From owner-freebsd-stable  Sun Jul  4 15:28:12 1999
Delivered-To: freebsd-stable@freebsd.org
Received: from alcanet.com.au (border.alcanet.com.au [203.62.196.10])
	by hub.freebsd.org (Postfix) with ESMTP id 2BD9E14FB3
	for <freebsd-stable@FreeBSD.ORG>; Sun,  4 Jul 1999 15:28:06 -0700 (PDT)
	(envelope-from jeremyp@gsmx07.alcatel.com.au)
Received: by border.alcanet.com.au id <40334>; Mon, 5 Jul 1999 08:10:31 +1000
Date: Mon, 5 Jul 1999 08:27:59 +1000
From: Peter Jeremy <jeremyp@gsmx07.alcatel.com.au>
Subject: Re: MMX
In-reply-to: <377C84BF.33E147F2@cs.strath.ac.uk>
To: freebsd-stable@FreeBSD.ORG
Message-Id: <99Jul5.081031est.40334@border.alcanet.com.au>
Sender: owner-freebsd-stable@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

Roger Hardiman <roger@cs.strath.ac.uk> wrote:
>Here is the almost proper way to determine if a CPU has MMX support.
>I say _alomost_ because it uses the CPU-ID instruction.
>This instruction is not on 386 machines or some early 486 machines.

The Intel 486 manual (that I have at home) includes `official' code
for determining what processor (8086 thru pre-CPUID 80486) you are
running on.  I'd suspect a later version would include testing for
the presence of CPUID.  I believe the code will run in user-mode.

That said, I think the correct approach is a hook to allow a user-mode
program to query the kernel (which has already determined what CPU
it's running on).  I'm not sure whether this should be an extension
of an existing syscall, or a sysctl variable.

Peter


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