From owner-freebsd-new-bus  Wed Jan 26 13:39: 8 2000
Delivered-To: freebsd-new-bus@freebsd.org
Received: from rover.village.org (rover.village.org [204.144.255.49])
	by hub.freebsd.org (Postfix) with ESMTP id A9A68154A3
	for <new-bus@FreeBSD.ORG>; Wed, 26 Jan 2000 13:39:04 -0800 (PST)
	(envelope-from imp@harmony.village.org)
Received: from harmony.village.org (harmony.village.org [10.0.0.6])
	by rover.village.org (8.9.3/8.9.3) with ESMTP id OAA01265;
	Wed, 26 Jan 2000 14:39:00 -0700 (MST)
	(envelope-from imp@harmony.village.org)
Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id OAA02463; Wed, 26 Jan 2000 14:38:36 -0700 (MST)
Message-Id: <200001262138.OAA02463@harmony.village.org>
To: Jeroen Ruigrok/Asmodai <asmodai@wxs.nl>
Subject: Re: newbus from a documentation view 
Cc: "Matthew N. Dodd" <winter@jurai.net>, new-bus@FreeBSD.ORG
In-reply-to: Your message of "Wed, 26 Jan 2000 18:23:10 +0100."
		<20000126182309.H290@daemon.ninth-circle.org> 
References: <20000126182309.H290@daemon.ninth-circle.org>  <20000126094020.G290@daemon.ninth-circle.org> <Pine.BSF.4.21.0001260341430.462-100000@sasami.jurai.net> 
Date: Wed, 26 Jan 2000 14:38:36 -0700
From: Warner Losh <imp@village.org>
Sender: owner-freebsd-new-bus@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

In message <20000126182309.H290@daemon.ninth-circle.org> Jeroen Ruigrok/Asmodai writes:
: >> machine/bus_memio.h <also in alpha and i386 version>
: >> 	I see that this one is basically an empty header, deprecated in
: >> 	usage?
: 
: I see we #define bus_memio_h in machine/bus.h, so basically all drivers
: including machine/bus_memio.h and bus_pio.h are kinda out of synch with
: newbus/busspace?
: 
: >> machine/bus_pio.h <also in alpha and i386 version>
: >> 	I see that this one is also an empty header, also deprecated in
: >> 	usage?
: 
: See above.

The way that Justin had intended it is as follows.

If the device hardware supported I/O mapping, then you'd include
<machine/bus_pio.h> in your driver.  It would do what was needed to
bring in support for progammed I/O.  Likewise with
<machine/bus_mem.h>.  This would allow the bus_space_read_1, et al, to
be highly optimized for devices that could optimize them.  He told me
in private conversations that this made a big difference in the
performance he was seeing in the ahc driver at the time.  I don't know
how big big is, however.  I know there was also optimizations in the
api to bus_space_dma to allow things to work better on specialized
hardware.

I changed this a little to default to using both memory and progammed
I/O if neither was included to be more compatible with NetBSD and
newconfig.

Warner


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-new-bus" in the body of the message