Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762729AbXHMWxS (ORCPT ); Mon, 13 Aug 2007 18:53:18 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753578AbXHMWw4 (ORCPT ); Mon, 13 Aug 2007 18:52:56 -0400 Received: from netops-testserver-3-out.sgi.com ([192.48.171.28]:53655 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751117AbXHMWwz (ORCPT ); Mon, 13 Aug 2007 18:52:55 -0400 Date: Mon, 13 Aug 2007 15:52:54 -0700 (PDT) From: Christoph Lameter X-X-Sender: clameter@schroedinger.engr.sgi.com To: Andi Kleen cc: Mel Gorman , Lee.Schermerhorn@hp.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: Re: [PATCH 3/4] Embed zone_id information within the zonelist->zones pointer In-Reply-To: <20070813234217.GI3406@bingen.suse.de> Message-ID: References: <200708102013.49170.ak@suse.de> <200708110304.55433.ak@suse.de> <20070813225020.GE3406@bingen.suse.de> <20070813225841.GG3406@bingen.suse.de> <20070813230801.GH3406@bingen.suse.de> <20070813234217.GI3406@bingen.suse.de> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1458 Lines: 36 On Tue, 14 Aug 2007, Andi Kleen wrote: > > I am not sure what you mean by that. Ia64 ZONE_DMA == x86_84 ZONE_DMA32? > > Hmm, when I wrote GFP_DMA32 it was a #define GFP_DMA32 GFP_DMA > on ia64 so that drivers not need to ifdef. Someone nasty > seems to have removed that too. I guess it would be best > to readd. What would be the point? > But then it wouldn't make sense to have GFP_DMA on ia64 and GFP_DMA32 > on x86. Since driver writers are more likely to test on x86 > I would recommend ia64 having compatible semantics. It'll > save everybody trouble long term. This means it wouldn't > help on IA64 machines that don't have a DMA zone -- they > would still need to validate drivers especially -- but at least > the others. There are no compatible semantics. ZONE_DMA may mean something different for each arch depending on its need. An arch may not have a need for a 4GB boundary (such as s390). > Also from my driver review driver authors often seem to have > trouble understanding what GFP_DMA really does. With GFP_DMA32 it > is clearer that it applies to a address range and is not > some synonym for pci_map_* GFP_DMA32 is clear because there are no other arches to muddy up the waters. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/