Miklos Szeredi <miklos@xxxxxxxxxx> writes: > On Thu, 11 Sep 2008, ebiederm@xxxxxxxxxxxx (Eric W. Biederman) >> There is a weird corner case I'm trying to wrap my head around. >> unlink and rmdir do not work on dentries that are mount points >> in another mount namespace. >> >> Which is at least needed for the moment so we don't leak mounts. >> >> Once we have unprivileged mounts does that introduce a DOS attack? > > Hmm, yes. That's a tough one... > > I think if the dentry has only user mounts, unlink should go ahead and > on success dissolve any mounts on the dentry. Does that sound > workable? I don't think only user mounts is the right filter. We have support for lazy unmounts so it is possible to handle that case. Technically all we need to do is transform d_mounted from a counter to a hlist_head and thread yet another list through struct vfs_mount to track this. I need to think about the semantics a little more before I have a good feel of what makes sense. In particular do we want a full recursive lazy unmount or do we want to handle submounts in a different way. This also intersects in interesting ways with dcache pruning, and automounting. Eric -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
