From mboxrd@z Thu Jan  1 00:00:00 1970
From: Pat LaVarre <p.lavarre@ieee.org>
Subject: big-endian udfct_1_0r2
Date: 08 Oct 2003 12:09:04 -0600
Sender: linux-fsdevel-owner@vger.kernel.org
Message-ID: <1065636544.7602.49.camel@patehci2>
References: <1065553341.8172.45.camel@patehci2>
	 <20031007204951.A25423@falcon.csc.calpoly.edu>
	 <1065631289.6625.34.camel@patehci2>  <1065631668.6625.48.camel@patehci2>
	 <1065635511.7602.38.camel@patehci2>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Cc: linux_udf@hpesjro.fc.hp.com
Return-path: <linux-fsdevel-owner@vger.kernel.org>
Received: from email-out1.iomega.com ([147.178.1.82]:50390 "EHLO
	email.iomega.com") by vger.kernel.org with ESMTP id S261689AbTJHSJR
	(ORCPT <rfc822;linux-fsdevel@vger.kernel.org>);
	Wed, 8 Oct 2003 14:09:17 -0400
To: linux-fsdevel@vger.kernel.org
In-Reply-To: <1065635511.7602.38.camel@patehci2>
List-Id: linux-fsdevel.vger.kernel.org

> http://www.extra.research.philips.com/udf/download.html
> For big endian platforms,
> ENDIAN_SWAP must be defined at compilation time.

Anyone here already working on fixing that feature?

I'm too new myself to know how socially correct C code in linux does
distinguish big-endian from little-endian.  Perhaps the first fragment
that leaps to my mind is:

static inline int lil() { int const i = 1; return *(char const *)&i; }

I'm too new to have tried that fragment in many gcc and linux.  I would
be disappointed, though not really surprised, to discover that the gcc
of wherever we see udf disks in fact doesn't support that kind of
constant as well as a #define constant.  Could be that an initialised
const union of char with int might work better.

Pat LaVarre