From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Wed, 1 Jan 2003 14:48:57 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Wed, 1 Jan 2003 14:48:57 -0500 Received: from dsl-67-48-44-237.telocity.com ([67.48.44.237]:46646 "EHLO lnuxlab.ath.cx") by vger.kernel.org with ESMTP id ; Wed, 1 Jan 2003 14:48:53 -0500 Date: Wed, 1 Jan 2003 15:14:51 -0500 To: Sowmya Adiga Cc: linux-kernel@vger.kernel.org Subject: Re: 2.5.53mm2 for AIMbench Message-ID: <20030101201451.GA3397@lnuxlab.ath.cx> References: <002c01c2b14b$0ecd8a10$6009720a@wipro.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <002c01c2b14b$0ecd8a10$6009720a@wipro.com> User-Agent: Mutt/1.3.28i From: khromy@lnuxlab.ath.cx (khromy) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Jan 01, 2003 at 09:35:37AM +0530, Sowmya Adiga wrote: > Hi, > > I gave 60 second for each test while running AIM bench for > 2.5.53mm2 patch.But it ran each test for only 6 second.Is there any > change in kernel frequency with this release? > > Regards > sowmya adiga I had some timing problems also but Andrew pointed out the following: Seems that this is because different parts of the kernel are using different values of HZ (!). In include/asm-i386/param.h, please add: #ifdef __KERNEL__ +#include #ifdef CONFIG_1000HZ Here is a patch for that change: --- linux-2.5/include/asm-i386/param.h.bak 2002-12-30 00:39:35.000000000 -0500 +++ linux-2.5/include/asm-i386/param.h 2002-12-30 00:40:18.000000000 -0500 @@ -3,6 +3,8 @@ #ifdef __KERNEL__ +#include + #ifdef CONFIG_1000HZ # define HZ 1000 /* Internal kernel timer frequency */ #else -- L1: khromy ;khromy(at)lnuxlab.ath.cx