From owner-freebsd-questions@FreeBSD.ORG Mon Feb 4 13:40:58 2008 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 4028916A468 for ; Mon, 4 Feb 2008 13:40:58 +0000 (UTC) (envelope-from enno@bashful.metva.com.au) Received: from bashful.metva.com.au (bashful.metva.com.au [202.0.82.3]) by mx1.freebsd.org (Postfix) with ESMTP id D4FA713C4F4 for ; Mon, 4 Feb 2008 13:40:57 +0000 (UTC) (envelope-from enno@bashful.metva.com.au) Received: from bashful.metva.com.au (localhost.metva.com [127.0.0.1]) by bashful.metva.com.au (Postfix) with ESMTP id E5E9533C1A; Tue, 5 Feb 2008 00:40:50 +1100 (EST) Received: (from enno@localhost) by bashful.metva.com.au (8.13.8/8.13.8/Submit) id m14DeoqS032917; Tue, 5 Feb 2008 00:40:50 +1100 (EST) (envelope-from enno) Date: Tue, 5 Feb 2008 00:40:50 +1100 From: Enno Davids To: "Heiko Wundram (Beenic)" Message-ID: <20080204134050.GB1128@bashful.metva.com.au> References: <1563a4fd0802040403x2b71eaa1yd3d8f78e7742843b@mail.gmail.com> <200802041320.14955.wundram@beenic.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200802041320.14955.wundram@beenic.net> User-Agent: Mutt/1.4.2.2i Cc: freebsd-questions@freebsd.org Subject: Re: Endianness of freeBSD X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 04 Feb 2008 13:40:58 -0000 On Mon, Feb 04, 2008 at 01:20:14PM +0100, Heiko Wundram (Beenic) wrote: | |As I said above: it depends on the hardware. There is even hardware (ARM, in |particular) which can run in little- or big-endian mode, depending on how it |is initialized. If I recall correctly some of the MIPS chips had/ve an endian selector bit on each page table entry in the MMU. The idea was to map each of the I/O devices through the MMU and let that bit help in talking to the hardware (i.e. Intel peripherals expecting little endian and motorola chips wanting the opposite.) E.