Re: [PATCH 3/3] xfs: port ondisk structure checks from xfs/122 to the kernel

["Darrick J. Wong" <djwong@kernel.org>]

On Mon, Nov 04, 2024 at 05:29:04AM +0100, Christoph Hellwig wrote:
> On Fri, Nov 01, 2024 at 03:19:05PM -0700, Darrick J. Wong wrote:
> > +#define XFS_CHECK_SB_OFFSET(field, offset) \
> > +	XFS_CHECK_OFFSET(struct xfs_dsb, field, offset); \
> > +	XFS_CHECK_OFFSET(struct xfs_sb, field, offset);
> 
> Despite comments to the contrary, xfs_sb is purely an in-memory structure
> and nothing cares about it having the same layout as xfs_dsb.  As we've
> kept them in sync so far I'm fine with adding this check under the
> expectation that I can remove it again when I finally start removing
> struct xfs_sb, which is long overdue.

Sounds good to me -- the only reason it's here is because the current
codebase requires it.  And that's only after years of djwong-wtf
actually having unsynchronized xfs_sb/xfs_dsb, so I think the codebase
doesn't even require it, but I've not done an exhaustive search.

> Otherwise:
> 
> Reviewed-by: Christoph Hellwig <hch@lst.de>

Thanks!

--D