This needs to be consistent across kernel- and user-space, and I
believe we settled on continuing to allow cross-quota links (despite
the accounting issues) because otherwise the introduction of new
quotas makes it incredibly messy. Unless I missed a change, Zheng?
-Greg
On Fri, Aug 10, 2018 at 2:57 AM, Chengguang Xu <cgxu519@xxxxxxx> wrote:
> Currently, we allow making hard link bewteen different
> quota realms and it may cause inaccurate quota accouting.
> This patch adds quota realm check when doing hard link.
>
> Signed-off-by: Chengguang Xu <cgxu519@xxxxxxx>
> ---
> fs/ceph/dir.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
> index 82928cea0209..87fa996dbad4 100644
> --- a/fs/ceph/dir.c
> +++ b/fs/ceph/dir.c
> @@ -993,6 +993,10 @@ static int ceph_link(struct dentry *old_dentry, struct inode *dir,
> if (ceph_snap(dir) != CEPH_NOSNAP)
> return -EROFS;
>
> + /* don't allow cross-quota link */
> + if (!ceph_quota_is_same_realm(d_inode(old_dentry), dir))
> + return -EXDEV;
> +
> dout("link in dir %p old_dentry %p dentry %p\n", dir,
> old_dentry, dentry);
> req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LINK, USE_AUTH_MDS);
> --
> 2.17.1
>
