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

Pull ext4 updates from Ted Ts'o:
 "This merge window, we've added some performance improvements in how we
  handle inode locking in the read/write paths, and improving the
  performance of Direct I/O overwrites.

  We also now record the error code which caused the first and most
  recent ext4_error() report in the superblock, to make it easier to
  root cause problems in production systems.

  There are also many of the usual cleanups and miscellaneous bug fixes"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (49 commits)
  jbd2: clean __jbd2_journal_abort_hard() and __journal_abort_soft()
  jbd2: make sure ESHUTDOWN to be recorded in the journal superblock
  ext4, jbd2: ensure panic when aborting with zero errno
  jbd2: switch to use jbd2_journal_abort() when failed to submit the commit record
  jbd2_seq_info_next should increase position index
  jbd2: remove pointless assertion in __journal_remove_journal_head
  ext4,jbd2: fix comment and code style
  jbd2: delete the duplicated words in the comments
  ext4: fix extent_status trace points
  ext4: fix symbolic enum printing in trace output
  ext4: choose hardlimit when softlimit is larger than hardlimit in ext4_statfs_project()
  ext4: fix race conditions in ->d_compare() and ->d_hash()
  ext4: make dioread_nolock the default
  ext4: fix extent_status fragmentation for plain files
  jbd2: clear JBD2_ABORT flag before journal_reset to update log tail info when load journal
  ext4: drop ext4_kvmalloc()
  ext4: Add EXT4_IOC_FSGETXATTR/EXT4_IOC_FSSETXATTR to compat_ioctl
  ext4: remove unused macro MPAGE_DA_EXTENT_TAIL
  ext4: add missing braces in ext4_ext_drop_refs()
  ext4: fix some nonstandard indentation in extents.c
  ...

Documentation/admin-guide/ext4.rst

@@ -92,6 +92,8 @@ Currently Available
 * efficient new ordered mode in JBD2 and ext4 (avoid using buffer head to force
   the ordering)
 * Case-insensitive file name lookups
+* file-based encryption support (fscrypt)
+* file-based verity support (fsverity)
 
 [1] Filesystems with a block size of 1k may see a limit imposed by the
 directory hash tree having a maximum depth of two.

Documentation/filesystems/fscrypt.rst

@@ -1016,9 +1016,9 @@ astute users may notice some differences in behavior:
 - Direct I/O is not supported on encrypted files.  Attempts to use
   direct I/O on such files will fall back to buffered I/O.
 
-- The fallocate operations FALLOC_FL_COLLAPSE_RANGE,
+- The fallocate operations FALLOC_FL_COLLAPSE_RANGE and
-  FALLOC_FL_INSERT_RANGE, and FALLOC_FL_ZERO_RANGE are not supported
+  FALLOC_FL_INSERT_RANGE are not supported on encrypted files and will
-  on encrypted files and will fail with EOPNOTSUPP.
+  fail with EOPNOTSUPP.
 
 - Online defragmentation of encrypted files is not supported.  The
   EXT4_IOC_MOVE_EXT and F2FS_IOC_MOVE_RANGE ioctls will fail with

fs/ext4/Kconfig

@@ -4,12 +4,7 @@
 # kernels after the removal of ext3 driver.
 config EXT3_FS
 	tristate "The Extended 3 (ext3) filesystem"
-	# These must match EXT4_FS selects...
 	select EXT4_FS
-	select JBD2
-	select CRC16
-	select CRYPTO
-	select CRYPTO_CRC32C
 	help
 	  This config option is here only for backward compatibility. ext3
 	  filesystem is now handled by the ext4 driver.
@@ -33,7 +28,6 @@ config EXT3_FS_SECURITY
 
 config EXT4_FS
 	tristate "The Extended 4 (ext4) filesystem"
-	# Please update EXT3_FS selects when changing these
 	select JBD2
 	select CRC16
 	select CRYPTO

fs/ext4/balloc.c

@@ -371,7 +371,8 @@ static int ext4_validate_block_bitmap(struct super_block *sb,
 	if (buffer_verified(bh))
 		goto verified;
 	if (unlikely(!ext4_block_bitmap_csum_verify(sb, block_group,
-			desc, bh))) {
+						    desc, bh) ||
+		     ext4_simulate_fail(sb, EXT4_SIM_BBITMAP_CRC))) {
 		ext4_unlock_group(sb, block_group);
 		ext4_error(sb, "bg %u: bad block bitmap checksum", block_group);
 		ext4_mark_group_bitmap_corrupted(sb, block_group,
@@ -505,7 +506,9 @@ int ext4_wait_block_bitmap(struct super_block *sb, ext4_group_t block_group,
 	if (!desc)
 		return -EFSCORRUPTED;
 	wait_on_buffer(bh);
+	ext4_simulate_fail_bh(sb, bh, EXT4_SIM_BBITMAP_EIO);
 	if (!buffer_uptodate(bh)) {
+		ext4_set_errno(sb, EIO);
 		ext4_error(sb, "Cannot read block bitmap - "
 			   "block_group = %u, block_bitmap = %llu",
 			   block_group, (unsigned long long) bh->b_blocknr);

fs/ext4/dir.c

@@ -462,7 +462,6 @@ int ext4_htree_store_dirent(struct file *dir_file, __u32 hash,
 	new_fn->name_len = ent_name->len;
 	new_fn->file_type = dirent->file_type;
 	memcpy(new_fn->name, ent_name->name, ent_name->len);
-	new_fn->name[ent_name->len] = 0;
 
 	while (*p) {
 		parent = *p;
@@ -672,9 +671,11 @@ static int ext4_d_compare(const struct dentry *dentry, unsigned int len,
 			  const char *str, const struct qstr *name)
 {
 	struct qstr qstr = {.name = str, .len = len };
-	struct inode *inode = dentry->d_parent->d_inode;
+	const struct dentry *parent = READ_ONCE(dentry->d_parent);
+	const struct inode *inode = READ_ONCE(parent->d_inode);
 
-	if (!IS_CASEFOLDED(inode) || !EXT4_SB(inode->i_sb)->s_encoding) {
+	if (!inode || !IS_CASEFOLDED(inode) ||
+	    !EXT4_SB(inode->i_sb)->s_encoding) {
 		if (len != name->len)
 			return -1;
 		return memcmp(str, name->name, len);
@@ -687,10 +688,11 @@ static int ext4_d_hash(const struct dentry *dentry, struct qstr *str)
 {
 	const struct ext4_sb_info *sbi = EXT4_SB(dentry->d_sb);
 	const struct unicode_map *um = sbi->s_encoding;
+	const struct inode *inode = READ_ONCE(dentry->d_inode);
 	unsigned char *norm;
 	int len, ret = 0;
 
-	if (!IS_CASEFOLDED(dentry->d_inode) || !um)
+	if (!inode || !IS_CASEFOLDED(inode) || !um)
 		return 0;
 
 	norm = kmalloc(PATH_MAX, GFP_ATOMIC);

fs/ext4/ext4.h

@@ -1052,8 +1052,6 @@ struct ext4_inode_info {
 	/* allocation reservation info for delalloc */
 	/* In case of bigalloc, this refer to clusters rather than blocks */
 	unsigned int i_reserved_data_blocks;
-	ext4_lblk_t i_da_metadata_calc_last_lblock;
-	int i_da_metadata_calc_len;
 
 	/* pending cluster reservations for bigalloc file systems */
 	struct ext4_pending_tree i_pending_tree;
@@ -1343,7 +1341,8 @@ struct ext4_super_block {
 	__u8	s_lastcheck_hi;
 	__u8	s_first_error_time_hi;
 	__u8	s_last_error_time_hi;
-	__u8	s_pad[2];
+	__u8	s_first_error_errcode;
+	__u8    s_last_error_errcode;
 	__le16  s_encoding;		/* Filename charset encoding */
 	__le16  s_encoding_flags;	/* Filename charset encoding flags */
 	__le32	s_reserved[95];		/* Padding to the end of the block */
@@ -1556,6 +1555,9 @@ struct ext4_sb_info {
 	/* Barrier between changing inodes' journal flags and writepages ops. */
 	struct percpu_rw_semaphore s_journal_flag_rwsem;
 	struct dax_device *s_daxdev;
+#ifdef CONFIG_EXT4_DEBUG
+	unsigned long s_simulate_fail;
+#endif
 };
 
 static inline struct ext4_sb_info *EXT4_SB(struct super_block *sb)
@@ -1574,6 +1576,66 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino)
 		 ino <= le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count));
 }
 
+/*
+ * Simulate_fail codes
+ */
+#define EXT4_SIM_BBITMAP_EIO	1
+#define EXT4_SIM_BBITMAP_CRC	2
+#define EXT4_SIM_IBITMAP_EIO	3
+#define EXT4_SIM_IBITMAP_CRC	4
+#define EXT4_SIM_INODE_EIO	5
+#define EXT4_SIM_INODE_CRC	6
+#define EXT4_SIM_DIRBLOCK_EIO	7
+#define EXT4_SIM_DIRBLOCK_CRC	8
+
+static inline bool ext4_simulate_fail(struct super_block *sb,
+				     unsigned long code)
+{
+#ifdef CONFIG_EXT4_DEBUG
+	struct ext4_sb_info *sbi = EXT4_SB(sb);
+
+	if (unlikely(sbi->s_simulate_fail == code)) {
+		sbi->s_simulate_fail = 0;
+		return true;
+	}
+#endif
+	return false;
+}
+
+static inline void ext4_simulate_fail_bh(struct super_block *sb,
+					 struct buffer_head *bh,
+					 unsigned long code)
+{
+	if (!IS_ERR(bh) && ext4_simulate_fail(sb, code))
+		clear_buffer_uptodate(bh);
+}
+
+/*
+ * Error number codes for s_{first,last}_error_errno
+ *
+ * Linux errno numbers are architecture specific, so we need to translate
+ * them into something which is architecture independent.   We don't define
+ * codes for all errno's; just the ones which are most likely to be the cause
+ * of an ext4_error() call.
+ */
+#define EXT4_ERR_UNKNOWN	 1
+#define EXT4_ERR_EIO		 2
+#define EXT4_ERR_ENOMEM		 3
+#define EXT4_ERR_EFSBADCRC	 4
+#define EXT4_ERR_EFSCORRUPTED	 5
+#define EXT4_ERR_ENOSPC		 6
+#define EXT4_ERR_ENOKEY		 7
+#define EXT4_ERR_EROFS		 8
+#define EXT4_ERR_EFBIG		 9
+#define EXT4_ERR_EEXIST		10
+#define EXT4_ERR_ERANGE		11
+#define EXT4_ERR_EOVERFLOW	12
+#define EXT4_ERR_EBUSY		13
+#define EXT4_ERR_ENOTDIR	14
+#define EXT4_ERR_ENOTEMPTY	15
+#define EXT4_ERR_ESHUTDOWN	16
+#define EXT4_ERR_EFAULT		17
+
 /*
  * Inode dynamic state flags
  */
@@ -2628,7 +2690,6 @@ extern int ext4_issue_zeroout(struct inode *inode, ext4_lblk_t lblk,
 /* indirect.c */
 extern int ext4_ind_map_blocks(handle_t *handle, struct inode *inode,
 				struct ext4_map_blocks *map, int flags);
-extern int ext4_ind_calc_metadata_amount(struct inode *inode, sector_t lblock);
 extern int ext4_ind_trans_blocks(struct inode *inode, int nrblocks);
 extern void ext4_ind_truncate(handle_t *, struct inode *inode);
 extern int ext4_ind_remove_space(handle_t *handle, struct inode *inode,
@@ -2679,8 +2740,6 @@ extern struct buffer_head *ext4_sb_bread(struct super_block *sb,
 extern int ext4_seq_options_show(struct seq_file *seq, void *offset);
 extern int ext4_calculate_overhead(struct super_block *sb);
 extern void ext4_superblock_csum_set(struct super_block *sb);
-extern void *ext4_kvmalloc(size_t size, gfp_t flags);
-extern void *ext4_kvzalloc(size_t size, gfp_t flags);
 extern int ext4_alloc_flex_bg_array(struct super_block *sb,
 				    ext4_group_t ngroup);
 extern const char *ext4_decode_error(struct super_block *sb, int errno,
@@ -2688,6 +2747,7 @@ extern const char *ext4_decode_error(struct super_block *sb, int errno,
 extern void ext4_mark_group_bitmap_corrupted(struct super_block *sb,
 					     ext4_group_t block_group,
 					     unsigned int flags);
+extern void ext4_set_errno(struct super_block *sb, int err);
 
 extern __printf(4, 5)
 void __ext4_error(struct super_block *, const char *, unsigned int,
@@ -3254,7 +3314,6 @@ struct ext4_extent;
 #define EXT_MAX_BLOCKS	0xffffffff
 
 extern int ext4_ext_tree_init(handle_t *handle, struct inode *);
-extern int ext4_ext_writepage_trans_blocks(struct inode *, int);
 extern int ext4_ext_index_trans_blocks(struct inode *inode, int extents);
 extern int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			       struct ext4_map_blocks *map, int flags);
@@ -3271,14 +3330,9 @@ extern int ext4_convert_unwritten_io_end_vec(handle_t *handle,
 					     ext4_io_end_t *io_end);
 extern int ext4_map_blocks(handle_t *handle, struct inode *inode,
 			   struct ext4_map_blocks *map, int flags);
-extern int ext4_ext_calc_metadata_amount(struct inode *inode,
-					 ext4_lblk_t lblocks);
 extern int ext4_ext_calc_credits_for_single_extent(struct inode *inode,
 						   int num,
 						   struct ext4_ext_path *path);
-extern int ext4_can_extents_be_merged(struct inode *inode,
-				      struct ext4_extent *ex1,
-				      struct ext4_extent *ex2);
 extern int ext4_ext_insert_extent(handle_t *, struct inode *,
 				  struct ext4_ext_path **,
 				  struct ext4_extent *, int);
@@ -3294,8 +3348,6 @@ extern int ext4_get_es_cache(struct inode *inode,
 			     struct fiemap_extent_info *fieinfo,
 			     __u64 start, __u64 len);
 extern int ext4_ext_precache(struct inode *inode);
-extern int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len);
-extern int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len);
 extern int ext4_swap_extents(handle_t *handle, struct inode *inode1,
 				struct inode *inode2, ext4_lblk_t lblk1,
 			     ext4_lblk_t lblk2,  ext4_lblk_t count,
@@ -3390,6 +3442,7 @@ static inline void ext4_clear_io_unwritten_flag(ext4_io_end_t *io_end)
 }
 
 extern const struct iomap_ops ext4_iomap_ops;
+extern const struct iomap_ops ext4_iomap_overwrite_ops;
 extern const struct iomap_ops ext4_iomap_report_ops;
 
 static inline int ext4_buffer_uptodate(struct buffer_head *bh)

fs/ext4/ext4_extents.h

@@ -267,10 +267,5 @@ static inline void ext4_idx_store_pblock(struct ext4_extent_idx *ix,
 				     0xffff);
 }
 
-#define ext4_ext_dirty(handle, inode, path) \
-		__ext4_ext_dirty(__func__, __LINE__, (handle), (inode), (path))
-int __ext4_ext_dirty(const char *where, unsigned int line, handle_t *handle,
-		     struct inode *inode, struct ext4_ext_path *path);
-
 #endif /* _EXT4_EXTENTS */
 

fs/ext4/ext4_jbd2.c

@@ -7,6 +7,28 @@
 
 #include <trace/events/ext4.h>
 
+int ext4_inode_journal_mode(struct inode *inode)
+{
+	if (EXT4_JOURNAL(inode) == NULL)
+		return EXT4_INODE_WRITEBACK_DATA_MODE;	/* writeback */
+	/* We do not support data journalling with delayed allocation */
+	if (!S_ISREG(inode->i_mode) ||
+	    ext4_test_inode_flag(inode, EXT4_INODE_EA_INODE) ||
+	    test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ||
+	    (ext4_test_inode_flag(inode, EXT4_INODE_JOURNAL_DATA) &&
+	    !test_opt(inode->i_sb, DELALLOC))) {
+		/* We do not support data journalling for encrypted data */
+		if (S_ISREG(inode->i_mode) && IS_ENCRYPTED(inode))
+			return EXT4_INODE_ORDERED_DATA_MODE;  /* ordered */
+		return EXT4_INODE_JOURNAL_DATA_MODE;	/* journal data */
+	}
+	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA)
+		return EXT4_INODE_ORDERED_DATA_MODE;	/* ordered */
+	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)
+		return EXT4_INODE_WRITEBACK_DATA_MODE;	/* writeback */
+	BUG();
+}
+
 /* Just increment the non-pointer handle value */
 static handle_t *ext4_get_nojournal(void)
 {
@@ -58,6 +80,7 @@ static int ext4_journal_check_start(struct super_block *sb)
 	 * take the FS itself readonly cleanly.
 	 */
 	if (journal && is_journal_aborted(journal)) {
+		ext4_set_errno(sb, -journal->j_errno);
 		ext4_abort(sb, "Detected aborted journal");
 		return -EROFS;
 	}
@@ -249,6 +272,7 @@ int __ext4_forget(const char *where, unsigned int line, handle_t *handle,
 	if (err) {
 		ext4_journal_abort_handle(where, line, __func__,
 					  bh, handle, err);
+		ext4_set_errno(inode->i_sb, -err);
 		__ext4_abort(inode->i_sb, where, line,
 			   "error %d when attempting revoke", err);
 	}
@@ -320,6 +344,7 @@ int __ext4_handle_dirty_metadata(const char *where, unsigned int line,
 				es = EXT4_SB(inode->i_sb)->s_es;
 				es->s_last_error_block =
 					cpu_to_le64(bh->b_blocknr);
+				ext4_set_errno(inode->i_sb, EIO);
 				ext4_error_inode(inode, where, line,
 						 bh->b_blocknr,
 					"IO error syncing itable block");

fs/ext4/ext4_jbd2.h

@@ -463,27 +463,7 @@ int ext4_force_commit(struct super_block *sb);
 #define EXT4_INODE_ORDERED_DATA_MODE	0x02 /* ordered data mode */
 #define EXT4_INODE_WRITEBACK_DATA_MODE	0x04 /* writeback data mode */
 
-static inline int ext4_inode_journal_mode(struct inode *inode)
+int ext4_inode_journal_mode(struct inode *inode);
-{
-	if (EXT4_JOURNAL(inode) == NULL)
-		return EXT4_INODE_WRITEBACK_DATA_MODE;	/* writeback */
-	/* We do not support data journalling with delayed allocation */
-	if (!S_ISREG(inode->i_mode) ||
-	    ext4_test_inode_flag(inode, EXT4_INODE_EA_INODE) ||
-	    test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ||
-	    (ext4_test_inode_flag(inode, EXT4_INODE_JOURNAL_DATA) &&
-	    !test_opt(inode->i_sb, DELALLOC))) {
-		/* We do not support data journalling for encrypted data */
-		if (S_ISREG(inode->i_mode) && IS_ENCRYPTED(inode))
-			return EXT4_INODE_ORDERED_DATA_MODE;  /* ordered */
-		return EXT4_INODE_JOURNAL_DATA_MODE;	/* journal data */
-	}
-	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA)
-		return EXT4_INODE_ORDERED_DATA_MODE;	/* ordered */
-	if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)
-		return EXT4_INODE_WRITEBACK_DATA_MODE;	/* writeback */
-	BUG();
-}
 
 static inline int ext4_should_journal_data(struct inode *inode)
 {

fs/ext4/extents.c

@@ -161,8 +161,9 @@ static int ext4_ext_get_access(handle_t *handle, struct inode *inode,
  *  - ENOMEM
  *  - EIO
  */
-int __ext4_ext_dirty(const char *where, unsigned int line, handle_t *handle,
+static int __ext4_ext_dirty(const char *where, unsigned int line,
-		     struct inode *inode, struct ext4_ext_path *path)
+			    handle_t *handle, struct inode *inode,
+			    struct ext4_ext_path *path)
 {
 	int err;
 
@@ -179,6 +180,9 @@ int __ext4_ext_dirty(const char *where, unsigned int line, handle_t *handle,
 	return err;
 }
 
+#define ext4_ext_dirty(handle, inode, path) \
+		__ext4_ext_dirty(__func__, __LINE__, (handle), (inode), (path))
+
 static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode,
 			      struct ext4_ext_path *path,
 			      ext4_lblk_t block)
@@ -309,53 +313,6 @@ ext4_force_split_extent_at(handle_t *handle, struct inode *inode,
 			(nofail ? EXT4_GET_BLOCKS_METADATA_NOFAIL:0));
 }
 
-/*
- * Calculate the number of metadata blocks needed
- * to allocate @blocks
- * Worse case is one block per extent
- */
-int ext4_ext_calc_metadata_amount(struct inode *inode, ext4_lblk_t lblock)
-{
-	struct ext4_inode_info *ei = EXT4_I(inode);
-	int idxs;
-
-	idxs = ((inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header))
-		/ sizeof(struct ext4_extent_idx));
-
-	/*
-	 * If the new delayed allocation block is contiguous with the
-	 * previous da block, it can share index blocks with the
-	 * previous block, so we only need to allocate a new index
-	 * block every idxs leaf blocks.  At ldxs**2 blocks, we need
-	 * an additional index block, and at ldxs**3 blocks, yet
-	 * another index blocks.
-	 */
-	if (ei->i_da_metadata_calc_len &&
-	    ei->i_da_metadata_calc_last_lblock+1 == lblock) {
-		int num = 0;
-
-		if ((ei->i_da_metadata_calc_len % idxs) == 0)
-			num++;
-		if ((ei->i_da_metadata_calc_len % (idxs*idxs)) == 0)
-			num++;
-		if ((ei->i_da_metadata_calc_len % (idxs*idxs*idxs)) == 0) {
-			num++;
-			ei->i_da_metadata_calc_len = 0;
-		} else
-			ei->i_da_metadata_calc_len++;
-		ei->i_da_metadata_calc_last_lblock++;
-		return num;
-	}
-
-	/*
-	 * In the worst case we need a new set of index blocks at
-	 * every level of the inode's extent tree.
-	 */
-	ei->i_da_metadata_calc_len = 1;
-	ei->i_da_metadata_calc_last_lblock = lblock;
-	return ext_depth(inode) + 1;
-}
-
 static int
 ext4_ext_max_entries(struct inode *inode, int depth)
 {
@@ -492,6 +449,7 @@ static int __ext4_ext_check(const char *function, unsigned int line,
 	return 0;
 
 corrupted:
+	ext4_set_errno(inode->i_sb, -err);
 	ext4_error_inode(inode, function, line, 0,
 			 "pblk %llu bad header/extent: %s - magic %x, "
 			 "entries %u, max %u(%u), depth %u(%u)",
@@ -510,6 +468,30 @@ int ext4_ext_check_inode(struct inode *inode)
 	return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode), 0);
 }
 
+static void ext4_cache_extents(struct inode *inode,
+			       struct ext4_extent_header *eh)
+{
+	struct ext4_extent *ex = EXT_FIRST_EXTENT(eh);
+	ext4_lblk_t prev = 0;
+	int i;
+
+	for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) {
+		unsigned int status = EXTENT_STATUS_WRITTEN;
+		ext4_lblk_t lblk = le32_to_cpu(ex->ee_block);
+		int len = ext4_ext_get_actual_len(ex);
+
+		if (prev && (prev != lblk))
+			ext4_es_cache_extent(inode, prev, lblk - prev, ~0,
+					     EXTENT_STATUS_HOLE);
+
+		if (ext4_ext_is_unwritten(ex))
+			status = EXTENT_STATUS_UNWRITTEN;
+		ext4_es_cache_extent(inode, lblk, len,
+				     ext4_ext_pblock(ex), status);
+		prev = lblk + len;
+	}
+}
+
 static struct buffer_head *
 __read_extent_tree_block(const char *function, unsigned int line,
 			 struct inode *inode, ext4_fsblk_t pblk, int depth,
@@ -544,26 +526,7 @@ __read_extent_tree_block(const char *function, unsigned int line,
 	 */
 	if (!(flags & EXT4_EX_NOCACHE) && depth == 0) {
 		struct ext4_extent_header *eh = ext_block_hdr(bh);
-		struct ext4_extent *ex = EXT_FIRST_EXTENT(eh);
+		ext4_cache_extents(inode, eh);
-		ext4_lblk_t prev = 0;
-		int i;
-
-		for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) {
-			unsigned int status = EXTENT_STATUS_WRITTEN;
-			ext4_lblk_t lblk = le32_to_cpu(ex->ee_block);
-			int len = ext4_ext_get_actual_len(ex);
-
-			if (prev && (prev != lblk))
-				ext4_es_cache_extent(inode, prev,
-						     lblk - prev, ~0,
-						     EXTENT_STATUS_HOLE);
-
-			if (ext4_ext_is_unwritten(ex))
-				status = EXTENT_STATUS_UNWRITTEN;
-			ext4_es_cache_extent(inode, lblk, len,
-					     ext4_ext_pblock(ex), status);
-			prev = lblk + len;
-		}
 	}
 	return bh;
 errout:
@@ -649,8 +612,9 @@ static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path)
 	ext_debug("path:");
 	for (k = 0; k <= l; k++, path++) {
 		if (path->p_idx) {
-		  ext_debug("  %d->%llu", le32_to_cpu(path->p_idx->ei_block),
+			ext_debug("  %d->%llu",
-			    ext4_idx_pblock(path->p_idx));
+				  le32_to_cpu(path->p_idx->ei_block),
+				  ext4_idx_pblock(path->p_idx));
 		} else if (path->p_ext) {
 			ext_debug("  %d:[%d]%d:%llu ",
 				  le32_to_cpu(path->p_ext->ee_block),
@@ -731,11 +695,12 @@ void ext4_ext_drop_refs(struct ext4_ext_path *path)
 	if (!path)
 		return;
 	depth = path->p_depth;
-	for (i = 0; i <= depth; i++, path++)
+	for (i = 0; i <= depth; i++, path++) {
 		if (path->p_bh) {
 			brelse(path->p_bh);
 			path->p_bh = NULL;
 		}
+	}
 }
 
 /*
@@ -777,8 +742,8 @@ ext4_ext_binsearch_idx(struct inode *inode,
 
 		chix = ix = EXT_FIRST_INDEX(eh);
 		for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) {
-		  if (k != 0 &&
+			if (k != 0 && le32_to_cpu(ix->ei_block) <=
-		      le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) {
+			    le32_to_cpu(ix[-1].ei_block)) {
 				printk(KERN_DEBUG "k=%d, ix=0x%p, "
 				       "first=0x%p\n", k,
 				       ix, EXT_FIRST_INDEX(eh));
@@ -911,6 +876,8 @@ ext4_find_extent(struct inode *inode, ext4_lblk_t block,
 	path[0].p_bh = NULL;
 
 	i = depth;
+	if (!(flags & EXT4_EX_NOCACHE) && depth == 0)
+		ext4_cache_extents(inode, eh);
 	/* walk through the tree */
 	while (i) {
 		ext_debug("depth %d: num %d, max %d\n",
@@ -1632,17 +1599,16 @@ ext4_ext_next_allocated_block(struct ext4_ext_path *path)
 		return EXT_MAX_BLOCKS;
 
 	while (depth >= 0) {
+		struct ext4_ext_path *p = &path[depth];
+
 		if (depth == path->p_depth) {
 			/* leaf */
-			if (path[depth].p_ext &&
+			if (p->p_ext && p->p_ext != EXT_LAST_EXTENT(p->p_hdr))
-				path[depth].p_ext !=
+				return le32_to_cpu(p->p_ext[1].ee_block);
-					EXT_LAST_EXTENT(path[depth].p_hdr))
-			  return le32_to_cpu(path[depth].p_ext[1].ee_block);
 		} else {
 			/* index */
-			if (path[depth].p_idx !=
+			if (p->p_idx != EXT_LAST_INDEX(p->p_hdr))
-					EXT_LAST_INDEX(path[depth].p_hdr))
+				return le32_to_cpu(p->p_idx[1].ei_block);
-			  return le32_to_cpu(path[depth].p_idx[1].ei_block);
 		}
 		depth--;
 	}
@@ -1742,9 +1708,9 @@ static int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode,
 	return err;
 }
 
-int
+static int ext4_can_extents_be_merged(struct inode *inode,
-ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
+				      struct ext4_extent *ex1,
-				struct ext4_extent *ex2)
+				      struct ext4_extent *ex2)
 {
 	unsigned short ext1_ee_len, ext2_ee_len;
 
@@ -1758,11 +1724,6 @@ ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1,
 			le32_to_cpu(ex2->ee_block))
 		return 0;
 
-	/*
-	 * To allow future support for preallocated extents to be added
-	 * as an RO_COMPAT feature, refuse to merge to extents if
-	 * this can result in the top bit of ee_len being set.
-	 */
 	if (ext1_ee_len + ext2_ee_len > EXT_INIT_MAX_LEN)
 		return 0;
 
@@ -1870,13 +1831,14 @@ static void ext4_ext_try_to_merge_up(handle_t *handle,
 }
 
 /*
- * This function tries to merge the @ex extent to neighbours in the tree.
+ * This function tries to merge the @ex extent to neighbours in the tree, then
- * return 1 if merge left else 0.
+ * tries to collapse the extent tree into the inode.
  */
 static void ext4_ext_try_to_merge(handle_t *handle,
 				  struct inode *inode,
 				  struct ext4_ext_path *path,
-				  struct ext4_extent *ex) {
+				  struct ext4_extent *ex)
+{
 	struct ext4_extent_header *eh;
 	unsigned int depth;
 	int merge_done = 0;
@@ -3718,9 +3680,6 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 		max_zeroout = sbi->s_extent_max_zeroout_kb >>
 			(inode->i_sb->s_blocksize_bits - 10);
 
-	if (IS_ENCRYPTED(inode))
-		max_zeroout = 0;
-
 	/*
 	 * five cases:
 	 * 1. split the extent into three extents.
@@ -4706,6 +4665,10 @@ retry:
 	return ret > 0 ? ret2 : ret;
 }
 
+static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len);
+
+static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len);
+
 static long ext4_zero_range(struct file *file, loff_t offset,
 			    loff_t len, int mode)
 {
@@ -4723,9 +4686,6 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 
 	trace_ext4_zero_range(inode, offset, len, mode);
 
-	if (!S_ISREG(inode->i_mode))
-		return -EINVAL;
-
 	/* Call ext4_force_commit to flush all data in case of data=journal. */
 	if (ext4_should_journal_data(inode)) {
 		ret = ext4_force_commit(inode->i_sb);
@@ -4765,7 +4725,7 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 	}
 
 	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-	    (offset + len > i_size_read(inode) ||
+	    (offset + len > inode->i_size ||
 	     offset + len > EXT4_I(inode)->i_disksize)) {
 		new_size = offset + len;
 		ret = inode_newsize_ok(inode, new_size);
@@ -4849,7 +4809,7 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 		* Mark that we allocate beyond EOF so the subsequent truncate
 		* can proceed even if the new size is the same as i_size.
 		*/
-		if ((offset + len) > i_size_read(inode))
+		if (offset + len > inode->i_size)
 			ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
 	}
 	ext4_mark_inode_dirty(handle, inode);
@@ -4890,14 +4850,9 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 	 * range since we would need to re-encrypt blocks with a
 	 * different IV or XTS tweak (which are based on the logical
 	 * block number).
-	 *
-	 * XXX It's not clear why zero range isn't working, but we'll
-	 * leave it disabled for encrypted inodes for now.  This is a
-	 * bug we should fix....
 	 */
 	if (IS_ENCRYPTED(inode) &&
-	    (mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE |
+	    (mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE)))
-		     FALLOC_FL_ZERO_RANGE)))
 		return -EOPNOTSUPP;
 
 	/* Return error if mode is not supported */
@@ -4941,7 +4896,7 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len)
 	}
 
 	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-	    (offset + len > i_size_read(inode) ||
+	    (offset + len > inode->i_size ||
 	     offset + len > EXT4_I(inode)->i_disksize)) {
 		new_size = offset + len;
 		ret = inode_newsize_ok(inode, new_size);
@@ -5268,7 +5223,7 @@ ext4_ext_shift_path_extents(struct ext4_ext_path *path, ext4_lblk_t shift,
 {
 	int depth, err = 0;
 	struct ext4_extent *ex_start, *ex_last;
-	bool update = 0;
+	bool update = false;
 	depth = path->p_depth;
 
 	while (depth >= 0) {
@@ -5284,7 +5239,7 @@ ext4_ext_shift_path_extents(struct ext4_ext_path *path, ext4_lblk_t shift,
 				goto out;
 
 			if (ex_start == EXT_FIRST_EXTENT(path[depth].p_hdr))
-				update = 1;
+				update = true;
 
 			while (ex_start <= ex_last) {
 				if (SHIFT == SHIFT_LEFT) {
@@ -5472,7 +5427,7 @@ out:
  * This implements the fallocate's collapse range functionality for ext4
  * Returns: 0 and non-zero on error.
  */
-int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
+static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 {
 	struct super_block *sb = inode->i_sb;
 	ext4_lblk_t punch_start, punch_stop;
@@ -5489,12 +5444,8 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
 		return -EOPNOTSUPP;
 
-	/* Collapse range works only on fs block size aligned offsets. */
+	/* Collapse range works only on fs cluster size aligned regions. */
-	if (offset & (EXT4_CLUSTER_SIZE(sb) - 1) ||
+	if (!IS_ALIGNED(offset | len, EXT4_CLUSTER_SIZE(sb)))
-	    len & (EXT4_CLUSTER_SIZE(sb) - 1))
-		return -EINVAL;
-
-	if (!S_ISREG(inode->i_mode))
 		return -EINVAL;
 
 	trace_ext4_collapse_range(inode, offset, len);
@@ -5514,7 +5465,7 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 	 * There is no need to overlap collapse range with EOF, in which case
 	 * it is effectively a truncate operation
 	 */
-	if (offset + len >= i_size_read(inode)) {
+	if (offset + len >= inode->i_size) {
 		ret = -EINVAL;
 		goto out_mutex;
 	}
@@ -5592,7 +5543,7 @@ int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 		goto out_stop;
 	}
 
-	new_size = i_size_read(inode) - len;
+	new_size = inode->i_size - len;
 	i_size_write(inode, new_size);
 	EXT4_I(inode)->i_disksize = new_size;
 
@@ -5620,7 +5571,7 @@ out_mutex:
  * by len bytes.
  * Returns 0 on success, error otherwise.
  */
-int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
+static int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
 {
 	struct super_block *sb = inode->i_sb;
 	handle_t *handle;
@@ -5639,14 +5590,10 @@ int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
 	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
 		return -EOPNOTSUPP;
 
-	/* Insert range works only on fs block size aligned offsets. */
+	/* Insert range works only on fs cluster size aligned regions. */
-	if (offset & (EXT4_CLUSTER_SIZE(sb) - 1) ||
+	if (!IS_ALIGNED(offset | len, EXT4_CLUSTER_SIZE(sb)))
-			len & (EXT4_CLUSTER_SIZE(sb) - 1))
 		return -EINVAL;
 
-	if (!S_ISREG(inode->i_mode))
-		return -EOPNOTSUPP;
-
 	trace_ext4_insert_range(inode, offset, len);
 
 	offset_lblk = offset >> EXT4_BLOCK_SIZE_BITS(sb);
@@ -5666,14 +5613,14 @@ int ext4_insert_range(struct inode *inode, loff_t offset, loff_t len)
 		goto out_mutex;
 	}
 
-	/* Check for wrap through zero */
+	/* Check whether the maximum file size would be exceeded */
-	if (inode->i_size + len > inode->i_sb->s_maxbytes) {
+	if (len > inode->i_sb->s_maxbytes - inode->i_size) {
 		ret = -EFBIG;
 		goto out_mutex;
 	}
 
-	/* Offset should be less than i_size */
+	/* Offset must be less than i_size */
-	if (offset >= i_size_read(inode)) {
+	if (offset >= inode->i_size) {
 		ret = -EINVAL;
 		goto out_mutex;
 	}

fs/ext4/extents_status.h

@@ -209,6 +209,12 @@ static inline ext4_fsblk_t ext4_es_pblock(struct extent_status *es)
 	return es->es_pblk & ~ES_MASK;
 }
 
+static inline ext4_fsblk_t ext4_es_show_pblock(struct extent_status *es)
+{
+	ext4_fsblk_t pblock = ext4_es_pblock(es);
+	return pblock == ~ES_MASK ? 0 : pblock;
+}
+
 static inline void ext4_es_store_pblock(struct extent_status *es,
 					ext4_fsblk_t pb)
 {

fs/ext4/file.c

@@ -88,9 +88,10 @@ static ssize_t ext4_dax_read_iter(struct kiocb *iocb, struct iov_iter *to)
 	struct inode *inode = file_inode(iocb->ki_filp);
 	ssize_t ret;
 
-	if (!inode_trylock_shared(inode)) {
+	if (iocb->ki_flags & IOCB_NOWAIT) {
-		if (iocb->ki_flags & IOCB_NOWAIT)
+		if (!inode_trylock_shared(inode))
 			return -EAGAIN;
+	} else {
 		inode_lock_shared(inode);
 	}
 	/*
@@ -165,19 +166,25 @@ static int ext4_release_file(struct inode *inode, struct file *filp)
  * threads are at work on the same unwritten block, they must be synchronized
  * or one thread will zero the other's data, causing corruption.
  */
-static int
+static bool
-ext4_unaligned_aio(struct inode *inode, struct iov_iter *from, loff_t pos)
+ext4_unaligned_io(struct inode *inode, struct iov_iter *from, loff_t pos)
 {
 	struct super_block *sb = inode->i_sb;
-	int blockmask = sb->s_blocksize - 1;
+	unsigned long blockmask = sb->s_blocksize - 1;
-
-	if (pos >= ALIGN(i_size_read(inode), sb->s_blocksize))
-		return 0;
 
 	if ((pos | iov_iter_alignment(from)) & blockmask)
-		return 1;
+		return true;
 
-	return 0;
+	return false;
+}
+
+static bool
+ext4_extending_io(struct inode *inode, loff_t offset, size_t len)
+{
+	if (offset + len > i_size_read(inode) ||
+	    offset + len > EXT4_I(inode)->i_disksize)
+		return true;
+	return false;
 }
 
 /* Is IO overwriting allocated and initialized blocks? */
@@ -203,7 +210,8 @@ static bool ext4_overwrite_io(struct inode *inode, loff_t pos, loff_t len)
 	return err == blklen && (map.m_flags & EXT4_MAP_MAPPED);
 }
 
-static ssize_t ext4_write_checks(struct kiocb *iocb, struct iov_iter *from)
+static ssize_t ext4_generic_write_checks(struct kiocb *iocb,
+					 struct iov_iter *from)
 {
 	struct inode *inode = file_inode(iocb->ki_filp);
 	ssize_t ret;
@@ -227,11 +235,21 @@ static ssize_t ext4_write_checks(struct kiocb *iocb, struct iov_iter *from)
 		iov_iter_truncate(from, sbi->s_bitmap_maxbytes - iocb->ki_pos);
 	}
 
+	return iov_iter_count(from);
+}
+
+static ssize_t ext4_write_checks(struct kiocb *iocb, struct iov_iter *from)
+{
+	ssize_t ret, count;
+
+	count = ext4_generic_write_checks(iocb, from);
+	if (count <= 0)
+		return count;
+
 	ret = file_modified(iocb->ki_filp);
 	if (ret)
 		return ret;
-
+	return count;
-	return iov_iter_count(from);
 }
 
 static ssize_t ext4_buffered_write_iter(struct kiocb *iocb,
@@ -363,62 +381,137 @@ static const struct iomap_dio_ops ext4_dio_write_ops = {
 	.end_io = ext4_dio_write_end_io,
 };
 
+/*
+ * The intention here is to start with shared lock acquired then see if any
+ * condition requires an exclusive inode lock. If yes, then we restart the
+ * whole operation by releasing the shared lock and acquiring exclusive lock.
+ *
+ * - For unaligned_io we never take shared lock as it may cause data corruption
+ *   when two unaligned IO tries to modify the same block e.g. while zeroing.
+ *
+ * - 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.
+ */
+static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from,
+				     bool *ilock_shared, bool *extend)
+{
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+	loff_t offset;
+	size_t count;
+	ssize_t ret;
+
+restart:
+	ret = ext4_generic_write_checks(iocb, from);
+	if (ret <= 0)
+		goto out;
+
+	offset = iocb->ki_pos;
+	count = ret;
+	if (ext4_extending_io(inode, offset, count))
+		*extend = true;
+	/*
+	 * Determine whether the IO operation will overwrite allocated
+	 * and initialized blocks.
+	 * We need exclusive i_rwsem for changing security info
+	 * in file_modified().
+	 */
+	if (*ilock_shared && (!IS_NOSEC(inode) || *extend ||
+	     !ext4_overwrite_io(inode, offset, count))) {
+		inode_unlock_shared(inode);
+		*ilock_shared = false;
+		inode_lock(inode);
+		goto restart;
+	}
+
+	ret = file_modified(file);
+	if (ret < 0)
+		goto out;
+
+	return count;
+out:
+	if (*ilock_shared)
+		inode_unlock_shared(inode);
+	else
+		inode_unlock(inode);
+	return ret;
+}
+
 static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	ssize_t ret;
-	size_t count;
-	loff_t offset;
 	handle_t *handle;
 	struct inode *inode = file_inode(iocb->ki_filp);
-	bool extend = false, overwrite = false, unaligned_aio = false;
+	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 ilock_shared = true;
+
+	/*
+	 * We initially start with shared inode lock unless it is
+	 * unaligned IO which needs exclusive lock anyways.
+	 */
+	if (ext4_unaligned_io(inode, from, offset)) {
+		unaligned_io = true;
+		ilock_shared = false;
+	}
+	/*
+	 * Quick check here without any i_rwsem lock to see if it is extending
+	 * IO. A more reliable check is done in ext4_dio_write_checks() with
+	 * proper locking in place.
+	 */
+	if (offset + count > i_size_read(inode))
+		ilock_shared = false;
 
 	if (iocb->ki_flags & IOCB_NOWAIT) {
-		if (!inode_trylock(inode))
+		if (ilock_shared) {
-			return -EAGAIN;
+			if (!inode_trylock_shared(inode))
+				return -EAGAIN;
+		} else {
+			if (!inode_trylock(inode))
+				return -EAGAIN;
+		}
 	} else {
-		inode_lock(inode);
+		if (ilock_shared)
+			inode_lock_shared(inode);
+		else
+			inode_lock(inode);
 	}
 
+	/* Fallback to buffered I/O if the inode does not support direct I/O. */
 	if (!ext4_dio_supported(inode)) {
-		inode_unlock(inode);
+		if (ilock_shared)
-		/*
+			inode_unlock_shared(inode);
-		 * Fallback to buffered I/O if the inode does not support
+		else
-		 * direct I/O.
+			inode_unlock(inode);
-		 */
 		return ext4_buffered_write_iter(iocb, from);
 	}
 
-	ret = ext4_write_checks(iocb, from);
+	ret = ext4_dio_write_checks(iocb, from, &ilock_shared, &extend);
-	if (ret <= 0) {
+	if (ret <= 0)
-		inode_unlock(inode);
 		return ret;
-	}
 
-	/*
-	 * Unaligned asynchronous direct I/O must be serialized among each
-	 * other as the zeroing of partial blocks of two competing unaligned
-	 * asynchronous direct I/O writes can result in data corruption.
-	 */
 	offset = iocb->ki_pos;
-	count = iov_iter_count(from);
+	count = ret;
-	if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS) &&
-	    !is_sync_kiocb(iocb) && ext4_unaligned_aio(inode, from, offset)) {
-		unaligned_aio = true;
-		inode_dio_wait(inode);
-	}
 
 	/*
-	 * Determine whether the I/O will overwrite allocated and initialized
+	 * Unaligned direct IO must be serialized among each other as zeroing
-	 * blocks. If so, check to see whether it is possible to take the
+	 * of partial blocks of two competing unaligned IOs can result in data
-	 * dioread_nolock path.
+	 * corruption.
+	 *
+	 * So we make sure we don't allow any unaligned IO in flight.
+	 * For IOs where we need not wait (like unaligned non-AIO DIO),
+	 * below inode_dio_wait() may anyway become a no-op, since we start
+	 * with exclusive lock.
 	 */
-	if (!unaligned_aio && ext4_overwrite_io(inode, offset, count) &&
+	if (unaligned_io)
-	    ext4_should_dioread_nolock(inode)) {
+		inode_dio_wait(inode);
-		overwrite = true;
-		downgrade_write(&inode->i_rwsem);
-	}
 
-	if (offset + count > EXT4_I(inode)->i_disksize) {
+	if (extend) {
 		handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
 		if (IS_ERR(handle)) {
 			ret = PTR_ERR(handle);
@@ -431,18 +524,19 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 			goto out;
 		}
 
-		extend = true;
 		ext4_journal_stop(handle);
 	}
 
-	ret = iomap_dio_rw(iocb, from, &ext4_iomap_ops, &ext4_dio_write_ops,
+	if (ilock_shared)
-			   is_sync_kiocb(iocb) || unaligned_aio || extend);
+		iomap_ops = &ext4_iomap_overwrite_ops;
+	ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops,
+			   is_sync_kiocb(iocb) || unaligned_io || extend);
 
 	if (extend)
 		ret = ext4_handle_inode_extension(inode, offset, ret, count);
 
 out:
-	if (overwrite)
+	if (ilock_shared)
 		inode_unlock_shared(inode);
 	else
 		inode_unlock(inode);
@@ -487,9 +581,10 @@ ext4_dax_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	bool extend = false;
 	struct inode *inode = file_inode(iocb->ki_filp);
 
-	if (!inode_trylock(inode)) {
+	if (iocb->ki_flags & IOCB_NOWAIT) {
-		if (iocb->ki_flags & IOCB_NOWAIT)
+		if (!inode_trylock(inode))
 			return -EAGAIN;
+	} else {
 		inode_lock(inode);
 	}
 

fs/ext4/ialloc.c

@@ -94,7 +94,8 @@ static int ext4_validate_inode_bitmap(struct super_block *sb,
 		goto verified;
 	blk = ext4_inode_bitmap(sb, desc);
 	if (!ext4_inode_bitmap_csum_verify(sb, block_group, desc, bh,
-					   EXT4_INODES_PER_GROUP(sb) / 8)) {
+					   EXT4_INODES_PER_GROUP(sb) / 8) ||
+	    ext4_simulate_fail(sb, EXT4_SIM_IBITMAP_CRC)) {
 		ext4_unlock_group(sb, block_group);
 		ext4_error(sb, "Corrupt inode bitmap - block_group = %u, "
 			   "inode_bitmap = %llu", block_group, blk);
@@ -192,8 +193,10 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
 	get_bh(bh);
 	submit_bh(REQ_OP_READ, REQ_META | REQ_PRIO, bh);
 	wait_on_buffer(bh);
+	ext4_simulate_fail_bh(sb, bh, EXT4_SIM_IBITMAP_EIO);
 	if (!buffer_uptodate(bh)) {
 		put_bh(bh);
+		ext4_set_errno(sb, EIO);
 		ext4_error(sb, "Cannot read inode bitmap - "
 			   "block_group = %u, inode_bitmap = %llu",
 			   block_group, bitmap_blk);
@@ -1223,6 +1226,7 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
 	inode = ext4_iget(sb, ino, EXT4_IGET_NORMAL);
 	if (IS_ERR(inode)) {
 		err = PTR_ERR(inode);
+		ext4_set_errno(sb, -err);
 		ext4_error(sb, "couldn't read orphan inode %lu (err %d)",
 			   ino, err);
 		return inode;

fs/ext4/indirect.c

@@ -659,32 +659,6 @@ out:
 	return err;
 }
 
-/*
- * Calculate the number of metadata blocks need to reserve
- * to allocate a new block at @lblocks for non extent file based file
- */
-int ext4_ind_calc_metadata_amount(struct inode *inode, sector_t lblock)
-{
-	struct ext4_inode_info *ei = EXT4_I(inode);
-	sector_t dind_mask = ~((sector_t)EXT4_ADDR_PER_BLOCK(inode->i_sb) - 1);
-	int blk_bits;
-
-	if (lblock < EXT4_NDIR_BLOCKS)
-		return 0;
-
-	lblock -= EXT4_NDIR_BLOCKS;
-
-	if (ei->i_da_metadata_calc_len &&
-	    (lblock & dind_mask) == ei->i_da_metadata_calc_last_lblock) {
-		ei->i_da_metadata_calc_len++;
-		return 0;
-	}
-	ei->i_da_metadata_calc_last_lblock = lblock & dind_mask;
-	ei->i_da_metadata_calc_len = 1;
-	blk_bits = order_base_2(lblock);
-	return (blk_bits / EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb)) + 1;
-}
-
 /*
  * Calculate number of indirect blocks touched by mapping @nrblocks logically
  * contiguous blocks

fs/ext4/inline.c

@@ -98,6 +98,7 @@ int ext4_get_max_inline_size(struct inode *inode)
 
 	error = ext4_get_inode_loc(inode, &iloc);
 	if (error) {
+		ext4_set_errno(inode->i_sb, -error);
 		ext4_error_inode(inode, __func__, __LINE__, 0,
 				 "can't get inode location %lu",
 				 inode->i_ino);
@@ -849,7 +850,7 @@ out:
 
 /*
  * Prepare the write for the inline data.
- * If the the data can be written into the inode, we just read
+ * If the data can be written into the inode, we just read
  * the page and make it uptodate, and start the journal.
  * Otherwise read the page, makes it dirty so that it can be
  * handle in writepages(the i_disksize update is left to the
@@ -1761,6 +1762,7 @@ bool empty_inline_dir(struct inode *dir, int *has_inline_data)
 
 	err = ext4_get_inode_loc(dir, &iloc);
 	if (err) {
+		ext4_set_errno(dir->i_sb, -err);
 		EXT4_ERROR_INODE(dir, "error %d getting inode %lu block",
 				 err, dir->i_ino);
 		return true;

fs/ext4/inode.c

@@ -48,8 +48,6 @@
 
 #include <trace/events/ext4.h>
 
-#define MPAGE_DA_EXTENT_TAIL 0x01
-
 static __u32 ext4_inode_csum(struct inode *inode, struct ext4_inode *raw,
 			      struct ext4_inode_info *ei)
 {
@@ -271,6 +269,7 @@ void ext4_evict_inode(struct inode *inode)
 	if (inode->i_blocks) {
 		err = ext4_truncate(inode);
 		if (err) {
+			ext4_set_errno(inode->i_sb, -err);
 			ext4_error(inode->i_sb,
 				   "couldn't truncate inode %lu (err %d)",
 				   inode->i_ino, err);
@@ -402,7 +401,7 @@ int ext4_issue_zeroout(struct inode *inode, ext4_lblk_t lblk, ext4_fsblk_t pblk,
 {
 	int ret;
 
-	if (IS_ENCRYPTED(inode))
+	if (IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode))
 		return fscrypt_zeroout_range(inode, lblk, pblk, len);
 
 	ret = sb_issue_zeroout(inode->i_sb, pblk, len, GFP_NOFS);
@@ -2478,10 +2477,12 @@ update_disksize:
 			EXT4_I(inode)->i_disksize = disksize;
 		up_write(&EXT4_I(inode)->i_data_sem);
 		err2 = ext4_mark_inode_dirty(handle, inode);
-		if (err2)
+		if (err2) {
+			ext4_set_errno(inode->i_sb, -err2);
 			ext4_error(inode->i_sb,
 				   "Failed to mark inode %lu dirty",
 				   inode->i_ino);
+		}
 		if (!err)
 			err = err2;
 	}
@@ -3448,6 +3449,22 @@ static int ext4_iomap_begin(struct inode *inode, loff_t offset, loff_t length,
 	return 0;
 }
 
+static int ext4_iomap_overwrite_begin(struct inode *inode, loff_t offset,
+		loff_t length, unsigned flags, struct iomap *iomap,
+		struct iomap *srcmap)
+{
+	int ret;
+
+	/*
+	 * Even for writes we don't need to allocate blocks, so just pretend
+	 * we are reading to save overhead of starting a transaction.
+	 */
+	flags &= ~IOMAP_WRITE;
+	ret = ext4_iomap_begin(inode, offset, length, flags, iomap, srcmap);
+	WARN_ON_ONCE(iomap->type != IOMAP_MAPPED);
+	return ret;
+}
+
 static int ext4_iomap_end(struct inode *inode, loff_t offset, loff_t length,
 			  ssize_t written, unsigned flags, struct iomap *iomap)
 {
@@ -3469,6 +3486,11 @@ const struct iomap_ops ext4_iomap_ops = {
 	.iomap_end		= ext4_iomap_end,
 };
 
+const struct iomap_ops ext4_iomap_overwrite_ops = {
+	.iomap_begin		= ext4_iomap_overwrite_begin,
+	.iomap_end		= ext4_iomap_end,
+};
+
 static bool ext4_iomap_is_delalloc(struct inode *inode,
 				   struct ext4_map_blocks *map)
 {
@@ -3701,8 +3723,12 @@ static int __ext4_block_zero_page_range(handle_t *handle,
 		if (S_ISREG(inode->i_mode) && IS_ENCRYPTED(inode)) {
 			/* We expect the key to be set. */
 			BUG_ON(!fscrypt_has_encryption_key(inode));
-			WARN_ON_ONCE(fscrypt_decrypt_pagecache_blocks(
+			err = fscrypt_decrypt_pagecache_blocks(page, blocksize,
-					page, blocksize, bh_offset(bh)));
+							       bh_offset(bh));
+			if (err) {
+				clear_buffer_uptodate(bh);
+				goto unlock;
+			}
 		}
 	}
 	if (ext4_should_journal_data(inode)) {
@@ -3912,9 +3938,6 @@ int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length)
 	unsigned int credits;
 	int ret = 0;
 
-	if (!S_ISREG(inode->i_mode))
-		return -EOPNOTSUPP;
-
 	trace_ext4_punch_hole(inode, offset, length, 0);
 
 	ext4_clear_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA);
@@ -4240,6 +4263,8 @@ static int __ext4_get_inode_loc(struct inode *inode,
 	bh = sb_getblk(sb, block);
 	if (unlikely(!bh))
 		return -ENOMEM;
+	if (ext4_simulate_fail(sb, EXT4_SIM_INODE_EIO))
+		goto simulate_eio;
 	if (!buffer_uptodate(bh)) {
 		lock_buffer(bh);
 
@@ -4338,6 +4363,8 @@ make_io:
 		blk_finish_plug(&plug);
 		wait_on_buffer(bh);
 		if (!buffer_uptodate(bh)) {
+		simulate_eio:
+			ext4_set_errno(inode->i_sb, EIO);
 			EXT4_ERROR_INODE_BLOCK(inode, block,
 					       "unable to read itable block");
 			brelse(bh);
@@ -4551,7 +4578,9 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 					      sizeof(gen));
 	}
 
-	if (!ext4_inode_csum_verify(inode, raw_inode, ei)) {
+	if (!ext4_inode_csum_verify(inode, raw_inode, ei) ||
+	    ext4_simulate_fail(sb, EXT4_SIM_INODE_CRC)) {
+		ext4_set_errno(inode->i_sb, EFSBADCRC);
 		ext4_error_inode(inode, function, line, 0,
 				 "iget: checksum invalid");
 		ret = -EFSBADCRC;
@@ -5090,6 +5119,7 @@ int ext4_write_inode(struct inode *inode, struct writeback_control *wbc)
 		if (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync)
 			sync_dirty_buffer(iloc.bh);
 		if (buffer_req(iloc.bh) && !buffer_uptodate(iloc.bh)) {
+			ext4_set_errno(inode->i_sb, EIO);
 			EXT4_ERROR_INODE_BLOCK(inode, iloc.bh->b_blocknr,
 					 "IO error syncing inode");
 			err = -EIO;
@@ -5368,7 +5398,8 @@ int ext4_getattr(const struct path *path, struct kstat *stat,
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	unsigned int flags;
 
-	if (EXT4_FITS_IN_INODE(raw_inode, ei, i_crtime)) {
+	if ((request_mask & STATX_BTIME) &&
+	    EXT4_FITS_IN_INODE(raw_inode, ei, i_crtime)) {
 		stat->result_mask |= STATX_BTIME;
 		stat->btime.tv_sec = ei->i_crtime.tv_sec;
 		stat->btime.tv_nsec = ei->i_crtime.tv_nsec;

fs/ext4/ioctl.c

@@ -1377,6 +1377,8 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case EXT4_IOC_CLEAR_ES_CACHE:
 	case EXT4_IOC_GETSTATE:
 	case EXT4_IOC_GET_ES_CACHE:
+	case EXT4_IOC_FSGETXATTR:
+	case EXT4_IOC_FSSETXATTR:
 		break;
 	default:
 		return -ENOIOCTLCMD;

fs/ext4/mballoc.c

@@ -3895,6 +3895,7 @@ ext4_mb_discard_group_preallocations(struct super_block *sb,
 	bitmap_bh = ext4_read_block_bitmap(sb, group);
 	if (IS_ERR(bitmap_bh)) {
 		err = PTR_ERR(bitmap_bh);
+		ext4_set_errno(sb, -err);
 		ext4_error(sb, "Error %d reading block bitmap for %u",
 			   err, group);
 		return 0;
@@ -4063,6 +4064,7 @@ repeat:
 		err = ext4_mb_load_buddy_gfp(sb, group, &e4b,
 					     GFP_NOFS|__GFP_NOFAIL);
 		if (err) {
+			ext4_set_errno(sb, -err);
 			ext4_error(sb, "Error %d loading buddy information for %u",
 				   err, group);
 			continue;
@@ -4071,6 +4073,7 @@ repeat:
 		bitmap_bh = ext4_read_block_bitmap(sb, group);
 		if (IS_ERR(bitmap_bh)) {
 			err = PTR_ERR(bitmap_bh);
+			ext4_set_errno(sb, -err);
 			ext4_error(sb, "Error %d reading block bitmap for %u",
 					err, group);
 			ext4_mb_unload_buddy(&e4b);
@@ -4325,6 +4328,7 @@ ext4_mb_discard_lg_preallocations(struct super_block *sb,
 		err = ext4_mb_load_buddy_gfp(sb, group, &e4b,
 					     GFP_NOFS|__GFP_NOFAIL);
 		if (err) {
+			ext4_set_errno(sb, -err);
 			ext4_error(sb, "Error %d loading buddy information for %u",
 				   err, group);
 			continue;

fs/ext4/mmp.c

@@ -173,8 +173,10 @@ static int kmmpd(void *data)
 		 * (s_mmp_update_interval * 60) seconds.
 		 */
 		if (retval) {
-			if ((failed_writes % 60) == 0)
+			if ((failed_writes % 60) == 0) {
+				ext4_set_errno(sb, -retval);
 				ext4_error(sb, "Error writing to MMP block");
+			}
 			failed_writes++;
 		}
 
@@ -205,6 +207,7 @@ static int kmmpd(void *data)
 
 			retval = read_mmp_block(sb, &bh_check, mmp_block);
 			if (retval) {
+				ext4_set_errno(sb, -retval);
 				ext4_error(sb, "error reading MMP data: %d",
 					   retval);
 				goto exit_thread;
@@ -218,6 +221,7 @@ static int kmmpd(void *data)
 					     "Error while updating MMP info. "
 					     "The filesystem seems to have been"
 					     " multiply mounted.");
+				ext4_set_errno(sb, EBUSY);
 				ext4_error(sb, "abort");
 				put_bh(bh_check);
 				retval = -EBUSY;

fs/ext4/namei.c

@@ -109,7 +109,10 @@ static struct buffer_head *__ext4_read_dirblock(struct inode *inode,
 	struct ext4_dir_entry *dirent;
 	int is_dx_block = 0;
 
-	bh = ext4_bread(NULL, inode, block, 0);
+	if (ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_EIO))
+		bh = ERR_PTR(-EIO);
+	else
+		bh = ext4_bread(NULL, inode, block, 0);
 	if (IS_ERR(bh)) {
 		__ext4_warning(inode->i_sb, func, line,
 			       "inode #%lu: lblock %lu: comm %s: "
@@ -153,9 +156,11 @@ static struct buffer_head *__ext4_read_dirblock(struct inode *inode,
 	 * caller is sure it should be an index block.
 	 */
 	if (is_dx_block && type == INDEX) {
-		if (ext4_dx_csum_verify(inode, dirent))
+		if (ext4_dx_csum_verify(inode, dirent) &&
+		    !ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_CRC))
 			set_buffer_verified(bh);
 		else {
+			ext4_set_errno(inode->i_sb, EFSBADCRC);
 			ext4_error_inode(inode, func, line, block,
 					 "Directory index failed checksum");
 			brelse(bh);
@@ -163,9 +168,11 @@ static struct buffer_head *__ext4_read_dirblock(struct inode *inode,
 		}
 	}
 	if (!is_dx_block) {
-		if (ext4_dirblock_csum_verify(inode, bh))
+		if (ext4_dirblock_csum_verify(inode, bh) &&
+		    !ext4_simulate_fail(inode->i_sb, EXT4_SIM_DIRBLOCK_CRC))
 			set_buffer_verified(bh);
 		else {
+			ext4_set_errno(inode->i_sb, EFSBADCRC);
 			ext4_error_inode(inode, func, line, block,
 					 "Directory block failed checksum");
 			brelse(bh);
@@ -1002,7 +1009,6 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 	top = (struct ext4_dir_entry_2 *) ((char *) de +
 					   dir->i_sb->s_blocksize -
 					   EXT4_DIR_REC_LEN(0));
-#ifdef CONFIG_FS_ENCRYPTION
 	/* Check if the directory is encrypted */
 	if (IS_ENCRYPTED(dir)) {
 		err = fscrypt_get_encryption_info(dir);
@@ -1017,7 +1023,7 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 			return err;
 		}
 	}
-#endif
+
 	for (; de < top; de = ext4_next_entry(de, dir->i_sb->s_blocksize)) {
 		if (ext4_check_dir_entry(dir, NULL, de, bh,
 				bh->b_data, bh->b_size,
@@ -1065,9 +1071,7 @@ static int htree_dirblock_to_tree(struct file *dir_file,
 	}
 errout:
 	brelse(bh);
-#ifdef CONFIG_FS_ENCRYPTION
 	fscrypt_fname_free_buffer(&fname_crypto_str);
-#endif
 	return count;
 }
 
@@ -1527,6 +1531,7 @@ restart:
 			goto next;
 		wait_on_buffer(bh);
 		if (!buffer_uptodate(bh)) {
+			ext4_set_errno(sb, EIO);
 			EXT4_ERROR_INODE(dir, "reading directory lblock %lu",
 					 (unsigned long) block);
 			brelse(bh);
@@ -1537,6 +1542,7 @@ restart:
 		    !is_dx_internal_node(dir, block,
 					 (struct ext4_dir_entry *)bh->b_data) &&
 		    !ext4_dirblock_csum_verify(dir, bh)) {
+			ext4_set_errno(sb, EFSBADCRC);
 			EXT4_ERROR_INODE(dir, "checksumming directory "
 					 "block %lu", (unsigned long)block);
 			brelse(bh);

fs/ext4/page-io.c

@@ -512,17 +512,26 @@ int ext4_bio_write_page(struct ext4_io_submit *io,
 		gfp_t gfp_flags = GFP_NOFS;
 		unsigned int enc_bytes = round_up(len, i_blocksize(inode));
 
+		/*
+		 * Since bounce page allocation uses a mempool, we can only use
+		 * a waiting mask (i.e. request guaranteed allocation) on the
+		 * first page of the bio.  Otherwise it can deadlock.
+		 */
+		if (io->io_bio)
+			gfp_flags = GFP_NOWAIT | __GFP_NOWARN;
 	retry_encrypt:
 		bounce_page = fscrypt_encrypt_pagecache_blocks(page, enc_bytes,
 							       0, gfp_flags);
 		if (IS_ERR(bounce_page)) {
 			ret = PTR_ERR(bounce_page);
-			if (ret == -ENOMEM && wbc->sync_mode == WB_SYNC_ALL) {
+			if (ret == -ENOMEM &&
-				if (io->io_bio) {
+			    (io->io_bio || wbc->sync_mode == WB_SYNC_ALL)) {
+				gfp_flags = GFP_NOFS;
+				if (io->io_bio)
 					ext4_io_submit(io);
-					congestion_wait(BLK_RW_ASYNC, HZ/50);
+				else
-				}
+					gfp_flags |= __GFP_NOFAIL;
-				gfp_flags |= __GFP_NOFAIL;
+				congestion_wait(BLK_RW_ASYNC, HZ/50);
 				goto retry_encrypt;
 			}
 

fs/ext4/readpage.c

@@ -57,6 +57,7 @@ enum bio_post_read_step {
 	STEP_INITIAL = 0,
 	STEP_DECRYPT,
 	STEP_VERITY,
+	STEP_MAX,
 };
 
 struct bio_post_read_ctx {
@@ -106,10 +107,22 @@ static void verity_work(struct work_struct *work)
 {
 	struct bio_post_read_ctx *ctx =
 		container_of(work, struct bio_post_read_ctx, work);
+	struct bio *bio = ctx->bio;
 
-	fsverity_verify_bio(ctx->bio);
+	/*
+	 * fsverity_verify_bio() may call readpages() again, and although verity
+	 * will be disabled for that, decryption may still be needed, causing
+	 * another bio_post_read_ctx to be allocated.  So to guarantee that
+	 * mempool_alloc() never deadlocks we must free the current ctx first.
+	 * This is safe because verity is the last post-read step.
+	 */
+	BUILD_BUG_ON(STEP_VERITY + 1 != STEP_MAX);
+	mempool_free(ctx, bio_post_read_ctx_pool);
+	bio->bi_private = NULL;
 
-	bio_post_read_processing(ctx);
+	fsverity_verify_bio(bio);
+
+	__read_end_io(bio);
 }
 
 static void bio_post_read_processing(struct bio_post_read_ctx *ctx)
@@ -176,12 +189,11 @@ static inline bool ext4_need_verity(const struct inode *inode, pgoff_t idx)
 	       idx < DIV_ROUND_UP(inode->i_size, PAGE_SIZE);
 }
 
-static struct bio_post_read_ctx *get_bio_post_read_ctx(struct inode *inode,
+static void ext4_set_bio_post_read_ctx(struct bio *bio,
-						       struct bio *bio,
+				       const struct inode *inode,
-						       pgoff_t first_idx)
+				       pgoff_t first_idx)
 {
 	unsigned int post_read_steps = 0;
-	struct bio_post_read_ctx *ctx = NULL;
 
 	if (IS_ENCRYPTED(inode) && S_ISREG(inode->i_mode))
 		post_read_steps |= 1 << STEP_DECRYPT;
@@ -190,14 +202,14 @@ static struct bio_post_read_ctx *get_bio_post_read_ctx(struct inode *inode,
 		post_read_steps |= 1 << STEP_VERITY;
 
 	if (post_read_steps) {
-		ctx = mempool_alloc(bio_post_read_ctx_pool, GFP_NOFS);
+		/* Due to the mempool, this never fails. */
-		if (!ctx)
+		struct bio_post_read_ctx *ctx =
-			return ERR_PTR(-ENOMEM);
+			mempool_alloc(bio_post_read_ctx_pool, GFP_NOFS);
+
 		ctx->bio = bio;
 		ctx->enabled_steps = post_read_steps;
 		bio->bi_private = ctx;
 	}
-	return ctx;
 }
 
 static inline loff_t ext4_readpage_limit(struct inode *inode)
@@ -358,24 +370,16 @@ int ext4_mpage_readpages(struct address_space *mapping,
 			bio = NULL;
 		}
 		if (bio == NULL) {
-			struct bio_post_read_ctx *ctx;
-
 			/*
 			 * bio_alloc will _always_ be able to allocate a bio if
 			 * __GFP_DIRECT_RECLAIM is set, see bio_alloc_bioset().
 			 */
 			bio = bio_alloc(GFP_KERNEL,
 				min_t(int, nr_pages, BIO_MAX_PAGES));
-			ctx = get_bio_post_read_ctx(inode, bio, page->index);
+			ext4_set_bio_post_read_ctx(bio, inode, page->index);
-			if (IS_ERR(ctx)) {
-				bio_put(bio);
-				bio = NULL;
-				goto set_error_page;
-			}
 			bio_set_dev(bio, bdev);
 			bio->bi_iter.bi_sector = blocks[0] << (blkbits - 9);
 			bio->bi_end_io = mpage_end_io;
-			bio->bi_private = ctx;
 			bio_set_op_attrs(bio, REQ_OP_READ,
 						is_readahead ? REQ_RAHEAD : 0);
 		}

fs/ext4/resize.c

@@ -824,9 +824,8 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	if (unlikely(err))
 		goto errout;
 
-	n_group_desc = ext4_kvmalloc((gdb_num + 1) *
+	n_group_desc = kvmalloc((gdb_num + 1) * sizeof(struct buffer_head *),
-				     sizeof(struct buffer_head *),
+				GFP_KERNEL);
-				     GFP_NOFS);
 	if (!n_group_desc) {
 		err = -ENOMEM;
 		ext4_warning(sb, "not enough memory for %lu groups",
@@ -900,9 +899,8 @@ static int add_new_gdb_meta_bg(struct super_block *sb,
 	gdb_bh = ext4_sb_bread(sb, gdblock, 0);
 	if (IS_ERR(gdb_bh))
 		return PTR_ERR(gdb_bh);
-	n_group_desc = ext4_kvmalloc((gdb_num + 1) *
+	n_group_desc = kvmalloc((gdb_num + 1) * sizeof(struct buffer_head *),
-				     sizeof(struct buffer_head *),
+				GFP_KERNEL);
-				     GFP_NOFS);
 	if (!n_group_desc) {
 		brelse(gdb_bh);
 		err = -ENOMEM;

fs/ext4/super.c

@@ -154,7 +154,7 @@ ext4_sb_bread(struct super_block *sb, sector_t block, int op_flags)
 
 	if (bh == NULL)
 		return ERR_PTR(-ENOMEM);
-	if (buffer_uptodate(bh))
+	if (ext4_buffer_uptodate(bh))
 		return bh;
 	ll_rw_block(REQ_OP_READ, REQ_META | op_flags, 1, &bh);
 	wait_on_buffer(bh);
@@ -204,26 +204,6 @@ void ext4_superblock_csum_set(struct super_block *sb)
 	es->s_checksum = ext4_superblock_csum(sb, es);
 }
 
-void *ext4_kvmalloc(size_t size, gfp_t flags)
-{
-	void *ret;
-
-	ret = kmalloc(size, flags | __GFP_NOWARN);
-	if (!ret)
-		ret = __vmalloc(size, flags, PAGE_KERNEL);
-	return ret;
-}
-
-void *ext4_kvzalloc(size_t size, gfp_t flags)
-{
-	void *ret;
-
-	ret = kzalloc(size, flags | __GFP_NOWARN);
-	if (!ret)
-		ret = __vmalloc(size, flags | __GFP_ZERO, PAGE_KERNEL);
-	return ret;
-}
-
 ext4_fsblk_t ext4_block_bitmap(struct super_block *sb,
 			       struct ext4_group_desc *bg)
 {
@@ -367,6 +347,8 @@ static void __save_error_info(struct super_block *sb, const char *func,
 	ext4_update_tstamp(es, s_last_error_time);
 	strncpy(es->s_last_error_func, func, sizeof(es->s_last_error_func));
 	es->s_last_error_line = cpu_to_le32(line);
+	if (es->s_last_error_errcode == 0)
+		es->s_last_error_errcode = EXT4_ERR_EFSCORRUPTED;
 	if (!es->s_first_error_time) {
 		es->s_first_error_time = es->s_last_error_time;
 		es->s_first_error_time_hi = es->s_last_error_time_hi;
@@ -375,6 +357,7 @@ static void __save_error_info(struct super_block *sb, const char *func,
 		es->s_first_error_line = cpu_to_le32(line);
 		es->s_first_error_ino = es->s_last_error_ino;
 		es->s_first_error_block = es->s_last_error_block;
+		es->s_first_error_errcode = es->s_last_error_errcode;
 	}
 	/*
 	 * Start the daily error reporting function if it hasn't been
@@ -631,6 +614,66 @@ const char *ext4_decode_error(struct super_block *sb, int errno,
 	return errstr;
 }
 
+void ext4_set_errno(struct super_block *sb, int err)
+{
+	if (err < 0)
+		err = -err;
+
+	switch (err) {
+	case EIO:
+		err = EXT4_ERR_EIO;
+		break;
+	case ENOMEM:
+		err = EXT4_ERR_ENOMEM;
+		break;
+	case EFSBADCRC:
+		err = EXT4_ERR_EFSBADCRC;
+		break;
+	case EFSCORRUPTED:
+		err = EXT4_ERR_EFSCORRUPTED;
+		break;
+	case ENOSPC:
+		err = EXT4_ERR_ENOSPC;
+		break;
+	case ENOKEY:
+		err = EXT4_ERR_ENOKEY;
+		break;
+	case EROFS:
+		err = EXT4_ERR_EROFS;
+		break;
+	case EFBIG:
+		err = EXT4_ERR_EFBIG;
+		break;
+	case EEXIST:
+		err = EXT4_ERR_EEXIST;
+		break;
+	case ERANGE:
+		err = EXT4_ERR_ERANGE;
+		break;
+	case EOVERFLOW:
+		err = EXT4_ERR_EOVERFLOW;
+		break;
+	case EBUSY:
+		err = EXT4_ERR_EBUSY;
+		break;
+	case ENOTDIR:
+		err = EXT4_ERR_ENOTDIR;
+		break;
+	case ENOTEMPTY:
+		err = EXT4_ERR_ENOTEMPTY;
+		break;
+	case ESHUTDOWN:
+		err = EXT4_ERR_ESHUTDOWN;
+		break;
+	case EFAULT:
+		err = EXT4_ERR_EFAULT;
+		break;
+	default:
+		err = EXT4_ERR_UNKNOWN;
+	}
+	EXT4_SB(sb)->s_es->s_last_error_errcode = err;
+}
+
 /* __ext4_std_error decodes expected errors from journaling functions
  * automatically and invokes the appropriate error response.  */
 
@@ -655,6 +698,7 @@ void __ext4_std_error(struct super_block *sb, const char *function,
 		       sb->s_id, function, line, errstr);
 	}
 
+	ext4_set_errno(sb, -errno);
 	save_error_info(sb, function, line);
 	ext4_handle_error(sb);
 }
@@ -982,8 +1026,10 @@ static void ext4_put_super(struct super_block *sb)
 		aborted = is_journal_aborted(sbi->s_journal);
 		err = jbd2_journal_destroy(sbi->s_journal);
 		sbi->s_journal = NULL;
-		if ((err < 0) && !aborted)
+		if ((err < 0) && !aborted) {
+			ext4_set_errno(sb, -err);
 			ext4_abort(sb, "Couldn't clean up the journal");
+		}
 	}
 
 	ext4_unregister_sysfs(sb);
@@ -1085,8 +1131,6 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
 	ei->i_es_shk_nr = 0;
 	ei->i_es_shrink_lblk = 0;
 	ei->i_reserved_data_blocks = 0;
-	ei->i_da_metadata_calc_len = 0;
-	ei->i_da_metadata_calc_last_lblock = 0;
 	spin_lock_init(&(ei->i_block_reservation_lock));
 	ext4_init_pending_tree(&ei->i_pending_tree);
 #ifdef CONFIG_QUOTA
@@ -1548,6 +1592,7 @@ static const match_table_t tokens = {
 	{Opt_auto_da_alloc, "auto_da_alloc"},
 	{Opt_noauto_da_alloc, "noauto_da_alloc"},
 	{Opt_dioread_nolock, "dioread_nolock"},
+	{Opt_dioread_lock, "nodioread_nolock"},
 	{Opt_dioread_lock, "dioread_lock"},
 	{Opt_discard, "discard"},
 	{Opt_nodiscard, "nodiscard"},
@@ -3720,6 +3765,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		set_opt(sb, NO_UID32);
 	/* xattr user namespace & acls are now defaulted on */
 	set_opt(sb, XATTR_USER);
+	set_opt(sb, DIOREAD_NOLOCK);
 #ifdef CONFIG_EXT4_FS_POSIX_ACL
 	set_opt(sb, POSIX_ACL);
 #endif
@@ -3887,9 +3933,8 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 #endif
 
 	if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) {
-		printk_once(KERN_WARNING "EXT4-fs: Warning: mounting "
+		printk_once(KERN_WARNING "EXT4-fs: Warning: mounting with data=journal disables delayed allocation, dioread_nolock, and O_DIRECT support!\n");
-			    "with data=journal disables delayed "
+		clear_opt(sb, DIOREAD_NOLOCK);
-			    "allocation and O_DIRECT support!\n");
 		if (test_opt2(sb, EXPLICIT_DELALLOC)) {
 			ext4_msg(sb, KERN_ERR, "can't mount with "
 				 "both data=journal and delalloc");
@@ -5540,9 +5585,15 @@ static int ext4_statfs_project(struct super_block *sb,
 		return PTR_ERR(dquot);
 	spin_lock(&dquot->dq_dqb_lock);
 
-	limit = (dquot->dq_dqb.dqb_bsoftlimit ?
+	limit = 0;
-		 dquot->dq_dqb.dqb_bsoftlimit :
+	if (dquot->dq_dqb.dqb_bsoftlimit &&
-		 dquot->dq_dqb.dqb_bhardlimit) >> sb->s_blocksize_bits;
+	    (!limit || dquot->dq_dqb.dqb_bsoftlimit < limit))
+		limit = dquot->dq_dqb.dqb_bsoftlimit;
+	if (dquot->dq_dqb.dqb_bhardlimit &&
+	    (!limit || dquot->dq_dqb.dqb_bhardlimit < limit))
+		limit = dquot->dq_dqb.dqb_bhardlimit;
+	limit >>= sb->s_blocksize_bits;
+
 	if (limit && buf->f_blocks > limit) {
 		curblock = (dquot->dq_dqb.dqb_curspace +
 			    dquot->dq_dqb.dqb_rsvspace) >> sb->s_blocksize_bits;
@@ -5552,9 +5603,14 @@ static int ext4_statfs_project(struct super_block *sb,
 			 (buf->f_blocks - curblock) : 0;
 	}
 
-	limit = dquot->dq_dqb.dqb_isoftlimit ?
+	limit = 0;
-		dquot->dq_dqb.dqb_isoftlimit :
+	if (dquot->dq_dqb.dqb_isoftlimit &&
-		dquot->dq_dqb.dqb_ihardlimit;
+	    (!limit || dquot->dq_dqb.dqb_isoftlimit < limit))
+		limit = dquot->dq_dqb.dqb_isoftlimit;
+	if (dquot->dq_dqb.dqb_ihardlimit &&
+	    (!limit || dquot->dq_dqb.dqb_ihardlimit < limit))
+		limit = dquot->dq_dqb.dqb_ihardlimit;
+
 	if (limit && buf->f_files > limit) {
 		buf->f_files = limit;
 		buf->f_ffree =

fs/ext4/sysfs.c

@@ -29,6 +29,10 @@ typedef enum {
 	attr_last_error_time,
 	attr_feature,
 	attr_pointer_ui,
+	attr_pointer_ul,
+	attr_pointer_u64,
+	attr_pointer_u8,
+	attr_pointer_string,
 	attr_pointer_atomic,
 	attr_journal_task,
 } attr_id_t;
@@ -46,6 +50,7 @@ struct ext4_attr {
 	struct attribute attr;
 	short attr_id;
 	short attr_ptr;
+	unsigned short attr_size;
 	union {
 		int offset;
 		void *explicit_ptr;
@@ -154,12 +159,35 @@ static struct ext4_attr ext4_attr_##_name = {			\
 	},							\
 }
 
+#define EXT4_ATTR_STRING(_name,_mode,_size,_struct,_elname)	\
+static struct ext4_attr ext4_attr_##_name = {			\
+	.attr = {.name = __stringify(_name), .mode = _mode },	\
+	.attr_id = attr_pointer_string,				\
+	.attr_size = _size,					\
+	.attr_ptr = ptr_##_struct##_offset,			\
+	.u = {							\
+		.offset = offsetof(struct _struct, _elname),\
+	},							\
+}
+
 #define EXT4_RO_ATTR_ES_UI(_name,_elname)				\
 	EXT4_ATTR_OFFSET(_name, 0444, pointer_ui, ext4_super_block, _elname)
 
+#define EXT4_RO_ATTR_ES_U8(_name,_elname)				\
+	EXT4_ATTR_OFFSET(_name, 0444, pointer_u8, ext4_super_block, _elname)
+
+#define EXT4_RO_ATTR_ES_U64(_name,_elname)				\
+	EXT4_ATTR_OFFSET(_name, 0444, pointer_u64, ext4_super_block, _elname)
+
+#define EXT4_RO_ATTR_ES_STRING(_name,_elname,_size)			\
+	EXT4_ATTR_STRING(_name, 0444, _size, ext4_super_block, _elname)
+
 #define EXT4_RW_ATTR_SBI_UI(_name,_elname)	\
 	EXT4_ATTR_OFFSET(_name, 0644, pointer_ui, ext4_sb_info, _elname)
 
+#define EXT4_RW_ATTR_SBI_UL(_name,_elname)	\
+	EXT4_ATTR_OFFSET(_name, 0644, pointer_ul, ext4_sb_info, _elname)
+
 #define EXT4_ATTR_PTR(_name,_mode,_id,_ptr) \
 static struct ext4_attr ext4_attr_##_name = {			\
 	.attr = {.name = __stringify(_name), .mode = _mode },	\
@@ -194,7 +222,20 @@ EXT4_RW_ATTR_SBI_UI(warning_ratelimit_interval_ms, s_warning_ratelimit_state.int
 EXT4_RW_ATTR_SBI_UI(warning_ratelimit_burst, s_warning_ratelimit_state.burst);
 EXT4_RW_ATTR_SBI_UI(msg_ratelimit_interval_ms, s_msg_ratelimit_state.interval);
 EXT4_RW_ATTR_SBI_UI(msg_ratelimit_burst, s_msg_ratelimit_state.burst);
+#ifdef CONFIG_EXT4_DEBUG
+EXT4_RW_ATTR_SBI_UL(simulate_fail, s_simulate_fail);
+#endif
 EXT4_RO_ATTR_ES_UI(errors_count, s_error_count);
+EXT4_RO_ATTR_ES_U8(first_error_errcode, s_first_error_errcode);
+EXT4_RO_ATTR_ES_U8(last_error_errcode, s_last_error_errcode);
+EXT4_RO_ATTR_ES_UI(first_error_ino, s_first_error_ino);
+EXT4_RO_ATTR_ES_UI(last_error_ino, s_last_error_ino);
+EXT4_RO_ATTR_ES_U64(first_error_block, s_first_error_block);
+EXT4_RO_ATTR_ES_U64(last_error_block, s_last_error_block);
+EXT4_RO_ATTR_ES_UI(first_error_line, s_first_error_line);
+EXT4_RO_ATTR_ES_UI(last_error_line, s_last_error_line);
+EXT4_RO_ATTR_ES_STRING(first_error_func, s_first_error_func, 32);
+EXT4_RO_ATTR_ES_STRING(last_error_func, s_last_error_func, 32);
 EXT4_ATTR(first_error_time, 0444, first_error_time);
 EXT4_ATTR(last_error_time, 0444, last_error_time);
 EXT4_ATTR(journal_task, 0444, journal_task);
@@ -225,9 +266,22 @@ static struct attribute *ext4_attrs[] = {
 	ATTR_LIST(msg_ratelimit_interval_ms),
 	ATTR_LIST(msg_ratelimit_burst),
 	ATTR_LIST(errors_count),
+	ATTR_LIST(first_error_ino),
+	ATTR_LIST(last_error_ino),
+	ATTR_LIST(first_error_block),
+	ATTR_LIST(last_error_block),
+	ATTR_LIST(first_error_line),
+	ATTR_LIST(last_error_line),
+	ATTR_LIST(first_error_func),
+	ATTR_LIST(last_error_func),
+	ATTR_LIST(first_error_errcode),
+	ATTR_LIST(last_error_errcode),
 	ATTR_LIST(first_error_time),
 	ATTR_LIST(last_error_time),
 	ATTR_LIST(journal_task),
+#ifdef CONFIG_EXT4_DEBUG
+	ATTR_LIST(simulate_fail),
+#endif
 	NULL,
 };
 ATTRIBUTE_GROUPS(ext4);
@@ -280,7 +334,7 @@ static void *calc_ptr(struct ext4_attr *a, struct ext4_sb_info *sbi)
 
 static ssize_t __print_tstamp(char *buf, __le32 lo, __u8 hi)
 {
-	return snprintf(buf, PAGE_SIZE, "%lld",
+	return snprintf(buf, PAGE_SIZE, "%lld\n",
 			((time64_t)hi << 32) + le32_to_cpu(lo));
 }
 
@@ -318,6 +372,30 @@ static ssize_t ext4_attr_show(struct kobject *kobj,
 		else
 			return snprintf(buf, PAGE_SIZE, "%u\n",
 					*((unsigned int *) ptr));
+	case attr_pointer_ul:
+		if (!ptr)
+			return 0;
+		return snprintf(buf, PAGE_SIZE, "%lu\n",
+				*((unsigned long *) ptr));
+	case attr_pointer_u8:
+		if (!ptr)
+			return 0;
+		return snprintf(buf, PAGE_SIZE, "%u\n",
+				*((unsigned char *) ptr));
+	case attr_pointer_u64:
+		if (!ptr)
+			return 0;
+		if (a->attr_ptr == ptr_ext4_super_block_offset)
+			return snprintf(buf, PAGE_SIZE, "%llu\n",
+					le64_to_cpup(ptr));
+		else
+			return snprintf(buf, PAGE_SIZE, "%llu\n",
+					*((unsigned long long *) ptr));
+	case attr_pointer_string:
+		if (!ptr)
+			return 0;
+		return snprintf(buf, PAGE_SIZE, "%.*s\n", a->attr_size,
+				(char *) ptr);
 	case attr_pointer_atomic:
 		if (!ptr)
 			return 0;
@@ -361,6 +439,14 @@ static ssize_t ext4_attr_store(struct kobject *kobj,
 		else
 			*((unsigned int *) ptr) = t;
 		return len;
+	case attr_pointer_ul:
+		if (!ptr)
+			return 0;
+		ret = kstrtoul(skip_spaces(buf), 0, &t);
+		if (ret)
+			return ret;
+		*((unsigned long *) ptr) = t;
+		return len;
 	case attr_inode_readahead:
 		return inode_readahead_blks_store(sbi, buf, len);
 	case attr_trigger_test_error:

fs/ext4/xattr.c

@@ -1456,7 +1456,7 @@ ext4_xattr_inode_cache_find(struct inode *inode, const void *value,
 	if (!ce)
 		return NULL;
 
-	ea_data = ext4_kvmalloc(value_len, GFP_NOFS);
+	ea_data = kvmalloc(value_len, GFP_KERNEL);
 	if (!ea_data) {
 		mb_cache_entry_put(ea_inode_cache, ce);
 		return NULL;
@@ -2879,9 +2879,11 @@ int ext4_xattr_delete_inode(handle_t *handle, struct inode *inode,
 		bh = ext4_sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl, REQ_PRIO);
 		if (IS_ERR(bh)) {
 			error = PTR_ERR(bh);
-			if (error == -EIO)
+			if (error == -EIO) {
+				ext4_set_errno(inode->i_sb, EIO);
 				EXT4_ERROR_INODE(inode, "block %llu read error",
 						 EXT4_I(inode)->i_file_acl);
+			}
 			bh = NULL;
 			goto cleanup;
 		}

fs/jbd2/checkpoint.c

@@ -164,7 +164,7 @@ void __jbd2_log_wait_for_space(journal_t *journal)
 				       "journal space in %s\n", __func__,
 				       journal->j_devname);
 				WARN_ON(1);
-				jbd2_journal_abort(journal, 0);
+				jbd2_journal_abort(journal, -EIO);
 			}
 			write_lock(&journal->j_state_lock);
 		} else {

fs/jbd2/commit.c

@@ -782,7 +782,7 @@ start_journal_io:
 		err = journal_submit_commit_record(journal, commit_transaction,
 						 &cbh, crc32_sum);
 		if (err)
-			__jbd2_journal_abort_hard(journal);
+			jbd2_journal_abort(journal, err);
 	}
 
 	blk_finish_plug(&plug);
@@ -875,7 +875,7 @@ start_journal_io:
 		err = journal_submit_commit_record(journal, commit_transaction,
 						&cbh, crc32_sum);
 		if (err)
-			__jbd2_journal_abort_hard(journal);
+			jbd2_journal_abort(journal, err);
 	}
 	if (cbh)
 		err = journal_wait_on_commit_record(journal, cbh);

fs/jbd2/journal.c

@@ -96,7 +96,6 @@ EXPORT_SYMBOL(jbd2_journal_release_jbd_inode);
 EXPORT_SYMBOL(jbd2_journal_begin_ordered_truncate);
 EXPORT_SYMBOL(jbd2_inode_cache);
 
-static void __journal_abort_soft (journal_t *journal, int errno);
 static int jbd2_journal_create_slab(size_t slab_size);
 
 #ifdef CONFIG_JBD2_DEBUG
@@ -805,7 +804,7 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr,
 					"at offset %lu on %s\n",
 			       __func__, blocknr, journal->j_devname);
 			err = -EIO;
-			__journal_abort_soft(journal, err);
+			jbd2_journal_abort(journal, err);
 		}
 	} else {
 		*retp = blocknr; /* +journal->j_blk_offset */
@@ -982,6 +981,7 @@ static void *jbd2_seq_info_start(struct seq_file *seq, loff_t *pos)
 
 static void *jbd2_seq_info_next(struct seq_file *seq, void *v, loff_t *pos)
 {
+	(*pos)++;
 	return NULL;
 }
 
@@ -1710,6 +1710,11 @@ int jbd2_journal_load(journal_t *journal)
 		       journal->j_devname);
 		return -EFSCORRUPTED;
 	}
+	/*
+	 * clear JBD2_ABORT flag initialized in journal_init_common
+	 * here to update log tail information with the newest seq.
+	 */
+	journal->j_flags &= ~JBD2_ABORT;
 
 	/* OK, we've finished with the dynamic journal bits:
 	 * reinitialise the dynamic contents of the superblock in memory
@@ -1717,7 +1722,6 @@ int jbd2_journal_load(journal_t *journal)
 	if (journal_reset(journal))
 		goto recovery_error;
 
-	journal->j_flags &= ~JBD2_ABORT;
 	journal->j_flags |= JBD2_LOADED;
 	return 0;
 
@@ -2098,67 +2102,6 @@ int jbd2_journal_wipe(journal_t *journal, int write)
 	return err;
 }
 
-/*
- * Journal abort has very specific semantics, which we describe
- * for journal abort.
- *
- * Two internal functions, which provide abort to the jbd layer
- * itself are here.
- */
-
-/*
- * Quick version for internal journal use (doesn't lock the journal).
- * Aborts hard --- we mark the abort as occurred, but do _nothing_ else,
- * and don't attempt to make any other journal updates.
- */
-void __jbd2_journal_abort_hard(journal_t *journal)
-{
-	transaction_t *transaction;
-
-	if (journal->j_flags & JBD2_ABORT)
-		return;
-
-	printk(KERN_ERR "Aborting journal on device %s.\n",
-	       journal->j_devname);
-
-	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);
-	write_unlock(&journal->j_state_lock);
-}
-
-/* Soft abort: record the abort error status in the journal superblock,
- * but don't do any other IO. */
-static void __journal_abort_soft (journal_t *journal, int errno)
-{
-	int old_errno;
-
-	write_lock(&journal->j_state_lock);
-	old_errno = journal->j_errno;
-	if (!journal->j_errno || errno == -ESHUTDOWN)
-		journal->j_errno = errno;
-
-	if (journal->j_flags & JBD2_ABORT) {
-		write_unlock(&journal->j_state_lock);
-		if (!old_errno && old_errno != -ESHUTDOWN &&
-		    errno == -ESHUTDOWN)
-			jbd2_journal_update_sb_errno(journal);
-		return;
-	}
-	write_unlock(&journal->j_state_lock);
-
-	__jbd2_journal_abort_hard(journal);
-
-	if (errno) {
-		jbd2_journal_update_sb_errno(journal);
-		write_lock(&journal->j_state_lock);
-		journal->j_flags |= JBD2_REC_ERR;
-		write_unlock(&journal->j_state_lock);
-	}
-}
-
 /**
  * void jbd2_journal_abort () - Shutdown the journal immediately.
  * @journal: the journal to shutdown.
@@ -2198,16 +2141,51 @@ static void __journal_abort_soft (journal_t *journal, int errno)
  * failure to disk.  ext3_error, for example, now uses this
  * functionality.
  *
- * Errors which originate from within the journaling layer will NOT
- * supply an errno; a null errno implies that absolutely no further
- * writes are done to the journal (unless there are any already in
- * progress).
- *
  */
 
 void jbd2_journal_abort(journal_t *journal, int errno)
 {
-	__journal_abort_soft(journal, errno);
+	transaction_t *transaction;
+
+	/*
+	 * ESHUTDOWN always takes precedence because a file system check
+	 * caused by any other journal abort error is not required after
+	 * a shutdown triggered.
+	 */
+	write_lock(&journal->j_state_lock);
+	if (journal->j_flags & JBD2_ABORT) {
+		int old_errno = journal->j_errno;
+
+		write_unlock(&journal->j_state_lock);
+		if (old_errno != -ESHUTDOWN && errno == -ESHUTDOWN) {
+			journal->j_errno = errno;
+			jbd2_journal_update_sb_errno(journal);
+		}
+		return;
+	}
+
+	/*
+	 * Mark the abort as occurred and start current running transaction
+	 * to release all journaled buffer.
+	 */
+	pr_err("Aborting journal on device %s.\n", journal->j_devname);
+
+	journal->j_flags |= JBD2_ABORT;
+	journal->j_errno = errno;
+	transaction = journal->j_running_transaction;
+	if (transaction)
+		__jbd2_log_start_commit(journal, transaction->t_tid);
+	write_unlock(&journal->j_state_lock);
+
+	/*
+	 * Record errno to the journal super block, so that fsck and jbd2
+	 * layer could realise that a filesystem check is needed.
+	 */
+	jbd2_journal_update_sb_errno(journal);
+
+	write_lock(&journal->j_state_lock);
+	journal->j_flags |= JBD2_REC_ERR;
+	write_unlock(&journal->j_state_lock);
 }
 
 /**
@@ -2556,7 +2534,6 @@ static void __journal_remove_journal_head(struct buffer_head *bh)
 {
 	struct journal_head *jh = bh2jh(bh);
 
-	J_ASSERT_JH(jh, jh->b_jcount >= 0);
 	J_ASSERT_JH(jh, jh->b_transaction == NULL);
 	J_ASSERT_JH(jh, jh->b_next_transaction == NULL);
 	J_ASSERT_JH(jh, jh->b_cp_transaction == NULL);

fs/jbd2/transaction.c

@@ -525,7 +525,7 @@ EXPORT_SYMBOL(jbd2__journal_start);
  * modified buffers in the log.  We block until the log can guarantee
  * that much space. Additionally, if rsv_blocks > 0, we also create another
  * handle with rsv_blocks reserved blocks in the journal. This handle is
- * is stored in h_rsv_handle. It is not attached to any particular transaction
+ * stored in h_rsv_handle. It is not attached to any particular transaction
  * and thus doesn't block transaction commit. If the caller uses this reserved
  * handle, it has to set h_rsv_handle to NULL as otherwise jbd2_journal_stop()
  * on the parent handle will dispose the reserved one. Reserved handle has to
@@ -1595,7 +1595,7 @@ out:
  * Allow this call even if the handle has aborted --- it may be part of
  * the caller's cleanup after an abort.
  */
-int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
+int jbd2_journal_forget(handle_t *handle, struct buffer_head *bh)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal;

include/linux/jbd2.h

@@ -1403,7 +1403,6 @@ extern int	   jbd2_journal_skip_recovery	(journal_t *);
 extern void	   jbd2_journal_update_sb_errno(journal_t *);
 extern int	   jbd2_journal_update_sb_log_tail	(journal_t *, tid_t,
 				unsigned long, int);
-extern void	   __jbd2_journal_abort_hard	(journal_t *);
 extern void	   jbd2_journal_abort      (journal_t *, int);
 extern int	   jbd2_journal_errno      (journal_t *);
 extern void	   jbd2_journal_ack_err    (journal_t *);

include/trace/events/ext4.h

@@ -48,6 +48,16 @@ struct partial_cluster;
 	{ EXT4_GET_BLOCKS_KEEP_SIZE,		"KEEP_SIZE" },		\
 	{ EXT4_GET_BLOCKS_ZERO,			"ZERO" })
 
+/*
+ * __print_flags() requires that all enum values be wrapped in the
+ * TRACE_DEFINE_ENUM macro so that the enum value can be encoded in the ftrace
+ * ring buffer.
+ */
+TRACE_DEFINE_ENUM(BH_New);
+TRACE_DEFINE_ENUM(BH_Mapped);
+TRACE_DEFINE_ENUM(BH_Unwritten);
+TRACE_DEFINE_ENUM(BH_Boundary);
+
 #define show_mflags(flags) __print_flags(flags, "",	\
 	{ EXT4_MAP_NEW,		"N" },			\
 	{ EXT4_MAP_MAPPED,	"M" },			\
@@ -62,11 +72,18 @@ struct partial_cluster;
 	{ EXT4_FREE_BLOCKS_NOFREE_FIRST_CLUSTER,"1ST_CLUSTER" },\
 	{ EXT4_FREE_BLOCKS_NOFREE_LAST_CLUSTER,	"LAST_CLUSTER" })
 
+TRACE_DEFINE_ENUM(ES_WRITTEN_B);
+TRACE_DEFINE_ENUM(ES_UNWRITTEN_B);
+TRACE_DEFINE_ENUM(ES_DELAYED_B);
+TRACE_DEFINE_ENUM(ES_HOLE_B);
+TRACE_DEFINE_ENUM(ES_REFERENCED_B);
+
 #define show_extent_status(status) __print_flags(status, "",	\
 	{ EXTENT_STATUS_WRITTEN,	"W" },			\
 	{ EXTENT_STATUS_UNWRITTEN,	"U" },			\
 	{ EXTENT_STATUS_DELAYED,	"D" },			\
-	{ EXTENT_STATUS_HOLE,		"H" })
+	{ EXTENT_STATUS_HOLE,		"H" },			\
+	{ EXTENT_STATUS_REFERENCED,	"R" })
 
 #define show_falloc_mode(mode) __print_flags(mode, "|",		\
 	{ FALLOC_FL_KEEP_SIZE,		"KEEP_SIZE"},		\
@@ -2265,7 +2282,7 @@ DECLARE_EVENT_CLASS(ext4__es_extent,
 		__entry->ino	= inode->i_ino;
 		__entry->lblk	= es->es_lblk;
 		__entry->len	= es->es_len;
-		__entry->pblk	= ext4_es_pblock(es);
+		__entry->pblk	= ext4_es_show_pblock(es);
 		__entry->status	= ext4_es_status(es);
 	),
 
@@ -2354,7 +2371,7 @@ TRACE_EVENT(ext4_es_find_extent_range_exit,
 		__entry->ino	= inode->i_ino;
 		__entry->lblk	= es->es_lblk;
 		__entry->len	= es->es_len;
-		__entry->pblk	= ext4_es_pblock(es);
+		__entry->pblk	= ext4_es_show_pblock(es);
 		__entry->status	= ext4_es_status(es);
 	),
 
@@ -2408,7 +2425,7 @@ TRACE_EVENT(ext4_es_lookup_extent_exit,
 		__entry->ino	= inode->i_ino;
 		__entry->lblk	= es->es_lblk;
 		__entry->len	= es->es_len;
-		__entry->pblk	= ext4_es_pblock(es);
+		__entry->pblk	= ext4_es_show_pblock(es);
 		__entry->status	= ext4_es_status(es);
 		__entry->found	= found;
 	),
@@ -2576,7 +2593,7 @@ TRACE_EVENT(ext4_es_insert_delayed_block,
 		__entry->ino		= inode->i_ino;
 		__entry->lblk		= es->es_lblk;
 		__entry->len		= es->es_len;
-		__entry->pblk		= ext4_es_pblock(es);
+		__entry->pblk		= ext4_es_show_pblock(es);
 		__entry->status		= ext4_es_status(es);
 		__entry->allocated	= allocated;
 	),