Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757592AbXIFAxU (ORCPT ); Wed, 5 Sep 2007 20:53:20 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756639AbXIFAxM (ORCPT ); Wed, 5 Sep 2007 20:53:12 -0400 Received: from mga05.intel.com ([192.55.52.89]:32455 "EHLO fmsmga101.fm.intel.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1756489AbXIFAxL (ORCPT ); Wed, 5 Sep 2007 20:53:11 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.20,213,1186383600"; d="scan'208";a="294434899" Subject: Re: tbench regression - Why process scheduler has impact on tbench and why small per-cpu slab (SLUB) cache creates the scenario? From: "Zhang, Yanmin" To: Christoph Lameter Cc: LKML , mingo@elte.hu In-Reply-To: References: <1188953218.26438.34.camel@ymzhang> <1188969725.26438.46.camel@ymzhang> <1188983603.26438.55.camel@ymzhang> Content-Type: text/plain; charset=utf-8 Date: Thu, 06 Sep 2007 08:52:19 +0800 Message-Id: <1189039939.26438.65.camel@ymzhang> Mime-Version: 1.0 X-Mailer: Evolution 2.9.2 (2.9.2-2.fc7) Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1764 Lines: 41 On Wed, 2007-09-05 at 03:45 -0700, Christoph Lameter wrote: > On Wed, 5 Sep 2007, Zhang, Yanmin wrote: > > > > > However, the approach treats the slabs in the same policy. Could we > > > > implement a per-slab specific approach like direct b)? > > > > > > I am not sure what you mean by same policy. Same configuration for all > > > slabs? > > Yes. > > Ok. I could add the ability to specify parameters for some slabs. Thanks. That will be more flexible. > > > > Would it be possible to try the two other approaches that I suggested? I > > > think both of those may also solve the issue. Try booting with > > > slab_max_order=0 > > 1) I tried slab_max_order=0 and the regression becomes 12.5%. It's still > > not good. > > > > 2) I apllied patch > > slub-direct-pass-through-of-page-size-or-higher-kmalloc.patch to kernel > > 2.6.23-rc4. The new testing result is much better, only 1% less than > > 2.6.22. I retested 2.6.22 and booted kernel with "slub_max_order=3 slub_min_objects=8". The result is about 8.7% better than without booting parameters. So all with booting parameter "slub_max_order=3 slub_min_objects=8", 2.6.22 is about 5.8% better than 2.6.23-rc4. I suspect process scheduler is responsible for the 5.8% regressions. > > Ok. That seems to indicate that we should improve the alloc path in the > page allocator. The page allocator performance needs to be competitive on > page sized allocations. The problem will be largely going away when we > merge the pass through patch in 2.6.24. - 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/