From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263645AbTLHVCS (ORCPT ); Mon, 8 Dec 2003 16:02:18 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263647AbTLHVCR (ORCPT ); Mon, 8 Dec 2003 16:02:17 -0500 Received: from holomorphy.com ([199.26.172.102]:52701 "EHLO holomorphy.com") by vger.kernel.org with ESMTP id S263645AbTLHVCO (ORCPT ); Mon, 8 Dec 2003 16:02:14 -0500 Date: Mon, 8 Dec 2003 13:02:01 -0800 From: William Lee Irwin III To: Per Andreas Buer Cc: linux-kernel@vger.kernel.org Subject: Re: 2.4: mylex and > 2GB RAM Message-ID: <20031208210201.GP8039@holomorphy.com> Mail-Followup-To: William Lee Irwin III , Per Andreas Buer , linux-kernel@vger.kernel.org References: <1070897058.25490.56.camel@netstat.linpro.no> <20031208153641.GJ8039@holomorphy.com> <1070898870.25490.76.camel@netstat.linpro.no> <20031208162214.GW19856@holomorphy.com> <20031208202229.GO8039@holomorphy.com> <1070917304.1260.44.camel@localhost.localdomain> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1070917304.1260.44.camel@localhost.localdomain> Organization: The Domain of Holomorphy User-Agent: Mutt/1.5.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 2003-12-08 at 21:22, William Lee Irwin III wrote: >> It could potentially slow it down a lot more than a few percent. >> The main effect you would see is heavy low memory consumption (LowFree: >> going down to almost nothing) and very heavy cpu consumption. On Mon, Dec 08, 2003 at 10:01:45PM +0100, Per Andreas Buer wrote: > Right on. LowFree drops and when it reaches almost nothing the system > more or less goes freezes. > The DAC960 driver is not a SCSI driver so this means that there is > something wrong with the PCI-DMA transfers, right? > Replacing the DAC960 with another RAID-kontroller will not help because > it will use the same PCI-DMA transfers, right? Any hints on how I can > mend this? Actually, this suggests lowmem starvation due to bounce buffering. -- wli