--14l74oP8pwp5L87R
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Nov 29, 2005 at 02:32:12AM +0900, YAMAMOTO Takashi wrote:
> > So would I, but there were objections from several folks with valued
> > opinions.  Are those objections still held?
>=20
> but i have no objection against fixing the calculation
> as far as it's better than current.

Ok, I have committed the updated calculation.  I did it in two parts,
one to redo the existing result and one to take into account the
additional factors, just because :)

Thanks for your help refining this.

> i still think mandating -s is better.

A hybrid is possible: only if -s is specified, then 'trust' it as you
propose.  Or even another mount flag to make -s behave that way. But
even there, I still think you want a different number, as a kind of
minimum-size reservation, with -s as the upper limit still. At least,
that's what I'd want if I was to use it that way.

It all seems a bit over-wrought and unnecessary, and long past the
point of being a useful discussion; I don't mean to prolong it here.

--
Dan.
--14l74oP8pwp5L87R
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (NetBSD)

iD8DBQFDi4e0EAVxvV4N66cRAi5DAJ0VLyNU1p5Nq6Js6rawWePiYSOfAQCfSrFG
JkFka5do9MElqWGc4C69bG0=
=avOd
-----END PGP SIGNATURE-----

--14l74oP8pwp5L87R--