On 31 May 2000, Miquel van Smoorenburg wrote:
> In article <cistron.E12wuSf-0002BH-00@the-village.bc.nu>,
> Alan Cox <alan@lxorguk.ukuu.org.uk> wrote:
> >> Yes, MNT_NOT_IF_ROOT makes sense. However, I'ld rather postpone that one
> >> until the decision on mount(2) interface (or the interface of whatever new
> >> syscall(s) will be used). I mean, just look at mount(8) syntax:
> >
> >NOT_IF_ROOT ??? what we care about is remounts not root/no root
>
> NOT_IF_ROOT and MNT_REMOUNT are both confusing since they both can mean
> different things (root as uid 0 or root of fs, remount as in -o remount).
>
> Simply do not allow mounting the same filesystem twice on the same
> mountpoint. Mounting _another_ filesystem on an existing mountpoint
> should not be prohibited as that is needed for union mounts.
Well, union-mounts have nothing with the picture, since they _have_ to get
the flag of their own. Internally VFS doesn't give a damn for multiple
mounts and (almost) doesn't give a damn for mounting over the existing
mountpoint (almost == modulo a couple of harmless but surprising side
effects produced by the code that will be changed by union-mount support
anyway). So we are talking about the logics in wrapper around the real
stuff: it's pure API question - how to encode the desired actions into
the arguments passed.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed May 31 2000 - 21:00:26 EST