From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Mon, 30 Sep 2002 23:56:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Mon, 30 Sep 2002 23:56:52 -0400 Received: from e33.co.us.ibm.com ([32.97.110.131]:30627 "EHLO e33.co.us.ibm.com") by vger.kernel.org with ESMTP id ; Mon, 30 Sep 2002 23:56:51 -0400 Date: Tue, 1 Oct 2002 09:43:58 +0530 From: Maneesh Soni To: Andrew Morton Cc: linux-kernel@vger.kernel.org Subject: Re: 2.5.39-mm1 Message-ID: <20021001094358.E13877@in.ibm.com> Reply-To: maneesh@in.ibm.com References: <3D9804E1.76C9D4AE@digeo.com> <3D9896F6.8E584DC5@digeo.com> <200210010346.g913ktfP148022@northrelay01.pok.ibm.com> <3D991BD4.1191F6C6@digeo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <3D991BD4.1191F6C6@digeo.com>; from akpm@digeo.com on Mon, Sep 30, 2002 at 08:51:48PM -0700 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Sep 30, 2002 at 08:51:48PM -0700, Andrew Morton wrote: > Maneesh Soni wrote: > > > > On Mon, 30 Sep 2002 23:55:50 +0530, Andrew Morton wrote: > > > > > "Martin J. Bligh" wrote: > > >> > > >> Which looks about the same to me? Me slightly confused. > > > > > > I expect that with the node-local allocations you're not getting a lot > > > of benefit from the lock amortisation. Anton will. > > > > > > It's the lack of improvement of cache-niceness which is irksome. Perhaps > > > the heuristic should be based on recency-of-allocation and not > > > recency-of-freeing. I'll play with that. > > > > > >> Will try > > >> adding the original hot/cold stuff onto 39-mm1 if you like? > > > > > > Well, it's all in the noise floor, isn't it? Better off trying broader > > > tests. I had a play with netperf and the chatroom benchmark. But the > > > latter varied from 80,000 msgs/sec up to 350,000 between runs. -- > > > > Hello Andrew, > > > > chatroom benchmark gives more consistent results with some delay > > (sleep 60) between two runs. > > > > oh. Thanks. Why? Could be because of sockets not getting closed immediately. I see them in TIME_WAIT state right after the run. Maneesh -- Maneesh Soni IBM Linux Technology Center, IBM India Software Lab, Bangalore. Phone: +91-80-5044999 email: maneesh@in.ibm.com http://lse.sourceforge.net/