Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751317AbdH1Woy (ORCPT ); Mon, 28 Aug 2017 18:44:54 -0400 Received: from mail-pf0-f178.google.com ([209.85.192.178]:33127 "EHLO mail-pf0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751190AbdH1Wow (ORCPT ); Mon, 28 Aug 2017 18:44:52 -0400 Date: Mon, 28 Aug 2017 15:44:49 -0700 From: Brian Norris To: Rob Herring Cc: Jeffy Chen , "linux-kernel@vger.kernel.org" , "bhelgaas@google.com" , Shawn Lin , Doug Anderson , "devicetree@vger.kernel.org" , Heiko Stuebner , "linux-pci@vger.kernel.org" , "open list:ARM/Rockchip SoC..." , Mark Rutland , "linux-arm-kernel@lists.infradead.org" Subject: Re: [PATCH v4 3/4] dt-bindings: PCI: rockchip: Add support for pcie wake irq Message-ID: <20170828224447.GA3119@google.com> References: <20170822031934.8675-1-jeffy.chen@rock-chips.com> <20170822031934.8675-4-jeffy.chen@rock-chips.com> <20170825181439.d3hlwrcbmcss3sx6@rob-hp-laptop> <20170825181959.GA64040@google.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1322 Lines: 29 On Mon, Aug 28, 2017 at 04:32:55PM -0500, Rob Herring wrote: > On Fri, Aug 25, 2017 at 1:20 PM, Brian Norris wrote: > > On Fri, Aug 25, 2017 at 01:14:39PM -0500, Rob Herring wrote: > >> Use the wakeup source binding: > >> Documentation/devicetree/bindings/power/wakeup-source.txt > > > > And I suppose this means we'd fall under this paragraph? > > > > "However if the devices have dedicated interrupt as the wakeup source > > then they need to specify/identify the same using device specific > > interrupt name. In such cases only that interrupt can be used as wakeup > > interrupt." > > > > We don't expect *any* interrupt to qualify as PCI WAKE#; so we should > > still also document the interrupt name ("wake"?) in > > Documentation/devicetree/bindings/pci/pci.txt as Bjorn suggested, in > > addition to using the 'wakeup-source' property documented there. > > I believe the defined interrupt name is "wakeup" as example 1 shows. That's an example, not a definition. And the definition I quoted literally says "device specific interrupt name". The PCIe specification calls it "WAKE#" all over the place, so I figured that's a good name to use. "wakeup" is also fine I suppose, as long as we document that it must be PCIe WAKE# signal, as per the PCIe specfication. Brian