Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264446AbUAaAZb (ORCPT ); Fri, 30 Jan 2004 19:25:31 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264449AbUAaAZb (ORCPT ); Fri, 30 Jan 2004 19:25:31 -0500 Received: from h00a0cca1a6cf.ne.client2.attbi.com ([65.96.182.167]:23189 "EHLO h00a0cca1a6cf.ne.client2.attbi.com") by vger.kernel.org with ESMTP id S264446AbUAaAZN (ORCPT ); Fri, 30 Jan 2004 19:25:13 -0500 Date: Fri, 30 Jan 2004 19:24:12 -0500 From: timothy parkinson To: Bartlomiej Zolnierkiewicz Cc: markus reichelt , Linux Kernel Subject: Re: 2.6.1 "clock preempt"? Message-ID: <20040131002412.GA2293@h00a0cca1a6cf.ne.client2.attbi.com> Mail-Followup-To: Bartlomiej Zolnierkiewicz , markus reichelt , Linux Kernel References: <1074800554.21658.68.camel@cog.beaverton.ibm.com> <20040123203835.GA518@h00a0cca1a6cf.ne.client2.attbi.com> <20040127213026.GA1315@lists.notified.de> <200401310032.23285.bzolnier@elka.pw.edu.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200401310032.23285.bzolnier@elka.pw.edu.pl> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1239 Lines: 36 On Sat, Jan 31, 2004 at 12:32:23AM +0100, Bartlomiej Zolnierkiewicz wrote: > On Tuesday 27 of January 2004 22:30, markus reichelt wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > > Hash: SHA1 > > > > timothy parkinson wrote: > > > * Running with SpeedStep (this is a cpu thing i assume?) could cause > > > this. * Not having DMA enabled on your hard disk(s) could cause this. > > > See the hdparm utility to enable it. > > > * Incorrect TSC synchronization on SMP systems could cause this. > > > * Anything else? > > > > Yepp: > > > > Jan 27 20:12:12 tatooine kernel: Losing too many ticks! > > > > I had to set "CONFIG_IDE_TASK_IOCTL=y" in my .config in order to get > > it working. > > How's that possible? > This config option only exports HDIO_DRIVE_TASKFILE ioctl to user-space. > > --bart > without that, "hdparm -d1 /dev/hda" would claim that dma wasn't supported on my harddrive. i'm positive it's the only variable, i didn't change anything else! timothy - 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/