From mboxrd@z Thu Jan  1 00:00:00 1970
From: Hans de Goede <hdegoede@redhat.com>
Date: Tue, 10 Mar 2015 12:38:45 +0100
Subject: [U-Boot] [PATCH 1/4] sunxi: video: Fix VIDEO_LCD_PANEL_I2C
 being enabled by default
In-Reply-To: <1425805488.27083.24.camel@hellion.org.uk>
References: <1425737102-930-1-git-send-email-hdegoede@redhat.com>
 <1425805488.27083.24.camel@hellion.org.uk>
Message-ID: <54FED7C5.7090007@redhat.com>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

Hi,

On 08-03-15 10:04, Ian Campbell wrote:
> On Sat, 2015-03-07 at 15:04 +0100, Hans de Goede wrote:
>> Fix a type in board/sunxi/Kconfig which caused VIDEO_LCD_PANEL_I2C to be
>
> "typo"

Heh, so I made a typo in the word typo, fixed :)

>> +#define CONFIG_VIDEO_LCD_I2C_BUS	1 /* NA, but necessary to compile */
>
> Hrm, should the usage sites not be either ifdef'd or excluded at the
> Makefile level when VIDEO_LCD_PANEL_I2C is disabled?
>
> The only use is in
>                  if (IS_ENABLED(CONFIG_VIDEO_LCD_TL059WV5C0)) {
>                          unsigned int orig_i2c_bus = i2c_get_bus_num();
>                          i2c_set_bus_num(CONFIG_VIDEO_LCD_I2C_BUS);
>                          i2c_reg_write(0x5c, 0x04, 0x42); /* Turn on the LCD */
>                          i2c_set_bus_num(orig_i2c_bus);
>                  }
>
> Is the issue that the IS_ENABLED statically false but the compiler still
> wants the contents to be valid?

Right.

> How about a helper set_video_i2c_bus or some such which can be a nop if
> CONFIG_VIDEO_LCD_I2C_BUS is not defined, which would keep the ifdef out
> of this code?

Not a fan of that, the whole purpose of using IS_ENABLED is to have easier
to read code, I do not believe that adding a wrapper helps there.

> Or at least #define it to some obviously bogus value (e.g. ~0UL).

That is a good idea, I've changed it to -1 in my personal tree.

Regards,

Hans