From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750768AbWFGDEB (ORCPT ); Tue, 6 Jun 2006 23:04:01 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750776AbWFGDEB (ORCPT ); Tue, 6 Jun 2006 23:04:01 -0400 Received: from peabody.ximian.com ([130.57.169.10]:65171 "EHLO peabody.ximian.com") by vger.kernel.org with ESMTP id S1750768AbWFGDEB (ORCPT ); Tue, 6 Jun 2006 23:04:01 -0400 Subject: Re: [PATCH 8/9] PCI PM: clear IO and MEM when disabling a device From: Adam Belay To: Benjamin Herrenschmidt Cc: Greg KH , Andrew Morton , linux-kernel@vger.kernel.org, linux-pci@atrey.karlin.mff.cuni.cz In-Reply-To: <1149553936.559.6.camel@localhost.localdomain> References: <1149497176.7831.162.camel@localhost.localdomain> <1149553936.559.6.camel@localhost.localdomain> Content-Type: text/plain Date: Tue, 06 Jun 2006 23:16:36 -0400 Message-Id: <1149650196.7831.229.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.4.2.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2006-06-06 at 10:32 +1000, Benjamin Herrenschmidt wrote: > On Mon, 2006-06-05 at 04:46 -0400, Adam Belay wrote: > > This patch modifies pci_disable_device() to clear IO and MEM from the > > COMMAND PCI config register. This is required before entering D3, but > > also probably a good general practice for system suspend. > > And will break a great deal of platforms. Don't do it :) The problem is, > I think, mostly related to firmware issues. > > Ben. If the device must be enabled before passing control to firmware, then shouldn't its controlling driver never call pci_disable_device()? I'd like to have pci_disable_device() actually disable devices, as this is needed for D3. Thanks, Adam