From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Subject: Re: [PATCH 3/5] ACPI: Bind implicit GPE dependencies to PCI devices Date: Wed, 6 Oct 2010 01:37:29 +0200 Message-ID: <201010060137.29369.rjw@sisk.pl> References: <1286216549-5438-1-git-send-email-mjg@redhat.com> <1286216549-5438-4-git-send-email-mjg@redhat.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-2" Content-Transfer-Encoding: 7bit Return-path: Received: from ogre.sisk.pl ([217.79.144.158]:48419 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756357Ab0JEXij (ORCPT ); Tue, 5 Oct 2010 19:38:39 -0400 In-Reply-To: <1286216549-5438-4-git-send-email-mjg@redhat.com> Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: Matthew Garrett Cc: linux-acpi@vger.kernel.org, linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org On Monday, October 04, 2010, Matthew Garrett wrote: > Several platforms provide GPE information about devices in ACPI, but provide > no ACPI methods to handle these at runtime. Provide a default handler for > this case and bind it to PCI devices as they're discovered in ACPI space. > Keep the methods associated in case they have side effects. I think I'd prefer it if ACPICA maintained a list of devices to Notify() if the given GPE triggers. That would require us to redefine acpi_gpe_can_wake() a bit, but that should be generally more efficient. Thanks, Rafael