Fair enough.
Actually, I think we only use ATTR_ROOT and ATTR_SECURE for the
namespace flags.
So you could probably use: XFS_ATTR_NSP_ARGS
xfs_attr_leaf.h:#define XFS_ATTR_NSP_ARGS_MASK (ATTR_ROOT | ATTR_SECURE)
xfs_attr_leaf.h:#define XFS_ATTR_NSP_ARGS(flags) ((flags) &
XFS_ATTR_NSP_ARGS_MASK)
and something like:
if (!XFS_ATTR_NSP_ARGS(al_hreq.flags))
return -XFS_ERROR(EINVAL);
Though would probably then need to include the right header (xfs_attr_leaf.h)
for it...
--Tim
Christoph Hellwig wrote:
> xfs_attrlist_by_handle should only take the ATTR_ flags for the root
> namespaces. The ATTR_KERN* flags may change at anytime and expect special
> preconditions that can't be guaranteed for userspace-originating
> requests. For example passing down ATTR_KERNNOVAL through
> xfs_attrlist_by_handle will hit an assert in debug builds currently.
>
>
> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
>
> Index: linux-2.6-xfs/fs/xfs/linux-2.6/xfs_ioctl.c
> ===================================================================
> --- linux-2.6-xfs.orig/fs/xfs/linux-2.6/xfs_ioctl.c 2008-05-28
> 17:37:02.000000000 +0200
> +++ linux-2.6-xfs/fs/xfs/linux-2.6/xfs_ioctl.c 2008-05-28
> 17:42:18.000000000 +0200
> @@ -470,6 +470,12 @@ xfs_attrlist_by_handle(
> if (al_hreq.buflen > XATTR_LIST_MAX)
> return -XFS_ERROR(EINVAL);
>
> + /*
> + * Reject flags, only allow namespaces.
> + */
> + if (al_hreq.flags & ~(ATTR_ROOT|ATTR_TRUST|ATTR_SECURE))
> + return -XFS_ERROR(EINVAL);
> +
> error = xfs_vget_fsop_handlereq(mp, parinode, &al_hreq.hreq, &inode);
> if (error)
> goto out;
>
|