From mboxrd@z Thu Jan 1 00:00:00 1970 From: Takashi Iwai Subject: Re: [PATCH] fix build error of driver/mfd/tc6393xb.c Date: Mon, 07 Jul 2008 18:09:16 +0200 Message-ID: References: <200807040954.34838.david-b@pacbell.net> <20080705083122.GA7026@doriath.ww600.siemens.net> <200807051013.10078.david-b@pacbell.net> Mime-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Return-path: Received: from ns.suse.de ([195.135.220.2]:44221 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753413AbYGGQJV (ORCPT ); Mon, 7 Jul 2008 12:09:21 -0400 In-Reply-To: <200807051013.10078.david-b@pacbell.net> Sender: linux-next-owner@vger.kernel.org List-ID: To: David Brownell Cc: Dmitry Baryshkov , Stephen Rothwell , linux-next@vger.kernel.org, LKML At Sat, 5 Jul 2008 10:13:09 -0700, David Brownell wrote: > > On Saturday 05 July 2008, Dmitry Baryshkov wrote: > > > > > > I'm not sure whether it's a better fix at this moment. > > > > > > Which platform is it which is trying to declare that it has > > > an implementation (HAVE_GPIO_LIB) of an interface that it > > > doesn't support (GENERIC_GPIO unset) ?? > > > > the powerpc with PPC_MPC52xx=y, PPC_MPC5200_GPIO=y > > The most natural fix is to have the Kconfig for PPC_MPC52xx > select both GENERIC_GPIO and HAVE_GPIO_LIB. Yes. I, however, still feel uneasy by the fact that the combination GENERIC_GPIO=n and HAVE_GPIO_LIB=y doesn't work. We should either fix it to work or disallow the combination. thanks, Takashi