Re: [PATCH v3] btrfs: fix a check-integrity warning on write caching disabled disk

[David Sterba <dsterba@suse.cz>]

On Thu, Nov 04, 2021 at 11:34:59PM +0800, Wang Yugui wrote:
> > On Wed, Nov 03, 2021 at 08:07:22AM +0800, Wang Yugui wrote:
> > > > On Thu, Oct 28, 2021 at 06:32:54AM +0800, Wang Yugui wrote:
> > > > > When a disk has write caching disabled, we skip submission of a bio
> > > > > with flush and sync requests before writing the superblock, since
> > > > > it's not needed. However when the integrity checker is enabled,
> > > > > this results in reports that there are metadata blocks referred
> > > > > by a superblock that were not properly flushed. So don't skip the
> > > > > bio submission only when the integrity checker is enabled
> > > > > for the sake of simplicity, since this is a debug tool and
> > > > > not meant for use in non-debug builds.
> > > > > 
> > > > > xfstest/btrfs/220 trigger a check-integrity warning like the following
> > > > > when CONFIG_BTRFS_FS_CHECK_INTEGRITY=y and the disk with WCE=0.
> > > > 
> > > > Does this need the integrity checker to be also enabled by mount
> > > > options? I don't think compile time (ie the #ifdef) is enough, the
> > > > message is printed only when it's enabled based on check in
> > > > __btrfsic_submit_bio "if (!btrfsic_is_initialized) return", where the
> > > > rest of the function does all the verification.
> > > 
> > > Yes.  We need mount option 'check_int' or 'check_int_data' to  trigger
> > > this check-integrity warning.
> > 
> > So the mount options need to be checked here as well.
> 
> We just need to add the mount option(check_int/check_int_data) to
> changelog
> 
> when CONFIG_BTRFS_FS_CHECK_INTEGRITY=y and the disk with WCE=0.
> =>
> when CONFIG_BTRFS_FS_CHECK_INTEGRITY=y , and mount
> option(check_int/check_int_data), and the disk with WCE=0.
> 
> 
> The check of mount option(check_int/check_int_data) is done later
> inside btrfsic_submit_bio. No necessary to check mount
> option(check_int/check_int_data) here, we just need to match
> submit_bio(linux/bio.h).

Ah right, I missed it.