On Dec 26, 2007 4:43 AM, Bryan Wu <[EMAIL PROTECTED]> wrote:
> On Wed, 2007-12-26 at 04:32 -0500, Mike Frysinger wrote:
> > On Dec 26, 2007 2:10 AM,  <[EMAIL PROTECTED]> wrote:
> > >  Revision 4065 Author cooloney Date 2007-12-26 01:10:10 -0600 (Wed, 26 Dec
> > > 2007)
> > > Log Message TASK [#3478] enable generic GPIO based I2C driver in 
> > > EZKIT-BF561
> > >
> > > Diffstat Kconfig | 25 +++++++++++++++++++++++++
> > >  mach-bf561/boards/ezkit.c | 24 ++++++++++++++++++++++++
> > >  2 files changed, 49 insertions(+)
> > >
> > >
> > > Modified Paths
> > >
> > > trunk/arch/blackfin/Kconfig
> > > trunk/arch/blackfin/mach-bf561/boards/ezkit.c
> >
> > why not just provide an example in the BF561-EZKIT board and be done
> > with it ?  no need for the Kconfig entry.
>
> That is a trade-off. I know kernel world dislike these Kconfig things,
> but customers like it. They don't wanna to edit the source code.
>
> Currently, this change is required by a customer from Yi.
> Of course, we can remove these Kconfig in the future.

they're going to need to edit the code for their platform regardless.
looking at it in a larger context, Kconfig is confusing if you need
more than one instance of software I2C.  the driver itself has no such
limitation, but by putting the bindings into the Kconfig rather than
the board resources file, it looks like you can only have 1 bus.  it's
also confusing because there's a disconnect between the .config and
the actual settings.  the .config is respected only by the board
resources file, not the driver itself.  so it currently only works on
the BF561-EZKIT even though the option is there for all Blackfins.
customers can waste time wondering why their own custom board isnt
respecting the settings they've selected in the Kconfig.  by having it
all in the board resources, there is no confusion and it operates
exactly the same like any other platform driver.
-mike
_______________________________________________
Linux-kernel-commits mailing list
[email protected]
http://blackfin.uclinux.org/mailman/listinfo/linux-kernel-commits

Reply via email to