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

Pull ext4 updates from Ted Ts'o:
 "All cleanups and bug fixes; most notably, fix some problems discovered
  in ext4's NFS support, and fix an ioctl (EXT4_IOC_GROUP_ADD) used by
  old versions of e2fsprogs which we accidentally broke a while back.

  Also fixed some error paths in ext4's quota and inline data support.

  Finally, improve tail latency in jbd2's commit code"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: check for shutdown and r/o file system in ext4_write_inode()
  ext4: force inode writes when nfsd calls commit_metadata()
  ext4: avoid declaring fs inconsistent due to invalid file handles
  ext4: include terminating u32 in size of xattr entries when expanding inodes
  ext4: compare old and new mode before setting update_mode flag
  ext4: fix EXT4_IOC_GROUP_ADD ioctl
  ext4: hard fail dax mount on unsupported devices
  jbd2: update locking documentation for transaction_t
  ext4: remove redundant condition check
  jbd2: clean up indentation issue, replace spaces with tab
  ext4: clean up indentation issues, remove extraneous tabs
  ext4: missing unlock/put_page() in ext4_try_to_write_inline_data()
  ext4: fix possible use after free in ext4_quota_enable
  jbd2: avoid long hold times of j_state_lock while committing a transaction
  ext4: add ext4_sb_bread() to disambiguate ENOMEM cases

1 files changed, 57 insertions, 35 deletions

diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 53ff6c2a26ed..d6c142d73d99 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -140,6 +140,29 @@ MODULE_ALIAS_FS("ext3");
 MODULE_ALIAS("ext3");
 #define IS_EXT3_SB(sb) ((sb)->s_bdev->bd_holder == &ext3_fs_type)
 
+/*
+ * This works like sb_bread() except it uses ERR_PTR for error
+ * returns.  Currently with sb_bread it's impossible to distinguish
+ * between ENOMEM and EIO situations (since both result in a NULL
+ * return.
+ */
+struct buffer_head *
+ext4_sb_bread(struct super_block *sb, sector_t block, int op_flags)
+{
+	struct buffer_head *bh = sb_getblk(sb, block);
+
+	if (bh == NULL)
+		return ERR_PTR(-ENOMEM);
+	if (buffer_uptodate(bh))
+		return bh;
+	ll_rw_block(REQ_OP_READ, REQ_META | op_flags, 1, &bh);
+	wait_on_buffer(bh);
+	if (buffer_uptodate(bh))
+		return bh;
+	put_bh(bh);
+	return ERR_PTR(-EIO);
+}
+
 static int ext4_verify_csum_type(struct super_block *sb,
 				 struct ext4_super_block *es)
 {
@@ -1000,14 +1023,13 @@ static void ext4_put_super(struct super_block *sb)
 		invalidate_bdev(sbi->journal_bdev);
 		ext4_blkdev_remove(sbi);
 	}
-	if (sbi->s_ea_inode_cache) {
-		ext4_xattr_destroy_cache(sbi->s_ea_inode_cache);
-		sbi->s_ea_inode_cache = NULL;
-	}
-	if (sbi->s_ea_block_cache) {
-		ext4_xattr_destroy_cache(sbi->s_ea_block_cache);
-		sbi->s_ea_block_cache = NULL;
-	}
+
+	ext4_xattr_destroy_cache(sbi->s_ea_inode_cache);
+	sbi->s_ea_inode_cache = NULL;
+
+	ext4_xattr_destroy_cache(sbi->s_ea_block_cache);
+	sbi->s_ea_block_cache = NULL;
+
 	if (sbi->s_mmp_tsk)
 		kthread_stop(sbi->s_mmp_tsk);
 	brelse(sbi->s_sbh);
@@ -1151,20 +1173,11 @@ static struct inode *ext4_nfs_get_inode(struct super_block *sb,
 {
 	struct inode *inode;
 
-	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
-		return ERR_PTR(-ESTALE);
-	if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count))
-		return ERR_PTR(-ESTALE);
-
-	/* iget isn't really right if the inode is currently unallocated!!
-	 *
-	 * ext4_read_inode will return a bad_inode if the inode had been
-	 * deleted, so we should be safe.
-	 *
+	/*
 	 * Currently we don't know the generation for parent directory, so
 	 * a generation of 0 means "accept any"
 	 */
-	inode = ext4_iget_normal(sb, ino);
+	inode = ext4_iget(sb, ino, EXT4_IGET_HANDLE);
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
 	if (generation && inode->i_generation != generation) {
@@ -1189,6 +1202,16 @@ static struct dentry *ext4_fh_to_parent(struct super_block *sb, struct fid *fid,
 				    ext4_nfs_get_inode);
 }
 
+static int ext4_nfs_commit_metadata(struct inode *inode)
+{
+	struct writeback_control wbc = {
+		.sync_mode = WB_SYNC_ALL
+	};
+
+	trace_ext4_nfs_commit_metadata(inode);
+	return ext4_write_inode(inode, &wbc);
+}
+
 /*
  * Try to release metadata pages (indirect blocks, directories) which are
  * mapped via the block device.  Since these pages could have journal heads
@@ -1393,6 +1416,7 @@ static const struct export_operations ext4_export_ops = {
 	.fh_to_dentry = ext4_fh_to_dentry,
 	.fh_to_parent = ext4_fh_to_parent,
 	.get_parent = ext4_get_parent,
+	.commit_metadata = ext4_nfs_commit_metadata,
 };
 
 enum {
@@ -1939,7 +1963,7 @@ static int handle_mount_opt(struct super_block *sb, char *opt, int token,
 #ifdef CONFIG_FS_DAX
 		ext4_msg(sb, KERN_WARNING,
 		"DAX enabled. Warning: EXPERIMENTAL, use at your own risk");
-			sbi->s_mount_opt |= m->mount_opt;
+		sbi->s_mount_opt |= m->mount_opt;
 #else
 		ext4_msg(sb, KERN_INFO, "dax option not supported");
 		return -1;
@@ -3842,12 +3866,12 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 		if (ext4_has_feature_inline_data(sb)) {
 			ext4_msg(sb, KERN_ERR, "Cannot use DAX on a filesystem"
 					" that may contain inline data");
-			sbi->s_mount_opt &= ~EXT4_MOUNT_DAX;
+			goto failed_mount;
 		}
 		if (!bdev_dax_supported(sb->s_bdev, blocksize)) {
 			ext4_msg(sb, KERN_ERR,
-				"DAX unsupported by block device. Turning off DAX.");
-			sbi->s_mount_opt &= ~EXT4_MOUNT_DAX;
+				"DAX unsupported by block device.");
+			goto failed_mount;
 		}
 	}
 
@@ -4328,7 +4352,7 @@ no_journal:
 	 * so we can safely mount the rest of the filesystem now.
 	 */
 
-	root = ext4_iget(sb, EXT4_ROOT_INO);
+	root = ext4_iget(sb, EXT4_ROOT_INO, EXT4_IGET_SPECIAL);
 	if (IS_ERR(root)) {
 		ext4_msg(sb, KERN_ERR, "get root inode failed");
 		ret = PTR_ERR(root);
@@ -4522,14 +4546,12 @@ failed_mount4:
 	if (EXT4_SB(sb)->rsv_conversion_wq)
 		destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
 failed_mount_wq:
-	if (sbi->s_ea_inode_cache) {
-		ext4_xattr_destroy_cache(sbi->s_ea_inode_cache);
-		sbi->s_ea_inode_cache = NULL;
-	}
-	if (sbi->s_ea_block_cache) {
-		ext4_xattr_destroy_cache(sbi->s_ea_block_cache);
-		sbi->s_ea_block_cache = NULL;
-	}
+	ext4_xattr_destroy_cache(sbi->s_ea_inode_cache);
+	sbi->s_ea_inode_cache = NULL;
+
+	ext4_xattr_destroy_cache(sbi->s_ea_block_cache);
+	sbi->s_ea_block_cache = NULL;
+
 	if (sbi->s_journal) {
 		jbd2_journal_destroy(sbi->s_journal);
 		sbi->s_journal = NULL;
@@ -4598,7 +4620,7 @@ static struct inode *ext4_get_journal_inode(struct super_block *sb,
 	 * happen if we iget() an unused inode, as the subsequent iput()
 	 * will try to delete it.
 	 */
-	journal_inode = ext4_iget(sb, journal_inum);
+	journal_inode = ext4_iget(sb, journal_inum, EXT4_IGET_SPECIAL);
 	if (IS_ERR(journal_inode)) {
 		ext4_msg(sb, KERN_ERR, "no journal found");
 		return NULL;
@@ -5680,7 +5702,7 @@ static int ext4_quota_enable(struct super_block *sb, int type, int format_id,
 	if (!qf_inums[type])
 		return -EPERM;
 
-	qf_inode = ext4_iget(sb, qf_inums[type]);
+	qf_inode = ext4_iget(sb, qf_inums[type], EXT4_IGET_SPECIAL);
 	if (IS_ERR(qf_inode)) {
 		ext4_error(sb, "Bad quota inode # %lu", qf_inums[type]);
 		return PTR_ERR(qf_inode);
@@ -5690,9 +5712,9 @@ static int ext4_quota_enable(struct super_block *sb, int type, int format_id,
 	qf_inode->i_flags |= S_NOQUOTA;
 	lockdep_set_quota_inode(qf_inode, I_DATA_SEM_QUOTA);
 	err = dquot_enable(qf_inode, type, format_id, flags);
-	iput(qf_inode);
 	if (err)
 		lockdep_set_quota_inode(qf_inode, I_DATA_SEM_NORMAL);
+	iput(qf_inode);
 
 	return err;
 }