On Mon, Feb 08, 2016 at 11:55:06PM -0800, Darrick J. Wong wrote: > On Tue, Feb 09, 2016 at 06:43:30PM +1100, Dave Chinner wrote: > > On Mon, Feb 08, 2016 at 05:13:03PM -0800, Darrick J. Wong wrote: > > > Include the refcount and rmap structures in the golden output. > > > > > > Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx> > > > --- > > > tests/xfs/122 | 3 +++ > > > tests/xfs/122.out | 4 ++++ > > > tests/xfs/group | 2 +- > > > 3 files changed, 8 insertions(+), 1 deletion(-) > > > > > > > > > diff --git a/tests/xfs/122 b/tests/xfs/122 > > > index e6697a2..758cb50 100755 > > > --- a/tests/xfs/122 > > > +++ b/tests/xfs/122 > > > @@ -90,6 +90,9 @@ xfs_da3_icnode_hdr > > > xfs_dir3_icfree_hdr > > > xfs_dir3_icleaf_hdr > > > xfs_name > > > +xfs_owner_info > > > +xfs_refcount_irec > > > +xfs_rmap_irec > > > xfs_alloctype_t > > > xfs_buf_cancel_t > > > xfs_bmbt_rec_32_t > > > > So this is going to cause failures on any userspace that doesn't > > know about these new types, right? > > > > Should these be conditional in some way? > > I wasn't sure how to handle this -- I could just keep the patch at the head of > my stack (unreleased) until xfsprogs pulls in the appropriate libxfs pieces? > So long as we're not dead certain of the final format of the rmapbt and > refcountbt, there's probably not a lot of value in putting this in (yet). Well, I'm more concerned about running on older/current distros that don't have support for them in userspace. My brain is mush right now, so I don't have any brilliant ideas (hence the question, rather than also presenting a posible solution). I'll have a think; maybe we can make use of the configurable .out file code we have now? Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
