Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 7 Aug 2002 16:34:13 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 7 Aug 2002 16:34:12 -0400 Received: from dsl-213-023-022-051.arcor-ip.net ([213.23.22.51]:54955 "EHLO starship") by vger.kernel.org with ESMTP id ; Wed, 7 Aug 2002 16:34:08 -0400 Content-Type: text/plain; charset=US-ASCII From: Daniel Phillips To: Andrew Morton Subject: Re: [PATCH] Rmap speedup Date: Wed, 7 Aug 2002 22:39:15 +0200 X-Mailer: KMail [version 1.3.2] Cc: linux-kernel@vger.kernel.org, wli@holomorphy.com, Rik van Riel References: <3D5177CB.D8CA77C2@zip.com.au> In-Reply-To: MIME-Version: 1.0 Content-Transfer-Encoding: 7BIT Message-Id: Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1113 Lines: 25 On Wednesday 07 August 2002 22:17, Daniel Phillips wrote: > On Wednesday 07 August 2002 21:40, Andrew Morton wrote: > > > Vectoring up the pte chain nodes as > > > you do here doesn't help much because the internal fragmentation > > > roughly equals the reduction in link fields. > > > > Are you sure about that? The vectoring is only a loss for very low > > sharing levels, at which the space consumption isn't a problem anyway. > > At high levels of sharing it's almost a halving. > > Your vector will only be half full on average. Ah, the internal fragmentation only exists in the first node, yes I see. So I'm correct at typical sharing levels, but for massive sharing, yes it approaches half the space, which is significant. That's also when rmap should show the most advantage over virtual scanning, so we really truly do have to benchmark that. -- Daniel - 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/