Subject: Re: letting userland issue FUA writes
To: Jason Thorpe <thorpej@shagadelic.org>
From: Daniel Carosone <dan@geek.com.au>
List: tech-kern
Date: 03/20/2006 14:15:17
--+aJhGqIeocQJJWbN
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Sun, Mar 19, 2006 at 07:07:06PM -0800, Bill Studenmund wrote:

> Since we are adding flags, the "p"/"non-p*" variants collapse into a flag=
=20
> being set or cleared. :-)

Yeah, that was my assumption too.

> My question was if we want both iovec and non-iovec versions.

I'm thinking a readx/writex that take an iovec and flags.

Then all the previous interfaces can be implemented internally in
terms of readx/writex (p/non-p as a flag, non-iovec as a
single synthesised iovec, etc) if worthwhile.

Worth thinking ahead of what other flags we might want, and what they
might imply for the interface of the new iovec structure...

--
Dan.


--+aJhGqIeocQJJWbN
Content-Type: application/pgp-signature
Content-Disposition: inline

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

iD8DBQFEHh5FEAVxvV4N66cRAvkwAKDX1Xx121CZNyL/PSkqylTjeAzzGQCfYbdA
lKh2ykCxmKY+u4wsbydFeE4=
=mjnP
-----END PGP SIGNATURE-----

--+aJhGqIeocQJJWbN--