From mboxrd@z Thu Jan 1 00:00:00 1970 From: Al Viro Subject: Re: [PATCH] qla1820 iomem annotations Date: Thu, 21 Oct 2004 18:30:57 +0100 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <20041021173057.GC24336@parcelfarce.linux.theplanet.co.uk> References: <20041020235837.GC23987@parcelfarce.linux.theplanet.co.uk> <20041021090032.GA1941@infradead.org> <20041021151406.GB24336@parcelfarce.linux.theplanet.co.uk> <20041021172547.GA8242@infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:45727 "EHLO www.linux.org.uk") by vger.kernel.org with ESMTP id S270732AbUJURa6 (ORCPT ); Thu, 21 Oct 2004 13:30:58 -0400 Content-Disposition: inline In-Reply-To: <20041021172547.GA8242@infradead.org> List-Id: linux-scsi@vger.kernel.org To: Christoph Hellwig Cc: Linus Torvalds , linux-scsi@vger.kernel.org On Thu, Oct 21, 2004 at 06:25:47PM +0100, Christoph Hellwig wrote: > On Thu, Oct 21, 2004 at 04:14:06PM +0100, Al Viro wrote: > > On Thu, Oct 21, 2004 at 10:00:32AM +0100, Christoph Hellwig wrote: > > > This breaks old-style PIO support in this driver which is required for the > > > Visual Workstations. > > > > Where? __iomem is #defined to nothing for gcc. So that leaves only > > the chunk below as theoretically possible source of changes, right? > > Sorry, this was totlly misworked. What I actually meant is that it gives > the sparse warnings again in the PIO-config, so you trade sparse warnings > for one variant vs ones for another variant. Huh? We _already_ have these guys as pointers; so PIO variants were not any happier with non-iomem ones. Details, please...