Subject: Re: NetBSD 1.6 Release Schedule
To: Wojciech Puchar <wojtek@chylonia.3miasto.net>
From: Ignatios Souvatzis <ignatios@theory.cs.uni-bonn.de>
List: port-i386
Date: 06/20/2002 14:18:06
--OBd5C1Lgu00Gd/Tn
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Thu, Jun 20, 2002 at 01:19:51PM +0200, Wojciech Puchar wrote:
> exactly. in simplest case one queue for reads, one for writes.
>=20
> then making request in 1:1 method for drives with write cache (drive will
> then order reads and writes in good sequence).

You've just broken not only the filesystems' method to make sure the
filesystem can be repaired after a crash, but also the filesystems' method
to makre sure it sees consistent data during normal operation.

Alas, it's more complicated than that.

Btw: is there any special reason to discuss this on port-i386?

Regards,
	-is

--OBd5C1Lgu00Gd/Tn
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: 2.6.i

iQEVAgUBPRHH+zCn4om+4LhpAQEZtwf9GsiKe3Hvf0lCzb6T99fiu1vkZ9bOY1a6
ZBd8fHugf/Iv8ismsUsQFTHs1kbtepno36Hn4S3AtNQsl+7Bi9XJvSVBT9xkVyKh
15kVkDMvNMo4LOpJQmtABa3kRmi17UxsswELjWz4ZIwJXc18mF8/QGhqP3wguOgg
oWzSeRQdXMT8ug/drf7pK4WtC2turQC2N4cgdUcvFg0q4q0VJRh8f2oIqOQPv6dT
bLNe7pjwEtMh3jRUPjxN+cwZbUplAx+y6D/e2v5YmjtbQowh4rykiV9GxEYl5Yb1
B8FtYU63iEIB/2b3JUtSj1C2ZxOK3L7/8vlqdRBWleeeifiZb34vFw==
=tFL7
-----END PGP SIGNATURE-----

--OBd5C1Lgu00Gd/Tn--