From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Subject: Re: [PATCH 1/5] ACPI: Allow handlers to be installed at the same time as methods Date: Wed, 6 Oct 2010 21:30:27 +0200 Message-ID: <201010062130.27564.rjw@sisk.pl> References: <1286216549-5438-1-git-send-email-mjg@redhat.com> <4911F71203A09E4D9981D27F9D830858B4893B89@orsmsx503.amr.corp.intel.com> <20101006161854.GA30584@srcf.ucam.org> Mime-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20101006161854.GA30584@srcf.ucam.org> Sender: linux-pci-owner@vger.kernel.org To: Matthew Garrett Cc: "Moore, Robert" , "linux-acpi@vger.kernel.org" , "linux-kernel@vger.kernel.org" , "linux-pci@vger.kernel.org" List-Id: linux-acpi@vger.kernel.org On Wednesday, October 06, 2010, Matthew Garrett wrote: > On Wed, Oct 06, 2010 at 09:14:07AM -0700, Moore, Robert wrote: > > > Although there exists a single Windows document mentions this behavior as a windows "feature", I'm not 100% convinced that this is actually true. AFAIK, we've never seen a machine that depends on an "implicit notify" on a device when a wake GPE happens. > > Such behaviour would be irrelevant for system sleep/wake - the only > requirement is in runtime power management. > > > It appears to me that the biggest issue right now is the fact that a Notify() must be performed on a Device object, and the problem is how to associate the GPE with the device object. > > The other important aspect of this is that a single GPE may correspond > to multiple devices. The methods will generally cope with this by either > sending multiple notifies, executing some SMM code to identify the > relevant device or reading PCI configuration registers to identify the > source of the wakeup. We need to handle that case as well. So, as I said, we can modify acpi_gpe_can_wake() to pass a device object to ACPICA. Then, the device object will be added to the list of devices to Notify() if the GPE is signaled. Plain and simple. Thanks, Rafael