Re: [PATCH] xfs: hold quota inode ILOCK_EXCL until the end of dqalloc

[Dave Chinner <david@fromorbit.com>]

On Tue, Jan 04, 2022 at 03:42:16PM -0800, Darrick J. Wong wrote:
> From: Darrick J. Wong <djwong@kernel.org>
> 
> Online fsck depends on callers holding ILOCK_EXCL from the time they
> decide to update a block mapping until after they've updated the reverse
> mapping records to guarantee the stability of both mapping records.
> Unfortunately, the quota code drops ILOCK_EXCL at the first transaction
> roll in the dquot allocation process, which breaks that assertion.  This
> leads to sporadic failures in the online rmap repair code if the repair
> code grabs the AGF after bmapi_write maps a new block into the quota
> file's data fork but before it can finish the deferred rmap update.
> 
> Fix this by rewriting the function to hold the ILOCK until after the
> transaction commit like all other bmap updates do, and get rid of the
> dqread wrapper that does nothing but complicate the codebase.
> 
> Signed-off-by: Darrick J. Wong <djwong@kernel.org>
> ---
>  fs/xfs/xfs_dquot.c |   79 ++++++++++++++++++----------------------------------
>  1 file changed, 28 insertions(+), 51 deletions(-)

Yup, much nicer. I was just pondering if I should clean up the weird
transaction stuff when looking at the xfs_bmapi_write() call in this
function, but now I don't have to :)

Reviewed-by: Dave Chinner <dchinner@redhat.com>
-- 
Dave Chinner
david@fromorbit.com