From: u.kleine-koenig@pengutronix.de (Uwe =?iso-8859-1?Q?Kleine-K=F6nig?=) Date: Sun, 24 Oct 2010 21:20:53 +0200 Subject: [PATCHv5 2/3] ARM: imx: Add gpio-keys to plat-mxc In-Reply-To: <19651.61503.118511.923142@ipc1.ka-ro> References: <1287766575-4092-1-git-send-email-Dinh.Nguyen@freescale.com> <1287766575-4092-2-git-send-email-Dinh.Nguyen@freescale.com> <20101022182812.GA8554@pengutronix.de> <86A0E76937111F4C92FABEC0A2098851051D6090@az33exm21> <20101022194215.GB8554@pengutronix.de> <86A0E76937111F4C92FABEC0A2098851051D6116@az33exm21> <19651.61503.118511.923142@ipc1.ka-ro> Message-ID: <20101024192053.GB25810@pengutronix.de> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Lothar, On Sun, Oct 24, 2010 at 10:37:19AM +0200, Lothar Wa?mann wrote: > Hi, > > >> How do you suggest I define this pin for Babbage board only? Do I add to iomux-mx51.h > > >> #define MX51_BABBAGE_PAD_EIM_A27__GPIO_2_21 IOMUX_PAD(0x45c, 0x0c8, 1, 0x0, > > >0,MX51_GPIO_PAD_CTRL | PAD_CTL_PUS_100K_UP) ? > > >> > > >> Or do I define the pin in the board file with a 100K_UP? Like how eukrea_mbimxsd-baseboard.c did for > > >MBIMXSD_GPIO_3_31? > > >I'd suggest something local to your machine file. Unfortunately I don't > > >see an easy way to reuse the generic MX51_PAD_EIM_A27__GPIO_2_21 to > > >define your local symbol. So I think something like you wrote above > > >should be fine. > > > > > >Too bad that > > > > > > #define MX51_BABBAGE_PAD_EIM_A27__GPIO_2_21 (MX51_PAD_EIM_A27__GPIO_2_21 | PAD_CTL_PUS_100K_UP) > > > > > >doesn't work. > > > > Yes, I also wish I didn't have to redefine the pin just because of PAD setting change. I am going to send out another patch that basically sets the pad_desc.pad_ctrl=PAD_SETTING in mxc_init_board() function. I think it's a bit cleaner than having to redefine the pin. > > > IMO the 'struct pad_desc' should be replaced with an u64 bitmap that > would allow for the pullup and other definitions that may be board > specific to be simply or'ed to the pad definition. > > I could try to provide a patch if this is agreed on. Hmm, I thought about that, too, when writing my mail and when I added the needed space for a struct pad_desc I stopped when it became >64. Somehow I must have got it wrong. After rechecking now it's ==64, so yes, I'd like it. Best regards Uwe -- Pengutronix e.K. | Uwe Kleine-K?nig | Industrial Linux Solutions | http://www.pengutronix.de/ |