Re: [patch 2.6.25-rc2-git] make i2c_register_board_info() a NOP when CONFIG_I2C=n

From: Jean Delvare <khali-PUYAD+kWke1g9hUCZPvPmw@public.gmane.org>
To: David Brownell <david-b-yBeKhBN/0LDR7s880joybQ@public.gmane.org>
Cc: i2c-GZX6beZjE8VD60Wz+7aTrA@public.gmane.org
Subject: Re: [patch 2.6.25-rc2-git] make i2c_register_board_info() a NOP when CONFIG_I2C=n
Date: Fri, 22 Feb 2008 12:04:58 +0100	[thread overview]
Message-ID: <20080222120458.28b0d6fe@hyperion.delvare> (raw)
In-Reply-To: <200802220222.18270.david-b-yBeKhBN/0LDR7s880joybQ@public.gmane.org>

On Fri, 22 Feb 2008 02:22:18 -0800, David Brownell wrote:
> On Friday 22 February 2008, Jean Delvare wrote:
> > Hi David,
> > 
> > On Fri, 22 Feb 2008 00:23:16 -0800, David Brownell wrote:
> > > Don't require platform code to be #ifdeffed according to whether
> > > I2C is enabled or not ... if it's not enabled, let GCC compile out
> > > all I2C device declarations.  (Issue noted on an NSLU2 build that
> > > didn't configure I2C.)
> > > 
> > > Signed-off-by: David Brownell <dbrownell-Rn4VEauK+AKRv+LV9MX5uipxlwaOVQ5f@public.gmane.org>
> > > ---
> > >  include/linux/i2c.h |    9 ++++++++-
> > >  1 file changed, 8 insertions(+), 1 deletion(-)
> > > 
> > > --- at91.orig/include/linux/i2c.h	2008-02-21 22:58:53.000000000 -0800
> > > +++ at91/include/linux/i2c.h	2008-02-21 23:02:57.000000000 -0800
> > > @@ -271,9 +271,16 @@ extern void i2c_unregister_device(struct
> > >   * This is done at arch_initcall time, before declaring any i2c adapters.
> > >   * Modules for add-on boards must use other calls.
> > >   */
> > > +#if defined(CONFIG_I2C) || defined(CONFIG_I2C_MODULE)
> > 
> > I'm fairly certain that #ifdef CONFIG_I2C is enough.
> 
> No; if I2C is modular, CONFIG_I2C isn't defined.

Mpff, you're right. I wonder why it was done this way, sounds totally
stupid to me given that in 99% of the cases you don't care if the thing
is modular or not. Oh well.

> Though maybe it could use CONFIG_I2C_BOARDINFO, which
> is defined as boolean.  In fact that's what the I2C
> Makefile uses, but it'd be less obvious here.

Now that you mention it, given that i2c_register_board_info() is
defined in i2c-boardinfo.c, it would indeed make a lot of sense to make
the conditional depend on CONFIG_I2C_BOARDINFO rather that CONFIG_I2C.

> > >  extern int
> > >  i2c_register_board_info(int busnum, struct i2c_board_info const *info, unsigned n);
> > > -
> > > +#else
> > > +static inline int
> > > +i2c_register_board_info(int busnum, struct i2c_board_info const *info, unsigned n)
> > > +{
> > > +	return 0;
> > > +}
> > > +#endif
> > >  
> > >  /*
> > >   * The following structs are for those who like to implement new bus drivers:
> > 
> > Other than that I am fine with this patch. Want it in 2.6.25 or 2.6.26
> > is enough?
> 
> I'd go for 2.6.25, since it's a build fix.

OK, will do.

-- 
Jean Delvare

_______________________________________________
i2c mailing list
i2c-GZX6beZjE8VD60Wz+7aTrA@public.gmane.org
http://lists.lm-sensors.org/mailman/listinfo/i2c