Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753107AbYAPVlh (ORCPT ); Wed, 16 Jan 2008 16:41:37 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751217AbYAPVl3 (ORCPT ); Wed, 16 Jan 2008 16:41:29 -0500 Received: from palinux.external.hp.com ([192.25.206.14]:41606 "EHLO mail.parisc-linux.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750753AbYAPVl3 (ORCPT ); Wed, 16 Jan 2008 16:41:29 -0500 Date: Wed, 16 Jan 2008 14:41:28 -0700 From: Matthew Wilcox To: Christoph Lameter Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org Subject: Re: SLUB: Increasing partial pages Message-ID: <20080116214127.GA11559@parisc-linux.org> References: <20080116195949.GO18741@parisc-linux.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1713 Lines: 42 On Wed, Jan 16, 2008 at 12:39:31PM -0800, Christoph Lameter wrote: > Ahhh.. Good to hear that the issue on x86_64 gets better. I am still > waiting for a test with the patchset that I did specifically to address > your regression: http://lkml.org/lkml/2007/10/27/245 (where I tried to > come up with an in kernel benchmark that exposes the issue even more)? It > is much more likely now that this patchset in mm addresses your regression > since the hackbench performance improvement fix already reduce it > partially. I sent you a mail on December 6th ... here are the contents of that mail: --- On Thu, Nov 29, 2007 at 07:54:44PM -0800, Christoph Lameter wrote: > Hmmmm... I have been running them for awhile and they have been in mm > for awhile. Never seen a boot issue. After investigation, there are two issues. Patch 8/10 oopses during the run, typically after about an hour. Patch 10/10 oopses during boot. I haven't been able to reproduce this on my quad-core machine, but it happens every time for them on their dual quad-core machine. --- Applying just patches 1-7 and 9 leads to a slight (0.34%) performance reduction compared to slub. That is 6.45% versus slab, reduces to 6.79% with the 8 patches applied. -- Intel are signing my paycheques ... these opinions are still mine "Bill, look, we understand that you're interested in selling us this operating system, but compare it to ours. We can't possibly take such a retrograde step." -- 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/