--phCU5ROyZO6kBE05
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Jan 03, 2006 at 11:51:23PM +0100, Quentin Garnier wrote:
> On Tue, Jan 03, 2006 at 04:12:05PM -0500, der Mouse wrote:
> > >> - keep new dev numbers strictly in sync (like sparc and sparc64 do)
> > > i think this is the crux of the problem; for a purely cosmetic reason
> > > a different device format was chosen.
> >=20
> > Purely cosmetic?  I thought it was done to avoid a flag day in /dev
> > when switching from an 8-partition kernel to a 16-partition kernel.
> >=20
> > Of course, you may consider that "purely cosmetic".  Personally, I'm
> > inclined to agree, but I suspect a lot of relatively Unix-na?ve users
> > would disagree.
>=20
> The choice was cosmetic for the amd64 port.  And that's what bitting us
> now.

I disagree. What was done for the i386 port should have been done=20
differently; it was a grotesque hack that should not have been propogated.

The amd64 port did the right thing by not carrying it forward.

Take care,

Bill

--phCU5ROyZO6kBE05
Content-Type: application/pgp-signature
Content-Disposition: inline

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

iD8DBQFDuxYtWz+3JHUci9cRAq4lAJ9CHqrdEn2IM/+sVYD0UJOI07i1ugCfYsD7
jeFmEalalFzw2pYof4pe4vM=
=JSF7
-----END PGP SIGNATURE-----

--phCU5ROyZO6kBE05--