From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1764634AbYEAT3W (ORCPT ); Thu, 1 May 2008 15:29:22 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1760340AbYEAT3O (ORCPT ); Thu, 1 May 2008 15:29:14 -0400 Received: from relay2.sgi.com ([192.48.171.30]:55721 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1758293AbYEAT3N (ORCPT ); Thu, 1 May 2008 15:29:13 -0400 Date: Thu, 1 May 2008 12:29:13 -0700 (PDT) From: Christoph Lameter X-X-Sender: clameter@schroedinger.engr.sgi.com To: Nick Piggin cc: "Ahmed S. Darwish" , Linux Memory Management List , Linux Kernel Mailing List Subject: Re: [patch] SLQB v2 In-Reply-To: <20080501015418.GC15179@wotan.suse.de> Message-ID: References: <20080410193137.GB9482@wotan.suse.de> <20080415034407.GA9120@ubuntu> <20080501015418.GC15179@wotan.suse.de> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 1 May 2008, Nick Piggin wrote: > > A small question for SLUB devs, would you accept a patch that does > > a similar thing by creating 'slub_page' instead of stuffing slub > > elements (freelist, inuse, ..) in 'mm_types::struct page' unions ? > > I'd like to see that. I have a patch for SLUB, actually. We could do that but then how do we make sure that both definitions stay in sync? So far I have thought that it is clearer if we have one def that shows how objects are overloaded. There is also the overloading of page flags that is now done separately in SLUB. I wonder if that needs to be moved into page-flags.h? Would clarify how page flags are overloaded. If someone inspects the contents of a page struct via debug then it would help if all the possible uses are in one place. If the stuff in tucked away in mm/sl?b.c then its difficult to find.