From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 14 Mar 2002 06:54:48 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 14 Mar 2002 06:54:38 -0500 Received: from mail.pha.ha-vel.cz ([195.39.72.3]:24329 "HELO mail.pha.ha-vel.cz") by vger.kernel.org with SMTP id ; Thu, 14 Mar 2002 06:54:26 -0500 Date: Thu, 14 Mar 2002 12:54:20 +0100 From: Vojtech Pavlik To: linux-kernel@vger.kernel.org Subject: [patch] Big IDE chipset driver update, please test! Message-ID: <20020314125420.A20797@ucw.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi! After Martin Dalecki poked me to start doing something I was planning long ago with the IDE chipset drivers, I indeed started. And the work slowly evolved in a quite big change. Here http://twilight.ucw.cz/ide-via-amd-piix-efar-nvidia-timing-11.diff is a patch that (among other things): * changes all the chipset drivers to use the new IDE mode-timing routines * replaces the Intel IDE driver completely, adding support for PIIX3 * removes the slc90e66 Efar Victory66 driver, as that chip is now handled by the Intel PIIX driver * replaces the AMD ide driver completely, adding support for AMD-8111 and nVidia nForce * fixes minor bugs and cleans up stuff here and there Please test this namely: If you have an Intel PIIX, PIIX3, ICH2 or ICH3 chip, or if you have the nVidia nForce or the Efar Victory66. I'm really interested if it works correctly on those. On both the Intel and AMD chipsets the driver is expected to give slightly (< 5%) better performance due to timing the transfers closer to the spec. -- Vojtech Pavlik SuSE Labs