From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933652AbXCMVgO (ORCPT ); Tue, 13 Mar 2007 17:36:14 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S933664AbXCMVgO (ORCPT ); Tue, 13 Mar 2007 17:36:14 -0400 Received: from gw.goop.org ([64.81.55.164]:57539 "EHLO mail.goop.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933652AbXCMVgN (ORCPT ); Tue, 13 Mar 2007 17:36:13 -0400 Message-ID: <45F7194B.5080705@goop.org> Date: Tue, 13 Mar 2007 14:36:11 -0700 From: Jeremy Fitzhardinge User-Agent: Thunderbird 1.5.0.10 (X11/20070302) MIME-Version: 1.0 To: Matt Mackall CC: David Miller , nickpiggin@yahoo.com.au, akpm@linux-foundation.org, clameter@sgi.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: [QUICKLIST 0/4] Arch independent quicklists V2 References: <20070313200313.GG10459@waste.org> <45F706BC.7060407@goop.org> <20070313202125.GO10394@waste.org> <20070313.140722.72711732.davem@davemloft.net> <20070313211435.GP10394@waste.org> In-Reply-To: <20070313211435.GP10394@waste.org> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Matt Mackall wrote: > Well you -could- do this: > > - reuse a long in struct page as a used map that divides the page up > into 32 or 64 segments > - every time you set a PTE, set the corresponding bit in the mask > - when we zap, only visit the regions set in the mask > > Thus, you avoid visiting most of a PMD page in the sparse case, > assuming PTEs aren't evenly spread across the PMD. > > This might not even be too horrible as the appropriate struct page > should be in cache with the appropriate bits of the mm already locked, > etc. > And do the same in pte pages for actual mapped pages? Or do you think they would be too densely populated for it to be worthwhile? J