On 2022/6/7 17:49, Filipe Manana wrote:

On Tue, Jun 07, 2022 at 10:31:46AM +0800, Qu Wenruo wrote:

...

[BUG] If we have a btrfs image with dirty log, along with an unsupported RO compatible flag:

log_root 30474240 ... compat_flags 0x0 compat_ro_flags 0x40000003 ( FREE_SPACE_TREE | FREE_SPACE_TREE_VALID | unknown flag: 0x40000000 )

Then even if we can only mount it RO, we will still cause metadata update for log replay:

BTRFS info (device dm-1): flagging fs with big metadata feature BTRFS info (device dm-1): using free space tree BTRFS info (device dm-1): has skinny extents BTRFS info (device dm-1): start tree-log replay

This is definitely against RO compact flag requirement.

[CAUSE] RO compact flag only forces us to do RO mount, but we will still do log replay for plain RO mount.

Thus this will result us to do log replay and update metadata.

This can be very problematic for new RO compat flag, for example older kernel can not understand v2 cache, and if we allow metadata update on RO mount and invalidate/corrupt v2 cache.

[FIX] Just set the nologreplay flag if there is any unsupported RO compact flag.

This will reject log replay no matter if we have dirty log or not, with the following message:

BTRFS info (device dm-1): disabling log replay due to unsupported ro compat features

Cc: stable@vger.kernel.org #4.9+ Signed-off-by: Qu Wenruo wqu@suse.com

---

fs/btrfs/disk-io.c | 8 ++++++++ 1 file changed, 8 insertions(+)

diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index fe309db9f5ff..d06f1a176b5b 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -3655,6 +3655,14 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device err = -EINVAL; goto fail_alloc; }

• /*

• * We have unsupported RO compat features, although RO mounted, we
  

• * should any metadata write, including the log replay.
  

• * Or we can screw up whatever the new feature requires.
  

• */
  

• if (features)

• btrfs_set_and_info(fs_info, NOLOGREPLAY,
  

• "disabling log replay due to unsupported ro compat features");
  

Well, this might be surprising for users.

On mount, it's expected that everything that was fsynced is available. Yes, there's a message printed informing the logs were not replayed, but this allows for applications to read stale data.

I think just failing the mount and printing a message telling that the fs needs to be explicitly mounted with -o nologreplay is less prone to having stale data being read and used.

OK, I'm fine with either method, as long as we reject the log when there is unsupported RO compat feature.

Thanks, Qu

Thanks.

...

if (sectorsize < PAGE_SIZE) { struct btrfs_subpage_info *subpage_info; -- 2.36.1