From mboxrd@z Thu Jan 1 00:00:00 1970 From: troy.kisky@boundarydevices.com (Troy Kisky) Date: Tue, 14 Aug 2012 11:09:58 -0700 Subject: [PATCH] pinctrl: imx5: start numbering pad from 0 In-Reply-To: References: <1344869278-27334-1-git-send-email-shawn.guo@linaro.org> <502954D2.908@boundarydevices.com> Message-ID: <502A9476.1040500@boundarydevices.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 8/14/2012 12:30 AM, Dong Aisheng wrote: > On 14 August 2012 03:26, Troy Kisky wrote: >> On 8/13/2012 7:47 AM, Shawn Guo wrote: >>> Unlike imx6q pinctrl driver that starts nubmering pad from 0, imx5 >>> pinctrl drivers number pad from 1. It causes problem/confusion when >>> driver accesses imx51_pinctrl_pads array using pin ID as the index. >>> >>> Change imx51_pads and imx53_pads numbering start from 0. >>> >>> Signed-off-by: Shawn Guo >>> --- >>> drivers/pinctrl/pinctrl-imx51.c | 490 >>> +++++++++++++++++++------------------- >>> drivers/pinctrl/pinctrl-imx53.c | 402 ++++++++++++++++---------------- >>> 2 files changed, 446 insertions(+), 446 deletions(-) >>> >>> diff --git a/drivers/pinctrl/pinctrl-imx51.c >>> b/drivers/pinctrl/pinctrl-imx51.c >>> index 9fd0216..fb84689 100644 >>> --- a/drivers/pinctrl/pinctrl-imx51.c >>> +++ b/drivers/pinctrl/pinctrl-imx51.c >>> @@ -23,251 +23,251 @@ >>> #include "pinctrl-imx.h" >>> enum imx51_pads { >>> - MX51_PAD_EIM_D16 = 1, >>> - MX51_PAD_EIM_D17 = 2, > ... >>> + MX53_PAD_GPIO_19 = 0, >>> + MX53_PAD_KEY_COL0 = 1, >> >> Why not skip the = xx altogether?? The enum will auto-increment. > Personally i'd like to keep it. > The reason is that pin id is basic property of a pin per pinctrl > subsystem's design > so explicitly define it looks more clear to me and i'm not sure but > it's possible that > the pin id may be used in device tree in the future(maybe some other > soc already uses it), > And defining it has no big harming. > > Regards > Dong Aisheng > . > Then maybe #defines should be used, so that device tree can include this file and be run through C pre-processor? Thanks Troy