--8P1HSweYDcXXzwPJ
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Fri, Sep 16, 2005 at 09:23:19AM +0900, YAMAMOTO Takashi wrote:
> > Actually, the root node usually gets identifier 2 in a number of file=
=20
> > systems. Node 1 is often used as roots "parent."
> >=20
> > Other file systems have taken the root =3D=3D 2 behavior and used inode=
 1 for=20
> > special things.
>=20
> yes.
> do you think tmpfs should follow the convention?

If it's easy, I'd say tmpfs should follow the convention.

> ie. do you know anything assuming it?

I'm not sure. I believe the kernel (outside any given fs) does not care,=20
but I am concerned about userland apps. I don't _know_ of anything that=20
will break, but I have no idea how far assumptions about ufs/ffs have=20
spread. And I am unsure what will happen if they are broken. :-|

Take care,

Bill

--8P1HSweYDcXXzwPJ
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFDMDSoWz+3JHUci9cRAuS5AJ9ZyvHU7vzF17qUwosbK25Ymc6aBwCfWbxQ
FBw8+zX3EAACFgNnkEBmZl8=
=ygm9
-----END PGP SIGNATURE-----

--8P1HSweYDcXXzwPJ--