Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kern…

…el/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;
@@ -4951,11 +4944,16 @@ 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 */
-			ret = -ESTALE;
+			/* 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;
 		}
 		/* The only unlinked inodes we let through here have
@@ -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

@@ -3872,9 +3872,16 @@ 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
@@ -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);