Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751334AbXBVR04 (ORCPT ); Thu, 22 Feb 2007 12:26:56 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751367AbXBVR0z (ORCPT ); Thu, 22 Feb 2007 12:26:55 -0500 Received: from outpipe-village-512-1.bc.nu ([81.2.110.250]:37479 "EHLO lxorguk.ukuu.org.uk" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1751334AbXBVR0y (ORCPT ); Thu, 22 Feb 2007 12:26:54 -0500 Date: Thu, 22 Feb 2007 18:29:39 +0000 From: Alan To: Jeff Garzik Cc: Robert Hancock , akpm@osdl.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] ACPI driver support for pata Message-ID: <20070222182939.56d43a68@lxorguk.ukuu.org.uk> In-Reply-To: <45DDCEC4.7050208@garzik.org> References: <45DD2E12.6000705@shaw.ca> <45DD3F20.7050907@pobox.com> <20070222142439.50cff3e9@lxorguk.ukuu.org.uk> <45DDCEC4.7050208@garzik.org> X-Mailer: Claws Mail 2.7.2 (GTK+ 2.10.8; i386-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1405 Lines: 35 On Thu, 22 Feb 2007 12:11:32 -0500 Jeff Garzik wrote: > Alan wrote: > > ACPI is the only way to do cable handling on the Nvidia PATA chipset. The > > You failed to quote the salient part of the message. Disliking a > separate pata_acpi driver in no way invalidates your statement (quoted > above). If you drive a device by the ACPI interface you don't get to fiddle with it directly or you end up in a murky world of undefined and ungood behaviour. Testing and vendor information both say pata_acpi is the right way to drive Nvidia PATA ports. Cable problem appears to be that - Tejun fixed the drive side detect logic to be stricter and correct - Large numbers of controllers with host side detect don't do drive side detect Changing the code to clip rates only for 40wire or unknown cables improves the behaviour. In the ACPI case I found another corner case with one BIOS which chooses to boot in PIO mode and which therefore fooled the cable detect trick used by pata_acpi. It now does further cable checking in ->mode_filter which combined with the patch in -mm to do the cable logic last means it works for that case too. Alan - 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/