From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Brown Subject: Re: [PATCH] Input: ads7846: add regulator support Date: Thu, 4 Feb 2010 18:59:22 +0000 Message-ID: <583BCA92-EE63-48E7-B2EF-D0281E7F676C@opensource.wolfsonmicro.com> References: <1265290758-2035-1-git-send-email-notasas@gmail.com> <20100204142439.GA1139@sirena.org.uk> <6ed0b2681002040652r722e3df2s3f032a9fe0d0c0fe@mail.gmail.com> <20100204162155.GB29982@rakim.wolfsonmicro.main> <20100204180807.GB10965@core.coreip.homeip.net> (sfid-20100204_180836_896233_FC3A8E05) Mime-Version: 1.0 (iPhone Mail 7E18) Content-Type: text/plain; charset=us-ascii; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Return-path: Received: from opensource.wolfsonmicro.com ([80.75.67.52]:53076 "EHLO opensource2.wolfsonmicro.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1757832Ab0BDS7E (ORCPT ); Thu, 4 Feb 2010 13:59:04 -0500 In-Reply-To: <20100204180807.GB10965@core.coreip.homeip.net> (sfid-20100204_180836_896233_FC3A8E05) Sender: linux-input-owner@vger.kernel.org List-Id: linux-input@vger.kernel.org To: Dmitry Torokhov Cc: Grazvydas Ignotas , "linux-input@vger.kernel.org" , "linux-omap@vger.kernel.org" , "lrg@slimlogic.co.uk" On 4 Feb 2010, at 18:08, Dmitry Torokhov wrote: > On Thu, Feb 04, 2010 at 04:21:55PM +0000, Mark Brown wrote: >> On Thu, Feb 04, 2010 at 04:52:26PM +0200, Grazvydas Ignotas wrote: >>> On Thu, Feb 4, 2010 at 4:24 PM, Mark Brown >> >>>> The updates to fix up the boards that need this are fairly >>>> straightforward and given that it's fairly easy to identify systems >>>> which are using the driver in mainline so I'd really prefer not >>>> to go >>>> down the route of trying to carry on in the face of error, it >>>> papers >>>> over stuff now but is not robust in the face of future changes. >> >>> What about warning and continuing only on ENODEV then? I really >>> don't >>> want to be responsible for potentially breaking touchscreen on ~30 >>> boards, where I have no idea about how things are wired up and how >>> the >>> regulators should be set up. >> >> I'm really not terribly enthusiastic about that approach - like I >> say, >> it just moves the problem about a bit and postpones the breakage in a >> potentially more obscure fashion. >> > > Maybe we should rely on platform data to tell us whether we need to > plug > into regulator framework and in this case fail hard on any error? That doesn't really help - it requres per driver code for the platform data and for conditionalising all the regulator API calls, results in inconsistent behaviour between drivers (does this driver have a no regulator switch?) and is looking a bit like the situation where you're passing around the regulator name to match device and supply. It seems like more work overall for users.