Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 14 Sep 2001 10:14:38 -0700
From:      Marcel Moolenaar <marcel@xcllnt.net>
To:        KATO Takenori <kato@nendai.nagoya-u.ac.jp>
Cc:        imp@harmony.village.org, ru@FreeBSD.org, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org
Subject:   Re: hw.machine vs hw.machine_arch
Message-ID:  <20010914101438.C465@dhcp01.pn.xcllnt.net>
In-Reply-To: <200109140914.f8E9EZO87758@tatu.nendai.nagoya-u.ac.jp>
References:  <200109131133.f8DBXoO84286@tatu.nendai.nagoya-u.ac.jp> <200109140648.f8E6mFt19413@harmony.village.org> <20010914012406.B16118@dhcp01.pn.xcllnt.net> <200109140914.f8E9EZO87758@tatu.nendai.nagoya-u.ac.jp>

next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Sep 14, 2001 at 06:14:30PM +0900, KATO Takenori wrote:
> From: Marcel Moolenaar <marcel@xcllnt.net>
> Subject: Re: hw.machine vs hw.machine_arch
> Date: Fri, 14 Sep 2001 01:24:06 -0700
> 
> > Both MIBs (ie hw.machine and machdep.ispc98) serve the same purpose
> > as far as I can see, so we should be able to remove one of them
> > (preferably machdep.ispc98).
> 
> Both machine and machine_arch are `i386' on PC-98 to avoid
> problem of third-party applications.

Do you know which applications?

> If a program uses the machine
> instead of the machine_arch, that program can not be used on PC-98
> box.

I don't expect that there are a lot of programs that explicitly
query hw.machine and bail out if it's not some fixed value. And,
technically speaking, it's the programs that are broken and not
the kernel, right?

> It happens not only when a program uses sysctl, also a program
> uses uname (uname returns machine instead of machine_arch.)

Yes, it can be argued that with the definitions of hw.machine_arch
and hw.machine, uname -m should really be returning hw.machine_arch

What I see at this time is a perfectly good definition of the
MIBs, but in the only real life example we have it's ignored and
violated and a third sysctl variable is needed (machdep.ispc98)
to compensate for the loss of information. I think something is
broken, whether that's the implementation or the definition is
open for debate :-)

-- 
 Marcel Moolenaar	  USPA: A-39004		 marcel@xcllnt.net

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010914101438.C465>