Received: from nic.funet.fi (nic.funet.fi [128.214.248.6]) by odie.barnet.ac.uk (8.8.2/8.8.0) with ESMTP id WAA31296 for <willy@odie.fluff.org>; Tue, 21 Jan 1997 22:47:44 GMT
Received: from vger.rutgers.edu ([128.6.190.2]) by nic.funet.fi with ESMTP id <66007-26991>; Wed, 22 Jan 1997 00:47:16 +0200
Received: by vger.rutgers.edu id <213285-2094>; Tue, 21 Jan 1997 17:36:12 -0500
Message-Id: <m0vmTsq-0006kEC@eyrie.demon.co.uk>
From: df@eyrie.demon.co.uk (Derek Fawcus)
Subject: Alignment Check (was Re: Arm Linux)
To: linux-arm@vger.rutgers.edu
Date: Tue, 21 Jan 1997 00:16:12 +0100 (GMT)
Cc: pjb27@cam.ac.uk, John.Tytgat@barco.com
In-Reply-To: <Pine.SOL.3.95.970120214519.1139A-100000@hammer.thor.cam.ac.uk> from "Philip Blundell" at Jan 20, 97 09:48:31 pm
X-Mailer: ELM [version 2.4 PL22]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Sender: owner-linux-arm@vger.rutgers.edu
Precedence: bulk
Status: RO
Philip Blundell wrote:
> On Mon, 20 Jan 1997, John Tytgat wrote:
> > On Mon, 20 Jan 1997, Philip Blundell wrote:
> > > else it *can* do. Old ARM machines silently discard the low two address
> > > bits when you ask them to do a word access. This differs, as I said
> >
> > Wouldn't it be the best that a StrongARM owner traces code like this (as
> > the StrongARM can abort on such constructs when configured) ?
>
> Well, yes, and so can the ARM710 when you turn the MMU on (Linux does).
Well, according to the datasheet - so can the ARM610. So basically anyone
with an ARM6 or above can do this.
Probably more useful - (for the debian ppl), is the fact that the 80x86
family (for x >= 4), can also do alignment checks. All one needs to do
is turn on bit 18 (AC) in the EFLAGS register, and then change the masks
that the kernel uses to it won't turn it off again.