From owner-cvs-all  Sun Jun 20 11:48: 4 1999
Delivered-To: cvs-all@freebsd.org
Received: by hub.freebsd.org (Postfix, from userid 618)
	id D11BF14C17; Sun, 20 Jun 1999 11:48:02 -0700 (PDT)
Subject: Re: cvs commit: src/sys/isa sio.c
In-Reply-To: <Pine.BSF.4.05.9906201426210.80685-100000@herring.nlsystems.com> from Doug Rabson at "Jun 20, 1999  2:28:22 pm"
To: dfr@nlsystems.com (Doug Rabson)
Date: Sun, 20 Jun 1999 11:48:02 -0700 (PDT)
Cc: peter@FreeBSD.ORG (Peter Wemm), cvs-committers@FreeBSD.ORG,
	cvs-all@FreeBSD.ORG
X-Mailer: ELM [version 2.4ME+ PL43 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1006      
Message-Id: <19990620184802.D11BF14C17@hub.freebsd.org>
From: wpaul@FreeBSD.ORG (Bill Paul)
Sender: owner-cvs-all@FreeBSD.ORG
Precedence: bulk

> On Sun, 20 Jun 1999, Peter Wemm wrote:
> 
> > peter       1999/06/20 06:10:11 PDT
> > 
> >   Modified files:
> >     sys/isa              sio.c 
> >   Log:
> >   Use the proper interfaces to find if a device is enabled, don't dig into
> >   the bus mechanism's private internals!
> 
> Well spotted. For future reference, bus_private.h may not be included by
> any drivers. Any driver which includes this header should be considered
> broken.

Granted an individual driver shouldn't be sticking its grubby mits
in struct device_t, but what about code to implement a custom bus?
When I was experimenting with the miibus stuff, I decided I needed to
enumerate the children of the miibus devclass manually and check the
return values of their probe routines so that I could prioritize
them (which subr_bus won't let me do). This requires bus_private.h.
My plan was to provide routine in sys/dev/miibus/miibus.c to do
this so that I wouldn't need to grope around struct device_t in any
drivers though.

-Bill


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