4_fc_tl_mem {
+	u16 fc_tag;
+	u16 fc_len;
+};
+
 static inline void tl_to_darg(struct dentry_info_args *darg,
-			      struct ext4_fc_tl *tl, u8 *val)
+			      struct ext4_fc_tl_mem *tl, u8 *val)
 {
 	struct ext4_fc_dentry_info fcd;
 
@@ -1345,16 +1351,18 @@ static inline void tl_to_darg(struct dentry_info_args *darg,
 	darg->dname_len = tl->fc_len - sizeof(struct ext4_fc_dentry_info);
 }
 
-static inline void ext4_fc_get_tl(struct ext4_fc_tl *tl, u8 *val)
+static inline void ext4_fc_get_tl(struct ext4_fc_tl_mem *tl, u8 *val)
 {
-	memcpy(tl, val, EXT4_FC_TAG_BASE_LEN);
-	tl->fc_len = le16_to_cpu(tl->fc_len);
-	tl->fc_tag = le16_to_cpu(tl->fc_tag);
+	struct ext4_fc_tl tl_disk;
+
+	memcpy(&tl_disk, val, EXT4_FC_TAG_BASE_LEN);
+	tl->fc_len = le16_to_cpu(tl_disk.fc_len);
+	tl->fc_tag = le16_to_cpu(tl_disk.fc_tag);
 }
 
 /* Unlink replay function */
-static int ext4_fc_replay_unlink(struct super_block *sb, struct ext4_fc_tl *tl,
-				 u8 *val)
+static int ext4_fc_replay_unlink(struct super_block *sb,
+				 struct ext4_fc_tl_mem *tl, u8 *val)
 {
 	struct inode *inode, *old_parent;
 	struct qstr entry;
@@ -1451,8 +1459,8 @@ static int ext4_fc_replay_link_internal(struct super_block *sb,
 }
 
 /* Link replay function */
-static int ext4_fc_replay_link(struct super_block *sb, struct ext4_fc_tl *tl,
-			       u8 *val)
+static int ext4_fc_replay_link(struct super_block *sb,
+			       struct ext4_fc_tl_mem *tl, u8 *val)
 {
 	struct inode *inode;
 	struct dentry_info_args darg;
@@ -1506,8 +1514,8 @@ static int ext4_fc_record_modified_inode(struct super_block *sb, int ino)
 /*
  * Inode replay function
  */
-static int ext4_fc_replay_inode(struct super_block *sb, struct ext4_fc_tl *tl,
-				u8 *val)
+static int ext4_fc_replay_inode(struct super_block *sb,
+				struct ext4_fc_tl_mem *tl, u8 *val)
 {
 	struct ext4_fc_inode fc_inode;
 	struct ext4_inode *raw_inode;
@@ -1609,8 +1617,8 @@ static int ext4_fc_replay_inode(struct super_block *sb, struct ext4_fc_tl *tl,
  * inode for which we are trying to create a dentry here, should already have
  * been replayed before we start here.
  */
-static int ext4_fc_replay_create(struct super_block *sb, struct ext4_fc_tl *tl,
-				 u8 *val)
+static int ext4_fc_replay_create(struct super_block *sb,
+				 struct ext4_fc_tl_mem *tl, u8 *val)
 {
 	int ret = 0;
 	struct inode *inode = NULL;
@@ -1708,7 +1716,7 @@ int ext4_fc_record_regions(struct super_block *sb, int ino,
 
 /* Replay add range tag */
 static int ext4_fc_replay_add_range(struct super_block *sb,
-				    struct ext4_fc_tl *tl, u8 *val)
+				    struct ext4_fc_tl_mem *tl, u8 *val)
 {
 	struct ext4_fc_add_range fc_add_ex;
 	struct ext4_extent newex, *ex;
@@ -1828,8 +1836,8 @@ static int ext4_fc_replay_add_range(struct super_block *sb,
 
 /* Replay DEL_RANGE tag */
 static int
-ext4_fc_replay_del_range(struct super_block *sb, struct ext4_fc_tl *tl,
-			 u8 *val)
+ext4_fc_replay_del_range(struct super_block *sb,
+			 struct ext4_fc_tl_mem *tl, u8 *val)
 {
 	struct inode *inode;
 	struct ext4_fc_del_range lrange;
@@ -2025,7 +2033,7 @@ static int ext4_fc_replay_scan(journal_t *journal,
 	struct ext4_fc_replay_state *state;
 	int ret = JBD2_FC_REPLAY_CONTINUE;
 	struct ext4_fc_add_range ext;
-	struct ext4_fc_tl tl;
+	struct ext4_fc_tl_mem tl;
 	struct ext4_fc_tail tail;
 	__u8 *start, *end, *cur, *val;
 	struct ext4_fc_head head;
@@ -2144,7 +2152,7 @@ static int ext4_fc_replay(journal_t *journal, struct buffer_head *bh,
 {
 	struct super_block *sb = journal->j_private;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	struct ext4_fc_tl tl;
+	struct ext4_fc_tl_mem tl;
 	__u8 *start, *end, *cur, *val;
 	int ret = JBD2_FC_REPLAY_CONTINUE;
 	struct ext4_fc_replay_state *state = &sbi->s_fc_replay_state;

fs/ext4/file.c

@@ -202,8 +202,9 @@ ext4_extending_io(struct inode *inode, loff_t offset, size_t len)
 	return false;
 }
 
-/* Is IO overwriting allocated and initialized blocks? */
-static bool ext4_overwrite_io(struct inode *inode, loff_t pos, loff_t len)
+/* Is IO overwriting allocated or initialized blocks? */
+static bool ext4_overwrite_io(struct inode *inode,
+			      loff_t pos, loff_t len, bool *unwritten)
 {
 	struct ext4_map_blocks map;
 	unsigned int blkbits = inode->i_blkbits;
@@ -217,12 +218,15 @@ static bool ext4_overwrite_io(struct inode *inode, loff_t pos, loff_t len)
 	blklen = map.m_len;
 
 	err = ext4_map_blocks(NULL, inode, &map, 0);
+	if (err != blklen)
+		return false;
 	/*
 	 * 'err==len' means that all of the blocks have been preallocated,
-	 * regardless of whether they have been initialized or not. To exclude
-	 * unwritten extents, we need to check m_flags.
+	 * regardless of whether they have been initialized or not. We need to
+	 * check m_flags to distinguish the unwritten extents.
 	 */
-	return err == blklen && (map.m_flags & EXT4_MAP_MAPPED);
+	*unwritten = !(map.m_flags & EXT4_MAP_MAPPED);
+	return true;
 }
 
 static ssize_t ext4_generic_write_checks(struct kiocb *iocb,
@@ -431,11 +435,16 @@ static const struct iomap_dio_ops ext4_dio_write_ops = {
  * - For extending writes case we don't take the shared lock, since it requires
  *   updating inode i_disksize and/or orphan handling with exclusive lock.
  *
- * - shared locking will only be true mostly with overwrites. Otherwise we will
- *   switch to exclusive i_rwsem lock.
+ * - shared locking will only be true mostly with overwrites, including
+ *   initialized blocks and unwritten blocks. For overwrite unwritten blocks
+ *   we protect splitting extents by i_data_sem in ext4_inode_info, so we can
+ *   also release exclusive i_rwsem lock.
+ *
+ * - Otherwise we will switch to exclusive i_rwsem lock.
  */
 static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from,
-				     bool *ilock_shared, bool *extend)
+				     bool *ilock_shared, bool *extend,
+				     bool *unwritten)
 {
 	struct file *file = iocb->ki_filp;
 	struct inode *inode = file_inode(file);
@@ -459,7 +468,7 @@ static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from,
 	 * in file_modified().
 	 */
 	if (*ilock_shared && (!IS_NOSEC(inode) || *extend ||
-	     !ext4_overwrite_io(inode, offset, count))) {
+	     !ext4_overwrite_io(inode, offset, count, unwritten))) {
 		if (iocb->ki_flags & IOCB_NOWAIT) {
 			ret = -EAGAIN;
 			goto out;
@@ -491,7 +500,7 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	loff_t offset = iocb->ki_pos;
 	size_t count = iov_iter_count(from);
 	const struct iomap_ops *iomap_ops = &ext4_iomap_ops;
-	bool extend = false, unaligned_io = false;
+	bool extend = false, unaligned_io = false, unwritten = false;
 	bool ilock_shared = true;
 
 	/*
@@ -534,7 +543,8 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 		return ext4_buffered_write_iter(iocb, from);
 	}
 
-	ret = ext4_dio_write_checks(iocb, from, &ilock_shared, &extend);
+	ret = ext4_dio_write_checks(iocb, from,
+				    &ilock_shared, &extend, &unwritten);
 	if (ret <= 0)
 		return ret;
 
@@ -582,7 +592,7 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 		ext4_journal_stop(handle);
 	}
 
-	if (ilock_shared)
+	if (ilock_shared && !unwritten)
 		iomap_ops = &ext4_iomap_overwrite_ops;
 	ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops,
 			   (unaligned_io || extend) ? IOMAP_DIO_FORCE_WAIT : 0,

fs/ext4/inode.c

@@ -4872,13 +4872,6 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 		goto bad_inode;
 	raw_inode = ext4_raw_inode(&iloc);
 
-	if ((ino == EXT4_ROOT_INO) && (raw_inode->i_links_count == 0)) {
-		ext4_error_inode(inode, function, line, 0,
-				 "iget: root inode unallocated");
-		ret = -EFSCORRUPTED;
-		goto bad_inode;
-	}
-
 	if ((flags & EXT4_IGET_HANDLE) &&
 	    (raw_inode->i_links_count == 0) && (raw_inode->i_mode == 0)) {
 		ret = -ESTALE;
@@ -4951,10 +4944,15 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 	 * NeilBrown 1999oct15
 	 */
 	if (inode->i_nlink == 0) {
-		if ((inode->i_mode == 0 ||
+		if ((inode->i_mode == 0 || flags & EXT4_IGET_SPECIAL ||
 		     !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) &&
 		    ino != EXT4_BOOT_LOADER_INO) {
-			/* this inode is deleted */
+			/* this inode is deleted or unallocated */
+			if (flags & EXT4_IGET_SPECIAL) {
+				ext4_error_inode(inode, function, line, 0,
+						 "iget: special inode unallocated");
+				ret = -EFSCORRUPTED;
+			} else
 				ret = -ESTALE;
 			goto bad_inode;
 		}
@@ -5788,7 +5786,7 @@ static int ext4_meta_trans_blocks(struct inode *inode, int lblocks,
 	ext4_group_t groups, ngroups = ext4_get_groups_count(inode->i_sb);
 	int gdpblocks;
 	int idxblocks;
-	int ret = 0;
+	int ret;
 
 	/*
 	 * How many index blocks need to touch to map @lblocks logical blocks

fs/ext4/ioctl.c

@@ -155,9 +155,6 @@ static int ext4_update_backup_sb(struct super_block *sb,
 	set_buffer_uptodate(bh);
 	unlock_buffer(bh);
 
-	if (err)
-		goto out_bh;
-
 	if (handle) {
 		err = ext4_handle_dirty_metadata(handle, NULL, bh);
 		if (err)

fs/ext4/namei.c

@@ -3872,10 +3872,17 @@ static int ext4_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 			if (new.dir != old.dir && EXT4_DIR_LINK_MAX(new.dir))
 				goto end_rename;
 		}
+		/*
+		 * We need to protect against old.inode directory getting
+		 * converted from inline directory format into a normal one.
+		 */
+		inode_lock_nested(old.inode, I_MUTEX_NONDIR2);
 		retval = ext4_rename_dir_prepare(handle, &old);
-		if (retval)
+		if (retval) {
+			inode_unlock(old.inode);
 			goto end_rename;
 		}
+	}
 	/*
 	 * If we're renaming a file within an inline_data dir and adding or
 	 * setting the new dirent causes a conversion from inline_data to
@@ -4006,6 +4013,8 @@ static int ext4_rename(struct mnt_idmap *idmap, struct inode *old_dir,
 	} else {
 		ext4_journal_stop(handle);
 	}
+	if (old.dir_bh)
+		inode_unlock(old.inode);
 release_bh:
 	brelse(old.dir_bh);
 	brelse(old.bh);

fs/ext4/super.c

@@ -2146,7 +2146,7 @@ static int ext4_parse_param(struct fs_context *fc, struct fs_parameter *param)
 		return 0;
 	case Opt_commit:
 		if (result.uint_32 == 0)
-			ctx->s_commit_interval = JBD2_DEFAULT_MAX_COMMIT_AGE;
+			result.uint_32 = JBD2_DEFAULT_MAX_COMMIT_AGE;
 		else if (result.uint_32 > INT_MAX / HZ) {
 			ext4_msg(NULL, KERN_ERR,
 				 "Invalid commit interval %d, "
@@ -2883,7 +2883,7 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct ext4_super_block *es = sbi->s_es;
-	int def_errors, def_mount_opt = sbi->s_def_mount_opt;
+	int def_errors;
 	const struct mount_opts *m;
 	char sep = nodefs ? '\n' : ',';
 
@@ -2895,15 +2895,28 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
 
 	for (m = ext4_mount_opts; m->token != Opt_err; m++) {
 		int want_set = m->flags & MOPT_SET;
+		int opt_2 = m->flags & MOPT_2;
+		unsigned int mount_opt, def_mount_opt;
+
 		if (((m->flags & (MOPT_SET|MOPT_CLEAR)) == 0) ||
 		    m->flags & MOPT_SKIP)
 			continue;
-		if (!nodefs && !(m->mount_opt & (sbi->s_mount_opt ^ def_mount_opt)))
-			continue; /* skip if same as the default */
+
+		if (opt_2) {
+			mount_opt = sbi->s_mount_opt2;
+			def_mount_opt = sbi->s_def_mount_opt2;
+		} else {
+			mount_opt = sbi->s_mount_opt;
+			def_mount_opt = sbi->s_def_mount_opt;
+		}
+		/* skip if same as the default */
+		if (!nodefs && !(m->mount_opt & (mount_opt ^ def_mount_opt)))
+			continue;
+		/* select Opt_noFoo vs Opt_Foo */
 		if ((want_set &&
-		     (sbi->s_mount_opt & m->mount_opt) != m->mount_opt) ||
-		    (!want_set && (sbi->s_mount_opt & m->mount_opt)))
-			continue; /* select Opt_noFoo vs Opt_Foo */
+		     (mount_opt & m->mount_opt) != m->mount_opt) ||
+		    (!want_set && (mount_opt & m->mount_opt)))
+			continue;
 		SEQ_OPTS_PRINT("%s", token2str(m->token));
 	}
 
@@ -2931,7 +2944,7 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
 	if (nodefs || sbi->s_stripe)
 		SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe);
 	if (nodefs || EXT4_MOUNT_DATA_FLAGS &
-			(sbi->s_mount_opt ^ def_mount_opt)) {
+			(sbi->s_mount_opt ^ sbi->s_def_mount_opt)) {
 		if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA)
 			SEQ_OPTS_PUTS("data=journal");
 		else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA)
@@ -4727,7 +4740,6 @@ static int ext4_group_desc_init(struct super_block *sb,
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	unsigned int db_count;
 	ext4_fsblk_t block;
-	int ret;
 	int i;
 
 	db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
@@ -4767,8 +4779,7 @@ static int ext4_group_desc_init(struct super_block *sb,
 			ext4_msg(sb, KERN_ERR,
 			       "can't read group descriptor %d", i);
 			sbi->s_gdb_count = i;
-			ret = PTR_ERR(bh);
-			goto out;
+			return PTR_ERR(bh);
 		}
 		rcu_read_lock();
 		rcu_dereference(sbi->s_group_desc)[i] = bh;
@@ -4777,13 +4788,10 @@ static int ext4_group_desc_init(struct super_block *sb,
 	sbi->s_gdb_count = db_count;
 	if (!ext4_check_descriptors(sb, logical_sb_block, first_not_zeroed)) {
 		ext4_msg(sb, KERN_ERR, "group descriptors corrupted!");
-		ret = -EFSCORRUPTED;
-		goto out;
+		return -EFSCORRUPTED;
 	}
+
 	return 0;
-out:
-	ext4_group_desc_free(sbi);
-	return ret;
 }
 
 static int ext4_load_and_init_journal(struct super_block *sb,
@@ -5075,6 +5083,7 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb)
 		goto failed_mount;
 
 	sbi->s_def_mount_opt = sbi->s_mount_opt;
+	sbi->s_def_mount_opt2 = sbi->s_mount_opt2;
 
 	err = ext4_check_opt_consistency(fc, sb);
 	if (err < 0)
@@ -5209,14 +5218,14 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb)
 	if (ext4_geometry_check(sb, es))
 		goto failed_mount;
 
-	err = ext4_group_desc_init(sb, es, logical_sb_block, &first_not_zeroed);
-	if (err)
-		goto failed_mount;
-
 	timer_setup(&sbi->s_err_report, print_daily_error_info, 0);
 	spin_lock_init(&sbi->s_error_lock);
 	INIT_WORK(&sbi->s_error_work, flush_stashed_error_work);
 
+	err = ext4_group_desc_init(sb, es, logical_sb_block, &first_not_zeroed);
+	if (err)
+		goto failed_mount3;
+
 	/* Register extent status tree shrinker */
 	if (ext4_es_register_shrinker(sbi))
 		goto failed_mount3;
@@ -5937,8 +5946,11 @@ static int ext4_load_journal(struct super_block *sb,
 	if (!really_read_only && journal_devnum &&
 	    journal_devnum != le32_to_cpu(es->s_journal_dev)) {
 		es->s_journal_dev = cpu_to_le32(journal_devnum);
-
-		/* Make sure we flush the recovery flag to disk. */
+		ext4_commit_super(sb);
+	}
+	if (!really_read_only && journal_inum &&
+	    journal_inum != le32_to_cpu(es->s_journal_inum)) {
+		es->s_journal_inum = cpu_to_le32(journal_inum);
 		ext4_commit_super(sb);
 	}
 

fs/ext4/xattr.c

@@ -184,27 +184,73 @@ ext4_xattr_handler(int name_index)
 }
 
 static int
-ext4_xattr_check_entries(struct ext4_xattr_entry *entry, void *end,
-			 void *value_start)
+check_xattrs(struct inode *inode, struct buffer_head *bh,
+	     struct ext4_xattr_entry *entry, void *end, void *value_start,
+	     const char *function, unsigned int line)
 {
 	struct ext4_xattr_entry *e = entry;
+	int err = -EFSCORRUPTED;
+	char *err_str;
+
+	if (bh) {
+		if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) ||
+		    BHDR(bh)->h_blocks != cpu_to_le32(1)) {
+			err_str = "invalid header";
+			goto errout;
+		}
+		if (buffer_verified(bh))
+			return 0;
+		if (!ext4_xattr_block_csum_verify(inode, bh)) {
+			err = -EFSBADCRC;
+			err_str = "invalid checksum";
+			goto errout;
+		}
+	} else {
+		struct ext4_xattr_ibody_header *header = value_start;
+
+		header -= 1;
+		if (end - (void *)header < sizeof(*header) + sizeof(u32)) {
+			err_str = "in-inode xattr block too small";
+			goto errout;
+		}
+		if (header->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC)) {
+			err_str = "bad magic number in in-inode xattr";
+			goto errout;
+		}
+	}
 
 	/* Find the end of the names list */
 	while (!IS_LAST_ENTRY(e)) {
 		struct ext4_xattr_entry *next = EXT4_XATTR_NEXT(e);
-		if ((void *)next >= end)
-			return -EFSCORRUPTED;
-		if (strnlen(e->e_name, e->e_name_len) != e->e_name_len)
-			return -EFSCORRUPTED;
+		if ((void *)next >= end) {
+			err_str = "e_name out of bounds";
+			goto errout;
+		}
+		if (strnlen(e->e_name, e->e_name_len) != e->e_name_len) {
+			err_str = "bad e_name length";
+			goto errout;
+		}
 		e = next;
 	}
 
 	/* Check the values */
 	while (!IS_LAST_ENTRY(entry)) {
 		u32 size = le32_to_cpu(entry->e_value_size);
+		unsigned long ea_ino = le32_to_cpu(entry->e_value_inum);
 
-		if (size > EXT4_XATTR_SIZE_MAX)
-			return -EFSCORRUPTED;
+		if (!ext4_has_feature_ea_inode(inode->i_sb) && ea_ino) {
+			err_str = "ea_inode specified without ea_inode feature enabled";
+			goto errout;
+		}
+		if (ea_ino && ((ea_ino == EXT4_ROOT_INO) ||
+			       !ext4_valid_inum(inode->i_sb, ea_ino))) {
+			err_str = "invalid ea_ino";
+			goto errout;
+		}
+		if (size > EXT4_XATTR_SIZE_MAX) {
+			err_str = "e_value size too large";
+			goto errout;
+		}
 
 		if (size != 0 && entry->e_value_inum == 0) {
 			u16 offs = le16_to_cpu(entry->e_value_offs);
@@ -216,66 +262,54 @@ ext4_xattr_check_entries(struct ext4_xattr_entry *entry, void *end,
 			 * the padded and unpadded sizes, since the size may
 			 * overflow to 0 when adding padding.
 			 */
-			if (offs > end - value_start)
-				return -EFSCORRUPTED;
+			if (offs > end - value_start) {
+				err_str = "e_value out of bounds";
+				goto errout;
+			}
 			value = value_start + offs;
 			if (value < (void *)e + sizeof(u32) ||
 			    size > end - value ||
-			    EXT4_XATTR_SIZE(size) > end - value)
-				return -EFSCORRUPTED;
+			    EXT4_XATTR_SIZE(size) > end - value) {
+				err_str = "overlapping e_value ";
+				goto errout;
+			}
 		}
 		entry = EXT4_XATTR_NEXT(entry);
 	}
-
+	if (bh)
+		set_buffer_verified(bh);
 	return 0;
+
+errout:
+	if (bh)
+		__ext4_error_inode(inode, function, line, 0, -err,
+				   "corrupted xattr block %llu: %s",
+				   (unsigned long long) bh->b_blocknr,
+				   err_str);
+	else
+		__ext4_error_inode(inode, function, line, 0, -err,
+				   "corrupted in-inode xattr: %s", err_str);
+	return err;
 }
 
 static inline int
 __ext4_xattr_check_block(struct inode *inode, struct buffer_head *bh,
 			 const char *function, unsigned int line)
 {
-	int error = -EFSCORRUPTED;
-
-	if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) ||
-	    BHDR(bh)->h_blocks != cpu_to_le32(1))
-		goto errout;
-	if (buffer_verified(bh))
-		return 0;
-
-	error = -EFSBADCRC;
-	if (!ext4_xattr_block_csum_verify(inode, bh))
-		goto errout;
-	error = ext4_xattr_check_entries(BFIRST(bh), bh->b_data + bh->b_size,
-					 bh->b_data);
-errout:
-	if (error)
-		__ext4_error_inode(inode, function, line, 0, -error,
-				   "corrupted xattr block %llu",
-				   (unsigned long long) bh->b_blocknr);
-	else
-		set_buffer_verified(bh);
-	return error;
+	return check_xattrs(inode, bh, BFIRST(bh), bh->b_data + bh->b_size,
+			    bh->b_data, function, line);
 }
 
 #define ext4_xattr_check_block(inode, bh) \
 	__ext4_xattr_check_block((inode), (bh),  __func__, __LINE__)
 
 
-static int
+static inline int
 __xattr_check_inode(struct inode *inode, struct ext4_xattr_ibody_header *header,
 			 void *end, const char *function, unsigned int line)
 {
-	int error = -EFSCORRUPTED;
-
-	if (end - (void *)header < sizeof(*header) + sizeof(u32) ||
-	    (header->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC)))
-		goto errout;
-	error = ext4_xattr_check_entries(IFIRST(header), end, IFIRST(header));
-errout:
-	if (error)
-		__ext4_error_inode(inode, function, line, 0, -error,
-				   "corrupted in-inode xattr");
-	return error;
+	return check_xattrs(inode, NULL, IFIRST(header), end, IFIRST(header),
+			    function, line);
 }
 
 #define xattr_check_inode(inode, header, end) \
@@ -388,6 +422,17 @@ static int ext4_xattr_inode_iget(struct inode *parent, unsigned long ea_ino,
 	struct inode *inode;
 	int err;
 
+	/*
+	 * We have to check for this corruption early as otherwise
+	 * iget_locked() could wait indefinitely for the state of our
+	 * parent inode.
+	 */
+	if (parent->i_ino == ea_ino) {
+		ext4_error(parent->i_sb,
+			   "Parent and EA inode have the same ino %lu", ea_ino);
+		return -EFSCORRUPTED;
+	}
+
 	inode = ext4_iget(parent->i_sb, ea_ino, EXT4_IGET_NORMAL);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
@@ -1438,6 +1483,13 @@ static struct inode *ext4_xattr_inode_create(handle_t *handle,
 	uid_t owner[2] = { i_uid_read(inode), i_gid_read(inode) };
 	int err;
 
+	if (inode->i_sb->s_root == NULL) {
+		ext4_warning(inode->i_sb,
+			     "refuse to create EA inode when umounting");
+		WARN_ON(1);
+		return ERR_PTR(-EINVAL);
+	}
+
 	/*
 	 * Let the next inode be the goal, so we try and allocate the EA inode
 	 * in the same group, or nearby one.
@@ -2567,9 +2619,8 @@ static int ext4_xattr_move_to_block(handle_t *handle, struct inode *inode,
 
 	is = kzalloc(sizeof(struct ext4_xattr_ibody_find), GFP_NOFS);
 	bs = kzalloc(sizeof(struct ext4_xattr_block_find), GFP_NOFS);
-	buffer = kvmalloc(value_size, GFP_NOFS);
 	b_entry_name = kmalloc(entry->e_name_len + 1, GFP_NOFS);
-	if (!is || !bs || !buffer || !b_entry_name) {
+	if (!is || !bs || !b_entry_name) {
 		error = -ENOMEM;
 		goto out;
 	}
@@ -2581,12 +2632,18 @@ static int ext4_xattr_move_to_block(handle_t *handle, struct inode *inode,
 
 	/* Save the entry name and the entry value */
 	if (entry->e_value_inum) {
+		buffer = kvmalloc(value_size, GFP_NOFS);
+		if (!buffer) {
+			error = -ENOMEM;
+			goto out;
+		}
+
 		error = ext4_xattr_inode_get(inode, entry, buffer, value_size);
 		if (error)
 			goto out;
 	} else {
 		size_t value_offs = le16_to_cpu(entry->e_value_offs);
-		memcpy(buffer, (void *)IFIRST(header) + value_offs, value_size);
+		buffer = (void *)IFIRST(header) + value_offs;
 	}
 
 	memcpy(b_entry_name, entry->e_name, entry->e_name_len);
@@ -2601,24 +2658,25 @@ static int ext4_xattr_move_to_block(handle_t *handle, struct inode *inode,
 	if (error)
 		goto out;
 
-	/* Remove the chosen entry from the inode */
-	error = ext4_xattr_ibody_set(handle, inode, &i, is);
-	if (error)
-		goto out;
-
 	i.value = buffer;
 	i.value_len = value_size;
 	error = ext4_xattr_block_find(inode, &i, bs);
 	if (error)
 		goto out;
 
-	/* Add entry which was removed from the inode into the block */
+	/* Move ea entry from the inode into the block */
 	error = ext4_xattr_block_set(handle, inode, &i, bs);
 	if (error)
 		goto out;
-	error = 0;
+
+	/* Remove the chosen entry from the inode */
+	i.value = NULL;
+	i.value_len = 0;
+	error = ext4_xattr_ibody_set(handle, inode, &i, is);
+
 out:
 	kfree(b_entry_name);
+	if (entry->e_value_inum && buffer)
 		kvfree(buffer);
 	if (is)
 		brelse(is->iloc.bh);

fs/jbd2/transaction.c

@@ -1010,36 +1010,28 @@ do_get_write_access(handle_t *handle, struct journal_head *jh,
 	 * ie. locked but not dirty) or tune2fs (which may actually have
 	 * the buffer dirtied, ugh.)  */
 
-	if (buffer_dirty(bh)) {
-		/*
-		 * First question: is this buffer already part of the current
-		 * transaction or the existing committing transaction?
-		 */
-		if (jh->b_transaction) {
-			J_ASSERT_JH(jh,
-				jh->b_transaction == transaction ||
-				jh->b_transaction ==
-					journal->j_committing_transaction);
-			if (jh->b_next_transaction)
-				J_ASSERT_JH(jh, jh->b_next_transaction ==
-							transaction);
+	if (buffer_dirty(bh) && jh->b_transaction) {
 		warn_dirty_buffer(bh);
-		}
 		/*
-		 * In any case we need to clean the dirty flag and we must
-		 * do it under the buffer lock to be sure we don't race
-		 * with running write-out.
+		 * We need to clean the dirty flag and we must do it under the
+		 * buffer lock to be sure we don't race with running write-out.
 		 */
 		JBUFFER_TRACE(jh, "Journalling dirty buffer");
 		clear_buffer_dirty(bh);
+		/*
+		 * The buffer is going to be added to BJ_Reserved list now and
+		 * nothing guarantees jbd2_journal_dirty_metadata() will be
+		 * ever called for it. So we need to set jbddirty bit here to
+		 * make sure the buffer is dirtied and written out when the
+		 * journaling machinery is done with it.
+		 */
 		set_buffer_jbddirty(bh);
 	}
 
-	unlock_buffer(bh);
-
 	error = -EROFS;
 	if (is_handle_aborted(handle)) {
 		spin_unlock(&jh->b_state_lock);
+		unlock_buffer(bh);
 		goto out;
 	}
 	error = 0;
@@ -1049,8 +1041,10 @@ do_get_write_access(handle_t *handle, struct journal_head *jh,
 	 * b_next_transaction points to it
 	 */
 	if (jh->b_transaction == transaction ||
-	    jh->b_next_transaction == transaction)
+	    jh->b_next_transaction == transaction) {
+		unlock_buffer(bh);
 		goto done;
+	}
 
 	/*
 	 * this is the first time this transaction is touching this buffer,
@@ -1074,10 +1068,24 @@ do_get_write_access(handle_t *handle, struct journal_head *jh,
 		 */
 		smp_wmb();
 		spin_lock(&journal->j_list_lock);
+		if (test_clear_buffer_dirty(bh)) {
+			/*
+			 * Execute buffer dirty clearing and jh->b_transaction
+			 * assignment under journal->j_list_lock locked to
+			 * prevent bh being removed from checkpoint list if
+			 * the buffer is in an intermediate state (not dirty
+			 * and jh->b_transaction is NULL).
+			 */
+			JBUFFER_TRACE(jh, "Journalling dirty buffer");
+			set_buffer_jbddirty(bh);
+		}
 		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
 		spin_unlock(&journal->j_list_lock);
+		unlock_buffer(bh);
 		goto done;
 	}
+	unlock_buffer(bh);
+
 	/*
 	 * If there is already a copy-out version of this buffer, then we don't
 	 * need to make another one