From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261210AbUBTOcV (ORCPT ); Fri, 20 Feb 2004 09:32:21 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261192AbUBTOcV (ORCPT ); Fri, 20 Feb 2004 09:32:21 -0500 Received: from parcelfarce.linux.theplanet.co.uk ([195.92.249.252]:49082 "EHLO www.linux.org.uk") by vger.kernel.org with ESMTP id S261210AbUBTOcU (ORCPT ); Fri, 20 Feb 2004 09:32:20 -0500 Message-ID: <40361A66.9060209@pobox.com> Date: Fri, 20 Feb 2004 09:32:06 -0500 From: Jeff Garzik User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030703 X-Accept-Language: en-us, en MIME-Version: 1.0 To: MALET JL CC: linux-kernel@vger.kernel.org Subject: Re: [drivers][sata-promise] TX4 has the cache enabled, it should be disabled References: <4031DB3E.8000406@laposte.net> <40354871.1060009@pobox.com> <40360FDD.1070607@laposte.net> In-Reply-To: <40360FDD.1070607@laposte.net> 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 MALET JL wrote: > I relooked into manual (see promise site support, user manual) and > there are two parameters for TX4 : write through/ write back , cache on > cache off..... I don't know which cache is it about but I'll redo some > test (argh you oblige me to reboot to windows! did you thought of my > uptime! LOL )this evening but as far as I remember the optimal values > where write through /no cache..... > on windows with cache on their was small hangs too.... > I noticed then that a prempible kernel is not a good idea in my case : > lot more "small hangs"...... This would be the drive's cache, not the card's cache. Jeff