Subject: Re: M_EXT_PAGES
To: YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp>
From: Jason Thorpe <thorpej@wasabisystems.com>
List: tech-kern
Date: 06/05/2004 13:50:14
--Apple-Mail-145-787232514
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII; format=flowed


On Jun 5, 2004, at 12:47 AM, YAMAMOTO Takashi wrote:

> currently, ext_pgs is merely used as a hint for sodopendfreel().
> i'd like to change it to be used by bus_dmamap_load_mbuf() as well.
> (x86 diff is attached)

This is great; I wholeheartedly support this patch.  I intended this 
usage for ext_pgs, but simply never had a chance to implement this 
optimization.

> while i think that the change itsself has its own benefits,
> eventual goals are:
> 	- delay pmap_kenter_pa() until it's really needed.
> 	  (likely on the first mtod()?)
> 	  if you're lucky enough, no pmap_kenter_pa() is needed at all.

Yes, this could really help both MP systems as well as systems with a 
virtually indexed cache.

> 	- use mbufs (or something similar) for disk i/o.

YES!  I would love to see mbuf chain support in the bio subsystem.  It 
could be especially helpful for network storage protocols (including 
NFS).

         -- Jason R. Thorpe <thorpej@wasabisystems.com>


--Apple-Mail-145-787232514
content-type: application/pgp-signature; x-mac-type=70674453;
	name=PGP.sig
content-description: This is a digitally signed message part
content-disposition: inline; filename=PGP.sig
content-transfer-encoding: 7bit

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

iD8DBQFAwjIKOpVKkaBm8XkRAstFAJ4pEACgZa7Z6mvB56o6KZC6oH7sywCeOkq/
lXZRmX7cn6l0QCJ6LqM9M90=
=QWi/
-----END PGP SIGNATURE-----

--Apple-Mail-145-787232514--