[Devel] [PATCH RHEL7 COMMIT] ms/ext4: handle unwritten or delalloc buffers before enabling data journaling

[Vasily Averin]

The commit is pushed to "branch-rh7-3.10.0-1160.31.1.vz7.181.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git
after rh7-3.10.0-1160.31.1.vz7.181.8
------>
commit 10151fe268075e193f27a4d32099409b79307dd9
Author: Daeho Jeong <daeho.jeong at samsung.com>
Date:   Tue Jun 22 21:51:16 2021 +0300

    ms/ext4: handle unwritten or delalloc buffers before enabling data journaling
    
        ms commit 4c54659269ec
    
        We already allocate delalloc blocks before changing the inode mode into
        "per-file data journal" mode to prevent delalloc blocks from remaining
        not allocated, but another issue concerned with "BH_Unwritten" status
        still exists. For example, by fallocate(), several buffers' status
        change into "BH_Unwritten", but these buffers cannot be processed by
        ext4_alloc_da_blocks(). So, they still remain in unwritten status after
        per-file data journaling is enabled and they cannot be changed into
        written status any more and, if they are journaled and eventually
        checkpointed, these unwritten buffer will cause a kernel panic by the
        below BUG_ON() function of submit_bh_wbc() when they are submitted
        during checkpointing.
    
        static int submit_bh_wbc(int rw, struct buffer_head *bh,...
        {
                ...
                BUG_ON(buffer_unwritten(bh));
    
        Moreover, when "dioread_nolock" option is enabled, the status of a
        buffer is changed into "BH_Unwritten" after write_begin() completes and
        the "BH_Unwritten" status will be cleared after I/O is done. Therefore,
        if a buffer's status is changed into unwrutten but the buffer's I/O is
        not submitted and completed, it can cause the same problem after
        enabling per-file data journaling. You can easily generate this bug by
        executing the following command.
    
        ./kvm-xfstests -C 10000 -m nodelalloc,dioread_nolock generic/269
    
        To resolve these problems and define a boundary between the previous
        mode and per-file data journaling mode, we need to flush and wait all
        the I/O of buffers of a file before enabling per-file data journaling
        of the file.
    
        Signed-off-by: Daeho Jeong <daeho.jeong at samsung.com>
        Signed-off-by: Theodore Ts'o <tytso at mit.edu>
        Reviewed-by: Jan Kara <jack at suse.cz>
    
    Signed-off-by: Kirill Tkhai <ktkhai at virtuozzo.com>
---
 fs/ext4/inode.c | 31 ++++++++++++++++++++-----------
 1 file changed, 20 insertions(+), 11 deletions(-)

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 5fe7026..57d3786 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -5612,22 +5612,29 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val)
 		return 0;
 	if (is_journal_aborted(journal))
 		return -EROFS;
-	/* We have to allocate physical blocks for delalloc blocks
-	 * before flushing journal. otherwise delalloc blocks can not
-	 * be allocated any more. even more truncate on delalloc blocks
-	 * could trigger BUG by flushing delalloc blocks in journal.
-	 * There is no delalloc block in non-journal data mode.
-	 */
-	if (val && test_opt(inode->i_sb, DELALLOC)) {
-		err = ext4_alloc_da_blocks(inode);
-		if (err < 0)
-			return err;
-	}
 
 	/* Wait for all existing dio workers */
 	ext4_inode_block_unlocked_dio(inode);
 	inode_dio_wait(inode);
 
+	/*
+	 * Before flushing the journal and switching inode's aops, we have
+	 * to flush all dirty data the inode has. There can be outstanding
+	 * delayed allocations, there can be unwritten extents created by
+	 * fallocate or buffered writes in dioread_nolock mode covered by
+	 * dirty data which can be converted only after flushing the dirty
+	 * data (and journalled aops don't know how to handle these cases).
+	 */
+	if (val) {
+		down_write(&EXT4_I(inode)->i_mmap_sem);
+		err = filemap_write_and_wait(inode->i_mapping);
+		if (err < 0) {
+			up_write(&EXT4_I(inode)->i_mmap_sem);
+			ext4_inode_resume_unlocked_dio(inode);
+			return err;
+		}
+	}
+
 	jbd2_journal_lock_updates(journal);
 
 	/*
@@ -5652,6 +5659,8 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val)
 	ext4_set_aops(inode);
 
 	jbd2_journal_unlock_updates(journal);
+	if (val)
+		up_write(&EXT4_I(inode)->i_mmap_sem);
 	ext4_inode_resume_unlocked_dio(inode);
 
 	/* Finally we can mark the inode as dirty. */