From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756416AbYGDQsD (ORCPT ); Fri, 4 Jul 2008 12:48:03 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753725AbYGDQrx (ORCPT ); Fri, 4 Jul 2008 12:47:53 -0400 Received: from mail.suse.de ([195.135.220.2]:56349 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753491AbYGDQrw (ORCPT ); Fri, 4 Jul 2008 12:47:52 -0400 Date: Fri, 04 Jul 2008 18:47:49 +0200 Message-ID: From: Takashi Iwai To: Dmitry Baryshkov Cc: Stephen Rothwell , David Brownell , linux-next@vger.kernel.org, LKML Subject: Re: [PATCH] fix build error of driver/mfd/tc6393xb.c In-Reply-To: <20080704152708.GA12382@doriath.ww600.siemens.net> References: <20080704114755.GA21747@doriath.ww600.siemens.net> <20080704124256.GB21747@doriath.ww600.siemens.net> <20080704132234.GA4576@doriath.ww600.siemens.net> <20080704142208.GB4576@doriath.ww600.siemens.net> <20080704152708.GA12382@doriath.ww600.siemens.net> User-Agent: Wanderlust/2.12.0 (Your Wildest Dreams) SEMI/1.14.6 (Maruoka) FLIM/1.14.7 (=?ISO-8859-4?Q?Sanj=F2?=) APEL/10.6 Emacs/22.2 (x86_64-suse-linux-gnu) MULE/5.0 (SAKAKI) MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org At Fri, 4 Jul 2008 19:27:08 +0400, Dmitry Baryshkov wrote: > > On Fri, Jul 04, 2008 at 05:24:52PM +0200, Takashi Iwai wrote: > > At Fri, 4 Jul 2008 18:22:08 +0400, > > Dmitry Baryshkov wrote: > > > > > > On Fri, Jul 04, 2008 at 03:59:54PM +0200, Takashi Iwai wrote: > > > > At Fri, 4 Jul 2008 17:22:35 +0400, > > > > Dmitry Baryshkov wrote: > > > > > > > > > > On Fri, Jul 04, 2008 at 02:59:18PM +0200, Takashi Iwai wrote: > > > > > > At Fri, 4 Jul 2008 16:42:56 +0400, > > > > > > Dmitry Baryshkov wrote: > > > > > > > > > > > > > > On Fri, Jul 04, 2008 at 02:04:30PM +0200, Takashi Iwai wrote: > > > > > > > > At Fri, 4 Jul 2008 15:47:55 +0400, > > > > > > > > Dmitry Baryshkov wrote: > > > > > > > > > > > > > > > > > > On Fri, Jul 04, 2008 at 01:20:42PM +0200, Takashi Iwai wrote: > > > > > > > > > > Fix the build of tc6393xb due to missing definition of struct > > > > > > > > > > gpio_chip. > > > > > > > > > > > > > > > > > > > > Signed-off-by: Takashi Iwai > > > > > > > > > > > > > > > > > > > > --- > > > > > > > > > > diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig > > > > > > > > > > index 9f93c29..2981da4 100644 > > > > > > > > > > --- a/drivers/mfd/Kconfig > > > > > > > > > > +++ b/drivers/mfd/Kconfig > > > > > > > > > > @@ -44,7 +44,7 @@ config HTC_PASIC3 > > > > > > > > > > > > > > > > > > > > config MFD_TC6393XB > > > > > > > > > > bool "Support Toshiba TC6393XB" > > > > > > > > > > - depends on HAVE_GPIO_LIB > > > > > > > > > > + depends on HAVE_GPIO_LIB && GENERIC_GPIO > > > > > > > > > > select MFD_CORE > > > > > > > > > > help > > > > > > > > > > Support for Toshiba Mobile IO Controller TC6393XB > > > > > > > > > > > > > > > > > > Which configuration does provide HAVE_GPIO_LIB but not GENERIC_GPIO? > > > > > > > > > > > > > > > > The below is the config with which I got the error. > > > > > > > > > > > > > > The proper fix probably is to make PPC_MPC5200_GPIO also select > > > > > > > GENERIC_GPIO. I don't think that building GPIO_LIB w/o GENERIC_GPIO > > > > > > > should be supported. > > > > > > > > > > > > [Added David to Cc] > > > > > > > > > > > > Then HAVE_GPIO_LIB should depend on GENERIC_GPIO. > > > > > > Or, linux/gpio.h should be fixed so that the code can be built > > > > > > properly without GENERIC_GPIO. > > > > > > > > > > IIRC selected symbols don't propagate their depends. > > > > > > > > Then HAVE_GPIO_LIB selects GENERIC_GPIO. Right now, it's a missing > > > > dependency. > > > > > > Please submit a patch where HAVE_GPIO_LIB selects GENERIC_GPIO. > > > > Hmm... just checking again, I found GENERIC_GPIO isn't defined in > > drivers/gpio/Kconfig but in each arch Kconfig. So we can't select it > > unconditionally. > > > > I'm not sure whether it's a better fix at this moment. > > Then we probably should move GENERIC_GPIO to drivers/gpio/Kconfig, allow > it to be selected from arch code which doesn't know anything about > gpiolib and allow it to be selected by HAVE_GPIO_LIB? Seconded. But, I'd like to have a comment from GPIO guys before moving on. Takashi