[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Zv6EwdDybm_I0QSh@google.com>
Date: Thu, 3 Oct 2024 04:49:21 -0700
From: Dmitry Torokhov <dmitry.torokhov@...il.com>
To: Linus Walleij <linus.walleij@...aro.org>
Cc: Arnd Bergmann <arnd@...db.de>, kernel test robot <lkp@...el.com>,
oe-kbuild-all@...ts.linux.dev, linux-kernel@...r.kernel.org,
Javier Martinez Canillas <javierm@...hat.com>
Subject: Re: drivers/input/touchscreen/cyttsp_i2c.c:47:18: error: implicit
declaration of function 'i2c_transfer'
On Wed, Oct 02, 2024 at 03:37:56PM +0200, Linus Walleij wrote:
> On Wed, Oct 2, 2024 at 11:31 AM Dmitry Torokhov
> <dmitry.torokhov@...il.com> wrote:
>
> > > drivers/input/touchscreen/cyttsp_i2c.c:118:26: warning: 'cyttsp_i2c_driver' defined but not used [-Wunused-variable]
> > > 118 | static struct i2c_driver cyttsp_i2c_driver = {
> > > | ^~~~~~~~~~~~~~~~~
> > > cc1: some warnings being treated as errors
> > >
> >
> > This is really weird: cyttsp_i2c.h includes linux/i2c.h, it has
> > dependency in Kconfig on I2C and CONFIG_I2C is present in the .config
> > referenced above, so all the declarations should be there.
>
> I usually ask Arnd about all kind of Kconfig business when
> things get weird.
>
> But could it be one of these awful things:
>
> config TOUCHSCREEN_CYTTSP_I2C
> tristate "support I2C bus connection"
> - depends on TOUCHSCREEN_CYTTSP_CORE && I2C
> + depends on TOUCHSCREEN_CYTTSP_CORE && (I2C=y ||
> I2C=TOUCHSCREEN_CYTTSP_CORE)
>
> Since I2C is tristate it could be a module and then this touchscreen
> has to be a module too.
"depends" should handle this and force touchscreen be a module if I2C is
a module. Also in the provided config I2C is built-in...
Thanks.
--
Dmitry
Powered by blists - more mailing lists