From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: iomapping a big endian area Date: Tue, 05 Apr 2005 09:05:32 -0500 Message-ID: <1112709932.5764.6.camel@mulgrave> References: <1112475134.5786.29.camel@mulgrave> <20050403013757.GB24234@parcelfarce.linux.theplanet.co.uk> <1112649465.5813.106.camel@mulgrave> <20050405072104.GA26208@infradead.org> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from stat16.steeleye.com ([209.192.50.48]:27778 "EHLO hancock.sc.steeleye.com") by vger.kernel.org with ESMTP id S261741AbVDEOFj (ORCPT ); Tue, 5 Apr 2005 10:05:39 -0400 In-Reply-To: <20050405072104.GA26208@infradead.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Christoph Hellwig Cc: Matthew Wilcox , SCSI Mailing List , Linux Kernel On Tue, 2005-04-05 at 08:21 +0100, Christoph Hellwig wrote: > What I really don't like is that you still need ifdefs and wrappers to > support BE and LE wired chips in the same driver. Shouldn't you set the > BE or LE flag at iomap() time and always use the same accessors? No, if you look how it works. On parisc, it can be either BE or LE depending on the chip wiring. There would be a case for selecting BE or LE by #define, but there's no case I know today where we have a driver that would always be BE. James