From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1161101AbWAHTUr (ORCPT ); Sun, 8 Jan 2006 14:20:47 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1161102AbWAHTUr (ORCPT ); Sun, 8 Jan 2006 14:20:47 -0500 Received: from relay02.mail-hub.dodo.com.au ([202.136.32.45]:64926 "EHLO relay02.mail-hub.dodo.com.au") by vger.kernel.org with ESMTP id S1161101AbWAHTUq (ORCPT ); Sun, 8 Jan 2006 14:20:46 -0500 From: Grant Coady To: Jan Engelhardt Cc: Bernd Eckenfels , linux-kernel@vger.kernel.org Subject: Re: Why is 2.4.32 four times faster than 2.6.14.6?? Date: Mon, 09 Jan 2006 06:20:41 +1100 Organization: http://bugsplatter.mine.nu/ Reply-To: gcoady@gmail.com Message-ID: References: In-Reply-To: X-Mailer: Forte Agent 2.0/32.652 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 8 Jan 2006 13:04:22 +0100 (MET), Jan Engelhardt wrote: > >>grant@deltree:~$ time grep -v 192\.168\. /var/log/apache/access_log |cut -c-96 >> >>real 0m1.671s >>user 0m0.550s >>sys 0m0.300s >>grant@deltree:~$ time grep -v 192\.168\. /var/log/apache/access_log |cut -c-96 >/dev/null >> >>real 0m0.510s >>user 0m0.420s >>sys 0m0.080s > >Given that the first command is the first one accessing access_log at >all, then: the second time, access_log is already cached and >therefore can be accessed faster. I did repeat measurements to check for variation due to caching, and that is not what is happening, we comparing going out over ssh terminal to dumping output locally. Grant.