Re: [PATCH 08/16] xfs_repair: don't fail directory repairs when grabbing inodes

["Darrick J. Wong" <darrick.wong@oracle.com>]

On Tue, Mar 06, 2018 at 11:43:29AM -0800, Darrick J. Wong wrote:
> On Tue, Mar 06, 2018 at 01:23:51PM -0600, Eric Sandeen wrote:
> > On 3/1/18 1:13 PM, Darrick J. Wong wrote:
> > > From: Darrick J. Wong <darrick.wong@oracle.com>
> > > 
> > > There are a few places where xfs_repair needs to be able to load a
> > > damaged directory inode to perform repairs.  Since inline data fork
> > > verifiers can now be customized, refactor libxfs_iget to enable
> > > repair to get at this so that we don't crash in phase 6.
> > > 
> > > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> > 
> > 
> > Hm but there /is/ no custom verifier here, right?  Just NULL or default?
> > In that case does it make sense to be passing ops around vs. just
> > adding a "bool validate" arg in place of the NULL ops dance you have now?
> 
> Originally it was just an iget flag, but I think Dave wanted repair to
> have the ability to pass in a custom ifork_ops.  Theoretically we could
> have an ifork verifier that does the same things as the libxfs ones
> *except* for the places where repair deliberately zeroes a field that it
> intends to set in a future pass.  I don't really like the prospect of
> maintaining code almost-duplication, but in the mean time this restores
> xfs_repair's ability to fix directory .. entries which we lost in
> 4.10).

As a more concrete example, a phase6 inline directory verifier might
look like this instead of the NULL ops we do now:

xfs_failaddr_t
phase6_dir2_sf_verify(
	struct xfs_inode		*ip)
{
	struct xfs_dir2_sf_hdr		*sfp;
	const struct xfs_dir_ops	*dops;
	struct xfs_ifork		*ifp;
	int				size;

	dops = xfs_dir_get_ops(mp, NULL);
	ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
	sfp = (struct xfs_dir2_sf_hdr *)ifp->if_u1.if_data;
	size = ifp->if_bytes;

	/*
	 * Give up if the directory is way too short.
	 */
	if (size <= offsetof(struct xfs_dir2_sf_hdr, parent) ||
	    size < xfs_dir2_sf_hdr_size(sfp->i8count))
		return __this_address;

	/*
	 * Phase 4 zeroes the .. entry if it's garbage and wants phase 6
	 * to fix it, so let that through.
	 */
	if (dops->sf_get_parent_ino(sfp) == 0)
		return NULL;

	/* Otherwise all regular libxfs checks apply. */
	return xfs_dir2_sf_verify(ip);
}

--D

> --D
> 
> > > ---
> > >  db/attrset.c        |    6 ++++--
> > >  include/xfs_inode.h |    6 ++++--
> > >  libxfs/rdwr.c       |   25 +++++++++++++++++--------
> > >  libxfs/trans.c      |    6 ++++--
> > >  libxfs/util.c       |    2 +-
> > >  repair/phase6.c     |   16 +++++++++++-----
> > >  6 files changed, 41 insertions(+), 20 deletions(-)
> > > 
> > > 
> > > diff --git a/db/attrset.c b/db/attrset.c
> > > index ad3c8f3..457317a 100644
> > > --- a/db/attrset.c
> > > +++ b/db/attrset.c
> > > @@ -151,7 +151,8 @@ attr_set_f(
> > >  		value = NULL;
> > >  	}
> > >  
> > > -	if (libxfs_iget(mp, NULL, iocur_top->ino, 0, &ip)) {
> > > +	if (libxfs_iget(mp, NULL, iocur_top->ino, 0, &ip,
> > > +			&xfs_default_ifork_ops)) {
> > >  		dbprintf(_("failed to iget inode %llu\n"),
> > >  			(unsigned long long)iocur_top->ino);
> > >  		goto out;
> > > @@ -226,7 +227,8 @@ attr_remove_f(
> > >  
> > >  	name = argv[optind];
> > >  
> > > -	if (libxfs_iget(mp, NULL, iocur_top->ino, 0, &ip)) {
> > > +	if (libxfs_iget(mp, NULL, iocur_top->ino, 0, &ip,
> > > +			&xfs_default_ifork_ops)) {
> > >  		dbprintf(_("failed to iget inode %llu\n"),
> > >  			(unsigned long long)iocur_top->ino);
> > >  		goto out;
> > > diff --git a/include/xfs_inode.h b/include/xfs_inode.h
> > > index 92829a2..f29f0f0 100644
> > > --- a/include/xfs_inode.h
> > > +++ b/include/xfs_inode.h
> > > @@ -162,9 +162,11 @@ extern void	libxfs_trans_ichgtime(struct xfs_trans *,
> > >  extern int	libxfs_iflush_int (struct xfs_inode *, struct xfs_buf *);
> > >  
> > >  /* Inode Cache Interfaces */
> > > -extern bool	libxfs_inode_verify_forks(struct xfs_inode *ip);
> > > +extern bool	libxfs_inode_verify_forks(struct xfs_inode *ip,
> > > +				struct xfs_ifork_ops *);
> > >  extern int	libxfs_iget(struct xfs_mount *, struct xfs_trans *, xfs_ino_t,
> > > -				uint, struct xfs_inode **);
> > > +				uint, struct xfs_inode **,
> > > +				struct xfs_ifork_ops *);
> > >  extern void	libxfs_iput(struct xfs_inode *);
> > >  
> > >  #define IRELE(ip) libxfs_iput(ip)
> > > diff --git a/libxfs/rdwr.c b/libxfs/rdwr.c
> > > index 3c5def2..416e89b 100644
> > > --- a/libxfs/rdwr.c
> > > +++ b/libxfs/rdwr.c
> > > @@ -1342,12 +1342,16 @@ extern kmem_zone_t	*xfs_inode_zone;
> > >   */
> > >  bool
> > >  libxfs_inode_verify_forks(
> > > -	struct xfs_inode	*ip)
> > > +	struct xfs_inode	*ip,
> > > +	struct xfs_ifork_ops	*ops)
> > >  {
> > >  	struct xfs_ifork	*ifp;
> > >  	xfs_failaddr_t		fa;
> > >  
> > > -	fa = xfs_ifork_verify_data(ip, &xfs_default_ifork_ops);
> > > +	if (!ops)
> > > +		return true;
> > > +
> > > +	fa = xfs_ifork_verify_data(ip, ops);
> > >  	if (fa) {
> > >  		ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
> > >  		xfs_inode_verifier_error(ip, -EFSCORRUPTED, "data fork",
> > > @@ -1355,7 +1359,7 @@ libxfs_inode_verify_forks(
> > >  		return false;
> > >  	}
> > >  
> > > -	fa = xfs_ifork_verify_attr(ip, &xfs_default_ifork_ops);
> > > +	fa = xfs_ifork_verify_attr(ip, ops);
> > >  	if (fa) {
> > >  		ifp = XFS_IFORK_PTR(ip, XFS_ATTR_FORK);
> > >  		xfs_inode_verifier_error(ip, -EFSCORRUPTED, "attr fork",
> > > @@ -1367,11 +1371,16 @@ libxfs_inode_verify_forks(
> > >  }
> > >  
> > >  int
> > > -libxfs_iget(xfs_mount_t *mp, xfs_trans_t *tp, xfs_ino_t ino, uint lock_flags,
> > > -		xfs_inode_t **ipp)
> > > +libxfs_iget(
> > > +	struct xfs_mount	*mp,
> > > +	struct xfs_trans	*tp,
> > > +	xfs_ino_t		ino,
> > > +	uint			lock_flags,
> > > +	struct xfs_inode	**ipp,
> > > +	struct xfs_ifork_ops	*ifork_ops)
> > >  {
> > > -	xfs_inode_t	*ip;
> > > -	int		error = 0;
> > > +	struct xfs_inode	*ip;
> > > +	int			error = 0;
> > >  
> > >  	ip = kmem_zone_zalloc(xfs_inode_zone, 0);
> > >  	if (!ip)
> > > @@ -1386,7 +1395,7 @@ libxfs_iget(xfs_mount_t *mp, xfs_trans_t *tp, xfs_ino_t ino, uint lock_flags,
> > >  		return error;
> > >  	}
> > >  
> > > -	if (!libxfs_inode_verify_forks(ip)) {
> > > +	if (!libxfs_inode_verify_forks(ip, ifork_ops)) {
> > >  		libxfs_iput(ip);
> > >  		return -EFSCORRUPTED;
> > >  	}
> > > diff --git a/libxfs/trans.c b/libxfs/trans.c
> > > index 0e7b7ae..67b1117 100644
> > > --- a/libxfs/trans.c
> > > +++ b/libxfs/trans.c
> > > @@ -243,9 +243,11 @@ libxfs_trans_iget(
> > >  	xfs_inode_log_item_t	*iip;
> > >  
> > >  	if (tp == NULL)
> > > -		return libxfs_iget(mp, tp, ino, lock_flags, ipp);
> > > +		return libxfs_iget(mp, tp, ino, lock_flags, ipp,
> > > +				&xfs_default_ifork_ops);
> > >  
> > > -	error = libxfs_iget(mp, tp, ino, lock_flags, &ip);
> > > +	error = libxfs_iget(mp, tp, ino, lock_flags, &ip,
> > > +			&xfs_default_ifork_ops);
> > >  	if (error)
> > >  		return error;
> > >  	ASSERT(ip != NULL);
> > > diff --git a/libxfs/util.c b/libxfs/util.c
> > > index aac558c..611ab9c 100644
> > > --- a/libxfs/util.c
> > > +++ b/libxfs/util.c
> > > @@ -494,7 +494,7 @@ libxfs_iflush_int(xfs_inode_t *ip, xfs_buf_t *bp)
> > >  		VFS_I(ip)->i_version++;
> > >  
> > >  	/* Check the inline fork data before we write out. */
> > > -	if (!libxfs_inode_verify_forks(ip))
> > > +	if (!libxfs_inode_verify_forks(ip, &xfs_default_ifork_ops))
> > >  		return -EFSCORRUPTED;
> > >  
> > >  	/*
> > > diff --git a/repair/phase6.c b/repair/phase6.c
> > > index 1a398aa..aff83bc 100644
> > > --- a/repair/phase6.c
> > > +++ b/repair/phase6.c
> > > @@ -927,7 +927,9 @@ mk_orphanage(xfs_mount_t *mp)
> > >  	 * would have been cleared in phase3 and phase4.
> > >  	 */
> > >  
> > > -	if ((i = -libxfs_iget(mp, NULL, mp->m_sb.sb_rootino, 0, &pip)))
> > > +	i = -libxfs_iget(mp, NULL, mp->m_sb.sb_rootino, 0, &pip,
> > > +			&xfs_default_ifork_ops);
> > > +	if (i)
> > >  		do_error(_("%d - couldn't iget root inode to obtain %s\n"),
> > >  			i, ORPHANAGE);
> > >  
> > > @@ -951,7 +953,9 @@ mk_orphanage(xfs_mount_t *mp)
> > >  	 * use iget/ijoin instead of trans_iget because the ialloc
> > >  	 * wrapper can commit the transaction and start a new one
> > >  	 */
> > > -/*	if ((i = -libxfs_iget(mp, NULL, mp->m_sb.sb_rootino, 0, &pip)))
> > > +/*	i = -libxfs_iget(mp, NULL, mp->m_sb.sb_rootino, 0, &pip,
> > > +			&xfs_default_ifork_ops);
> > > +	if (i)
> > >  		do_error(_("%d - couldn't iget root inode to make %s\n"),
> > >  			i, ORPHANAGE);*/
> > >  
> > > @@ -1066,7 +1070,8 @@ mv_orphanage(
> > >  	xname.len = snprintf((char *)fname, sizeof(fname), "%llu",
> > >  				(unsigned long long)ino);
> > >  
> > > -	err = -libxfs_iget(mp, NULL, orphanage_ino, 0, &orphanage_ip);
> > > +	err = -libxfs_iget(mp, NULL, orphanage_ino, 0, &orphanage_ip,
> > > +			&xfs_default_ifork_ops);
> > >  	if (err)
> > >  		do_error(_("%d - couldn't iget orphanage inode\n"), err);
> > >  	/*
> > > @@ -1078,7 +1083,8 @@ mv_orphanage(
> > >  		xname.len = snprintf((char *)fname, sizeof(fname), "%llu.%d",
> > >  					(unsigned long long)ino, ++incr);
> > >  
> > > -	if ((err = -libxfs_iget(mp, NULL, ino, 0, &ino_p)))
> > > +	err = -libxfs_iget(mp, NULL, ino, 0, &ino_p, &xfs_default_ifork_ops);
> > > +	if (err)
> > >  		do_error(_("%d - couldn't iget disconnected inode\n"), err);
> > >  
> > >  	xname.type = libxfs_mode_to_ftype(VFS_I(ino_p)->i_mode);
> > > @@ -2827,7 +2833,7 @@ process_dir_inode(
> > >  
> > >  	ASSERT(!is_inode_refchecked(irec, ino_offset) || dotdot_update);
> > >  
> > > -	error = -libxfs_iget(mp, NULL, ino, 0, &ip);
> > > +	error = -libxfs_iget(mp, NULL, ino, 0, &ip, NULL);
> > >  	if (error) {
> > >  		if (!no_modify)
> > >  			do_error(
> > > 
> > > --
> > > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> > > the body of a message to majordomo@vger.kernel.org
> > > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > > 
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> --
> To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html