Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261729AbUJ1Pcs (ORCPT ); Thu, 28 Oct 2004 11:32:48 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261718AbUJ1Pav (ORCPT ); Thu, 28 Oct 2004 11:30:51 -0400 Received: from gort.metaparadigm.com ([203.117.131.12]:7879 "EHLO gort.metaparadigm.com") by vger.kernel.org with ESMTP id S261729AbUJ1PZm (ORCPT ); Thu, 28 Oct 2004 11:25:42 -0400 Message-ID: <41810FAB.40107@metaparadigm.com> Date: Thu, 28 Oct 2004 23:26:35 +0800 From: Michael Clark Organization: Metaparadigm Pte Ltd User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20041007 Debian/1.7.3-5 X-Accept-Language: en MIME-Version: 1.0 To: Denis Vlasenko Cc: Linux Kernel Subject: Re: 2.6.9 page allocation failure. order:0, mode:0x20 References: <41808419.8070104@metaparadigm.com> <200410281129.09810.vda@port.imtp.ilyichevsk.odessa.ua> In-Reply-To: <200410281129.09810.vda@port.imtp.ilyichevsk.odessa.ua> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1270 Lines: 40 On 10/28/04 16:29, Denis Vlasenko wrote: > On Thursday 28 October 2004 08:31, Michael Clark wrote: > >>BTW - 2.6 is much more responsive than 2.4 while this is all >>going on - i'm just worried about these messages. > > > Which one was faster, and by how much? Both tests compiling 2.6.9 tree with make -j192 bzImage modules (.config posted earlier) from clean source after a reboot. 2CPUs, 2GB RAM, 2GB swap 2.4.27 real 15m38.504s user 21m5.720s sys 3m28.990s peaked at about 1.7GB swap usage 2.6.9 real 14m50.360s user 21m9.008s sys 2m40.580s peaked at 2.0GB swap usage - top said 0K swap free and it survived ;) 2.6.9 was 5% faster (although subjectively almost a magnitude more responsive ie. sshing into the box in the middle of this took about a minute with 2.4.27 and only about 10 or so seconds with 2.6.9, although i didn't time this). Seems 2.6's more proactive swapping helps a bit ie. swap more of the right stuff so as to swap less overall as 2.6 went about 20% deeper into swap. ~mc - 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/